Tau2 bench changes

When `sudo hermes gateway install --system --run-as-user <user>` generates
the systemd unit, get_hermes_home() resolves to /root/.hermes because
Path.home() returns root's home under sudo. The unit correctly sets
HOME= and User= via _system_service_identity(), but HERMES_HOME was
computed independently and pointed to root's config directory.

Add _hermes_home_for_target_user() which remaps the current HERMES_HOME
to the equivalent path under the target user's home. This handles:
- Default ~/.hermes → target user's ~/.hermes
- Profiles (e.g. ~/.hermes/profiles/coder) → preserves relative structure
- Custom paths (e.g. /opt/hermes) → kept as-is

Supersedes #3861 which only handled the default case and left profiles
broken (also flagged by Copilot review).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

.cursorrules

@@ -1,23 +0,0 @@
-Hermes-Agent is an agent harness for LLMs.
-
-When building, the tool functionality is in the tools/ directory, where each specific tool (or in some cases, tools that are built for the same execution category or api) are placed in a script each their own.
-
-Each tool is then consolidated in the model_tools.py file in the repo root.
-
-There is also a way to consolidate sets of tools in toolsets.py for the agent to use.
-
-The primary agent runner code is in run_agent, but other runners could be developed using the tools and framework.
-
-Always ensure consistency between tools, the model_tools.py and toolsets.py when changing any of them, otherwise they could become desynced in a way that is detrimental to functionality.
-
-The expected pathway for using API keys is to setup and place them in a .env file in the repo root.
-
-Test scripts will be placed in tests/
-
-The run_agent loop is setup to:
-- Process the enabled toolsets to provide to the model,
-- Pipe in a prompt or problem from the input to the agent,
-- Loop the LLM each time it calls a tool, until the model decides no more tools are needed and provides a natural language response,
-- Return that response.
-
-There are additional caveats for logging, where we restructure the "tools" as a system prompt for storage later into a format that can be used and handled properly later.

.dockerignore

@@ -0,0 +1,15 @@
+# Git
+.git
+.gitignore
+.gitmodules
+
+# Dependencies
+node_modules
+
+# CI/CD
+.github
+
+# Environment files
+.env
+
+*.md

.env.example

@@ -1,49 +1,351 @@
 # Hermes Agent Environment Configuration
 # Copy this file to .env and fill in your API keys
-# Get API keys from the URLs listed below
 
 # =============================================================================
-# REQUIRED API KEYS
+# LLM PROVIDER (OpenRouter)
+# =============================================================================
+# OpenRouter provides access to many models through one API
+# All LLM calls go through OpenRouter - no direct provider keys needed
+# Get your key at: https://openrouter.ai/keys
+# OPENROUTER_API_KEY=
+
+# Default model is configured in ~/.hermes/config.yaml (model.default).
+# Use 'hermes model' or 'hermes setup' to change it.
+# LLM_MODEL is no longer read from .env — this line is kept for reference only.
+# LLM_MODEL=anthropic/claude-opus-4.6
+
+# =============================================================================
+# LLM PROVIDER (z.ai / GLM)
+# =============================================================================
+# z.ai provides access to ZhipuAI GLM models (GLM-4-Plus, etc.)
+# Get your key at: https://z.ai or https://open.bigmodel.cn
+# GLM_API_KEY=
+# GLM_BASE_URL=https://api.z.ai/api/paas/v4  # Override default base URL
+
+# =============================================================================
+# LLM PROVIDER (Kimi / Moonshot)
+# =============================================================================
+# Kimi Code provides access to Moonshot AI coding models (kimi-k2.5, etc.)
+# Get your key at: https://platform.kimi.ai (Kimi Code console)
+# Keys prefixed sk-kimi- use the Kimi Code API (api.kimi.com) by default.
+# Legacy keys from platform.moonshot.ai need KIMI_BASE_URL override below.
+# KIMI_API_KEY=
+# KIMI_BASE_URL=https://api.kimi.com/coding/v1  # Default for sk-kimi- keys
+# KIMI_BASE_URL=https://api.moonshot.ai/v1      # For legacy Moonshot keys
+# KIMI_BASE_URL=https://api.moonshot.cn/v1       # For Moonshot China keys
+
+# =============================================================================
+# LLM PROVIDER (MiniMax)
+# =============================================================================
+# MiniMax provides access to MiniMax models (global endpoint)
+# Get your key at: https://www.minimax.io
+# MINIMAX_API_KEY=
+# MINIMAX_BASE_URL=https://api.minimax.io/v1  # Override default base URL
+
+# MiniMax China endpoint (for users in mainland China)
+# MINIMAX_CN_API_KEY=
+# MINIMAX_CN_BASE_URL=https://api.minimaxi.com/v1  # Override default base URL
+
+# =============================================================================
+# LLM PROVIDER (OpenCode Zen)
+# =============================================================================
+# OpenCode Zen provides curated, tested models (GPT, Claude, Gemini, MiniMax, GLM, Kimi)
+# Pay-as-you-go pricing. Get your key at: https://opencode.ai/auth
+# OPENCODE_ZEN_API_KEY=
+# OPENCODE_ZEN_BASE_URL=https://opencode.ai/zen/v1  # Override default base URL
+
+# =============================================================================
+# LLM PROVIDER (OpenCode Go)
+# =============================================================================
+# OpenCode Go provides access to open models (GLM-5, Kimi K2.5, MiniMax M2.5)
+# $10/month subscription. Get your key at: https://opencode.ai/auth
+# OPENCODE_GO_API_KEY=
+
+# =============================================================================
+# LLM PROVIDER (Hugging Face Inference Providers)
+# =============================================================================
+# Hugging Face routes to 20+ open models via unified OpenAI-compatible endpoint.
+# Free tier included ($0.10/month), no markup on provider rates.
+# Get your token at: https://huggingface.co/settings/tokens
+# Required permission: "Make calls to Inference Providers"
+# HF_TOKEN=
+# OPENCODE_GO_BASE_URL=https://opencode.ai/zen/go/v1  # Override default base URL
+
+# =============================================================================
+# TOOL API KEYS
 # =============================================================================
 
-# Anthropic API Key - Main agent model
-# Get at: https://console.anthropic.com/
-ANTHROPIC_API_KEY=
+# Exa API Key - AI-native web search and contents
+# Get at: https://exa.ai
+# EXA_API_KEY=
+
+# Parallel API Key - AI-native web search and extract
+# Get at: https://parallel.ai
+# PARALLEL_API_KEY=
 
 # Firecrawl API Key - Web search, extract, and crawl
 # Get at: https://firecrawl.dev/
-FIRECRAWL_API_KEY=
+# FIRECRAWL_API_KEY=
 
-# Nous Research API Key - Vision analysis and multi-model reasoning
-# Get at: https://inference-api.nousresearch.com/
-NOUS_API_KEY=
-
-# Morph API Key - Terminal/command execution tools
-# Get at: https://morph.so/
-MORPH_API_KEY=
 
 # FAL.ai API Key - Image generation
 # Get at: https://fal.ai/
-FAL_KEY=
+# FAL_KEY=
+
+# Honcho - Cross-session AI-native user modeling (optional)
+# Builds a persistent understanding of the user across sessions and tools.
+# Get at: https://app.honcho.dev
+# Also requires ~/.honcho/config.json with enabled=true (see README).
+# HONCHO_API_KEY=
 
 # =============================================================================
-# OPTIONAL API KEYS
+# TERMINAL TOOL CONFIGURATION
 # =============================================================================
+# Backend type: "local", "singularity", "docker", "modal", or "ssh"
+# Terminal backend is configured in ~/.hermes/config.yaml (terminal.backend).
+# Use 'hermes setup' or 'hermes config set terminal.backend docker' to change.
+# Supported: local, docker, singularity, modal, ssh
+#
+# Only override here if you need to force a backend without touching config.yaml:
+# TERMINAL_ENV=local
 
-# OpenAI API Key - Optional, for enhanced Hecate features
-# Get at: https://platform.openai.com/
-OPENAI_API_KEY=
+# Container images (for singularity/docker/modal backends)
+# TERMINAL_DOCKER_IMAGE=nikolaik/python-nodejs:python3.11-nodejs20
+# TERMINAL_SINGULARITY_IMAGE=docker://nikolaik/python-nodejs:python3.11-nodejs20
+TERMINAL_MODAL_IMAGE=nikolaik/python-nodejs:python3.11-nodejs20
+
+
+# Working directory for terminal commands
+# For local backend: "." means current directory (resolved automatically)
+# For remote backends (ssh/docker/modal/singularity): use an absolute path
+#   INSIDE the target environment, or leave unset for the backend's default
+#   (/root for modal, / for docker, ~ for ssh). Do NOT use a host-local path.
+# Usually managed by config.yaml (terminal.cwd) — uncomment to override
+# TERMINAL_CWD=.
+
+# Default command timeout in seconds
+TERMINAL_TIMEOUT=60
+
+# Cleanup inactive environments after this many seconds
+TERMINAL_LIFETIME_SECONDS=300
 
 # =============================================================================
-# OPTIONAL CONFIGURATION
+# SSH REMOTE EXECUTION (for TERMINAL_ENV=ssh)
 # =============================================================================
+# Run terminal commands on a remote server via SSH.
+# Agent code stays on your machine, commands execute remotely.
+#
+# SECURITY BENEFITS:
+# - Agent cannot read your .env file (API keys protected)
+# - Agent cannot modify its own code
+# - Remote server acts as isolated sandbox
+# - Can safely configure passwordless sudo on remote
+#
+# TERMINAL_SSH_HOST=192.168.1.100
+# TERMINAL_SSH_USER=agent
+# TERMINAL_SSH_PORT=22
+# TERMINAL_SSH_KEY=~/.ssh/id_rsa
 
-# Terminal Tool Settings
-HECATE_VM_LIFETIME_SECONDS=300
-HECATE_DEFAULT_SNAPSHOT_ID=snapshot_p5294qxt
+# =============================================================================
+# SUDO SUPPORT (works with ALL terminal backends)
+# =============================================================================
+# If set, enables sudo commands by piping password via `sudo -S`.
+# Works with: local, docker, singularity, modal, and ssh backends.
+# 
+# SECURITY WARNING: Password stored in plaintext. Only use on trusted machines.
+# 
+# ALTERNATIVES:
+# - For SSH backend: Configure passwordless sudo on the remote server
+# - For containers: Run as root inside the container (no sudo needed)
+# - For local: Configure /etc/sudoers for specific commands
+# - For CLI: Leave unset - you'll be prompted interactively with 45s timeout
+#
+# SUDO_PASSWORD=your_password_here
 
-# Debug Logging (set to "true" to enable, logs saved to ./logs/)
+# =============================================================================
+# MODAL CLOUD BACKEND (Optional - for TERMINAL_ENV=modal)
+# =============================================================================
+# Modal uses CLI authentication, not environment variables.
+# Run: pip install modal && modal setup
+# This will authenticate via browser and store credentials locally.
+# No API key needed in .env - Modal handles auth automatically.
+
+# =============================================================================
+# BROWSER TOOL CONFIGURATION (agent-browser + Browserbase)
+# =============================================================================
+# Browser automation requires Browserbase cloud service for remote browser execution.
+# This allows the agent to navigate websites, fill forms, and extract information.
+#
+# STEALTH MODES:
+# - Basic Stealth: ALWAYS active (random fingerprints, auto CAPTCHA solving)
+# - Advanced Stealth: Requires BROWSERBASE_ADVANCED_STEALTH=true (Scale Plan only)
+
+# Browserbase API Key - Cloud browser execution
+# Get at: https://browserbase.com/
+# BROWSERBASE_API_KEY=
+
+# Browserbase Project ID - From your Browserbase dashboard
+# BROWSERBASE_PROJECT_ID=
+
+# Enable residential proxies for better CAPTCHA solving (default: true)
+# Routes traffic through residential IPs, significantly improves success rate
+BROWSERBASE_PROXIES=true
+
+# Enable advanced stealth mode (default: false, requires Scale Plan)
+# Uses custom Chromium build to avoid bot detection altogether
+BROWSERBASE_ADVANCED_STEALTH=false
+
+# Browser session timeout in seconds (default: 300)
+# Sessions are cleaned up after this duration of inactivity
+BROWSER_SESSION_TIMEOUT=300
+
+# Browser inactivity timeout - auto-cleanup inactive sessions (default: 120 = 2 min)
+# Browser sessions are automatically closed after this period of no activity
+BROWSER_INACTIVITY_TIMEOUT=120
+
+# =============================================================================
+# SESSION LOGGING
+# =============================================================================
+# Session trajectories are automatically saved to logs/ directory
+# Format: logs/session_YYYYMMDD_HHMMSS_UUID.json
+# Contains full conversation history in trajectory format for debugging/replay
+
+# =============================================================================
+# VOICE TRANSCRIPTION & OPENAI TTS
+# =============================================================================
+# Required for voice message transcription (Whisper) and OpenAI TTS voices.
+# Uses OpenAI's API directly (not via OpenRouter).
+# Named VOICE_TOOLS_OPENAI_KEY to avoid interference with OpenRouter.
+# Get at: https://platform.openai.com/api-keys
+# VOICE_TOOLS_OPENAI_KEY=
+
+# =============================================================================
+# SLACK INTEGRATION
+# =============================================================================
+# Slack Bot Token - From Slack App settings (OAuth & Permissions)
+# Get at: https://api.slack.com/apps
+# SLACK_BOT_TOKEN=xoxb-...
+
+# Slack App Token - For Socket Mode (App-Level Tokens in Slack App settings)
+# SLACK_APP_TOKEN=xapp-...
+
+# Slack allowed users (comma-separated Slack user IDs)
+# SLACK_ALLOWED_USERS=
+
+# =============================================================================
+# TELEGRAM INTEGRATION
+# =============================================================================
+# Telegram Bot Token - From @BotFather (https://t.me/BotFather)
+# TELEGRAM_BOT_TOKEN=
+# TELEGRAM_ALLOWED_USERS=                  # Comma-separated user IDs
+# TELEGRAM_HOME_CHANNEL=                   # Default chat for cron delivery
+# TELEGRAM_HOME_CHANNEL_NAME=              # Display name for home channel
+
+# Webhook mode (optional — for cloud deployments like Fly.io/Railway)
+# Default is long polling. Setting TELEGRAM_WEBHOOK_URL switches to webhook mode.
+# TELEGRAM_WEBHOOK_URL=https://my-app.fly.dev/telegram
+# TELEGRAM_WEBHOOK_PORT=8443
+# TELEGRAM_WEBHOOK_SECRET=                 # Recommended for production
+
+# WhatsApp (built-in Baileys bridge — run `hermes whatsapp` to pair)
+# WHATSAPP_ENABLED=false
+# WHATSAPP_ALLOWED_USERS=15551234567
+
+# Email (IMAP/SMTP — send and receive emails as Hermes)
+# For Gmail: enable 2FA → create App Password at https://myaccount.google.com/apppasswords
+# EMAIL_ADDRESS=hermes@gmail.com
+# EMAIL_PASSWORD=xxxx xxxx xxxx xxxx
+# EMAIL_IMAP_HOST=imap.gmail.com
+# EMAIL_IMAP_PORT=993
+# EMAIL_SMTP_HOST=smtp.gmail.com
+# EMAIL_SMTP_PORT=587
+# EMAIL_POLL_INTERVAL=15
+# EMAIL_ALLOWED_USERS=your@email.com
+# EMAIL_HOME_ADDRESS=your@email.com
+
+# Gateway-wide: allow ALL users without an allowlist (default: false = deny)
+# Only set to true if you intentionally want open access.
+# GATEWAY_ALLOW_ALL_USERS=false
+
+# =============================================================================
+# RESPONSE PACING
+# =============================================================================
+# Human-like delays between message chunks on messaging platforms.
+# Makes the bot feel less robotic.
+# HERMES_HUMAN_DELAY_MODE=off     # off | natural | custom
+# HERMES_HUMAN_DELAY_MIN_MS=800   # Min delay in ms (custom mode)
+# HERMES_HUMAN_DELAY_MAX_MS=2500  # Max delay in ms (custom mode)
+
+# =============================================================================
+# DEBUG OPTIONS
+# =============================================================================
 WEB_TOOLS_DEBUG=false
 VISION_TOOLS_DEBUG=false
 MOA_TOOLS_DEBUG=false
 IMAGE_TOOLS_DEBUG=false
+
+# =============================================================================
+# CONTEXT COMPRESSION (Auto-shrinks long conversations)
+# =============================================================================
+# When conversation approaches model's context limit, middle turns are
+# automatically summarized to free up space.
+#
+# Context compression is configured in ~/.hermes/config.yaml under compression:
+# CONTEXT_COMPRESSION_ENABLED=true        # Enable auto-compression (default: true)
+# CONTEXT_COMPRESSION_THRESHOLD=0.85      # Compress at 85% of context limit
+# Model is set via compression.summary_model in config.yaml (default: google/gemini-3-flash-preview)
+
+# =============================================================================
+# RL TRAINING (Tinker + Atropos)
+# =============================================================================
+# Run reinforcement learning training on language models using the Tinker API.
+# Requires the rl-server to be running (from tinker-atropos package).
+
+# Tinker API Key - RL training service
+# Get at: https://tinker-console.thinkingmachines.ai/keys
+# TINKER_API_KEY=
+
+# Weights & Biases API Key - Experiment tracking and metrics
+# Get at: https://wandb.ai/authorize
+# WANDB_API_KEY=
+
+# RL API Server URL (default: http://localhost:8080)
+# Change if running the rl-server on a different host/port
+# RL_API_URL=http://localhost:8080
+
+# =============================================================================
+# SKILLS HUB (GitHub integration for skill search/install/publish)
+# =============================================================================
+
+# GitHub Personal Access Token — for higher API rate limits on skill search/install
+# Get at: https://github.com/settings/tokens (Fine-grained recommended)
+# GITHUB_TOKEN=ghp_xxxxxxxxxxxxxxxxxxxx
+
+# GitHub App credentials (optional — for bot identity on PRs)
+# GITHUB_APP_ID=
+# GITHUB_APP_PRIVATE_KEY_PATH=
+# GITHUB_APP_INSTALLATION_ID=
+
+# Groq API key (free tier — used for Whisper STT in voice mode)
+# GROQ_API_KEY=
+
+# =============================================================================
+# STT PROVIDER SELECTION
+# =============================================================================
+# Default STT provider is "local" (faster-whisper) — runs on your machine, no API key needed.
+# Install with: pip install faster-whisper
+# Model downloads automatically on first use (~150 MB for "base").
+# To use cloud providers instead, set GROQ_API_KEY or VOICE_TOOLS_OPENAI_KEY above.
+# Provider priority: local > groq > openai
+# Configure in config.yaml: stt.provider: local | groq | openai
+
+# =============================================================================
+# STT ADVANCED OVERRIDES (optional)
+# =============================================================================
+# Override default STT models per provider (normally set via stt.model in config.yaml)
+# STT_GROQ_MODEL=whisper-large-v3-turbo
+# STT_OPENAI_MODEL=whisper-1
+
+# Override STT provider endpoints (for proxies or self-hosted instances)
+# GROQ_BASE_URL=https://api.groq.com/openai/v1
+# STT_OPENAI_BASE_URL=https://api.openai.com/v1

.envrc

@@ -0,0 +1 @@
+use flake

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -0,0 +1,144 @@
+name: "🐛 Bug Report"
+description: Report a bug — something that's broken, crashes, or behaves incorrectly.
+title: "[Bug]: "
+labels: ["bug"]
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Thanks for reporting a bug! Please fill out the sections below so we can reproduce and fix it quickly.
+
+        **Before submitting**, please:
+        - [ ] Search [existing issues](https://github.com/NousResearch/hermes-agent/issues) to avoid duplicates
+        - [ ] Update to the latest version (`hermes update`) and confirm the bug still exists
+
+  - type: textarea
+    id: description
+    attributes:
+      label: Bug Description
+      description: A clear description of what's broken. Include error messages, tracebacks, or screenshots if relevant.
+      placeholder: |
+        What happened? What did you expect to happen instead?
+    validations:
+      required: true
+
+  - type: textarea
+    id: reproduction
+    attributes:
+      label: Steps to Reproduce
+      description: Minimal steps to trigger the bug. The more specific, the faster we can fix it.
+      placeholder: |
+        1. Run `hermes chat`
+        2. Send the message "..."
+        3. Agent calls tool X
+        4. Error appears: ...
+    validations:
+      required: true
+
+  - type: textarea
+    id: expected
+    attributes:
+      label: Expected Behavior
+      description: What should have happened instead?
+    validations:
+      required: true
+
+  - type: textarea
+    id: actual
+    attributes:
+      label: Actual Behavior
+      description: What actually happened? Include full error output if available.
+    validations:
+      required: true
+
+  - type: dropdown
+    id: component
+    attributes:
+      label: Affected Component
+      description: Which part of Hermes is affected?
+      multiple: true
+      options:
+        - CLI (interactive chat)
+        - Gateway (Telegram/Discord/Slack/WhatsApp)
+        - Setup / Installation
+        - Tools (terminal, file ops, web, code execution, etc.)
+        - Skills (skill loading, skill hub, skill guard)
+        - Agent Core (conversation loop, context compression, memory)
+        - Configuration (config.yaml, .env, hermes setup)
+        - Other
+    validations:
+      required: true
+
+  - type: dropdown
+    id: platform
+    attributes:
+      label: Messaging Platform (if gateway-related)
+      description: Which platform adapter is affected?
+      multiple: true
+      options:
+        - N/A (CLI only)
+        - Telegram
+        - Discord
+        - Slack
+        - WhatsApp
+
+  - type: input
+    id: os
+    attributes:
+      label: Operating System
+      description: e.g. Ubuntu 24.04, macOS 15.2, Windows 11
+      placeholder: Ubuntu 24.04
+    validations:
+      required: true
+
+  - type: input
+    id: python-version
+    attributes:
+      label: Python Version
+      description: Output of `python --version`
+      placeholder: "3.11.9"
+    validations:
+      required: true
+
+  - type: input
+    id: hermes-version
+    attributes:
+      label: Hermes Version
+      description: Output of `hermes version`
+      placeholder: "2.1.0"
+    validations:
+      required: true
+
+  - type: textarea
+    id: logs
+    attributes:
+      label: Relevant Logs / Traceback
+      description: Paste any error output, traceback, or log messages. This will be auto-formatted as code.
+      render: shell
+
+  - type: textarea
+    id: root-cause
+    attributes:
+      label: Root Cause Analysis (optional)
+      description: |
+        If you've dug into the code and identified the root cause, share it here.
+        Include file paths, line numbers, and code snippets if possible. This massively speeds up fixes.
+      placeholder: |
+        The bug is in `gateway/run.py` line 949. `len(history)` counts session_meta entries
+        but `agent_messages` was built from filtered history...
+
+  - type: textarea
+    id: proposed-fix
+    attributes:
+      label: Proposed Fix (optional)
+      description: If you have a fix in mind (or a PR ready), describe it here.
+      placeholder: |
+        Replace `.get()` with `.pop()` on line 289 of `gateway/platforms/base.py`
+        to actually clear the pending message after retrieval.
+
+  - type: checkboxes
+    id: pr-ready
+    attributes:
+      label: Are you willing to submit a PR for this?
+      options:
+        - label: I'd like to fix this myself and submit a PR

.github/ISSUE_TEMPLATE/config.yml

@@ -0,0 +1,11 @@
+blank_issues_enabled: true
+contact_links:
+  - name: 💬 Nous Research Discord
+    url: https://discord.gg/NousResearch
+    about: For quick questions, showcasing projects, sharing skills, and community chat.
+  - name: 📖 Documentation
+    url: https://github.com/NousResearch/hermes-agent/blob/main/README.md
+    about: Check the README and docs before opening an issue.
+  - name: 🤝 Contributing Guide
+    url: https://github.com/NousResearch/hermes-agent/blob/main/CONTRIBUTING.md
+    about: Read this before submitting a PR.

.github/ISSUE_TEMPLATE/feature_request.yml

@@ -0,0 +1,73 @@
+name: "✨ Feature Request"
+description: Suggest a new feature or improvement.
+title: "[Feature]: "
+labels: ["enhancement"]
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Thanks for the suggestion! Before submitting, please consider:
+
+        - **Is this a new skill?** Most capabilities should be [skills, not tools](https://github.com/NousResearch/hermes-agent/blob/main/CONTRIBUTING.md#should-it-be-a-skill-or-a-tool). If it's a specialized integration (crypto, NFT, niche SaaS), it belongs on the Skills Hub, not bundled.
+        - **Search [existing issues](https://github.com/NousResearch/hermes-agent/issues)** — someone may have already proposed this.
+
+  - type: textarea
+    id: problem
+    attributes:
+      label: Problem or Use Case
+      description: What problem does this solve? What are you trying to do that you can't today?
+      placeholder: |
+        I'm trying to use Hermes with [provider/platform/workflow] but currently
+        there's no way to...
+    validations:
+      required: true
+
+  - type: textarea
+    id: solution
+    attributes:
+      label: Proposed Solution
+      description: How do you think this should work? Be as specific as you can — CLI flags, config options, UI behavior.
+      placeholder: |
+        Add a `--foo` flag to `hermes chat` that enables...
+        Or: Add a config key `bar.baz` that controls...
+    validations:
+      required: true
+
+  - type: textarea
+    id: alternatives
+    attributes:
+      label: Alternatives Considered
+      description: What other approaches did you consider? Why is the proposed solution better?
+
+  - type: dropdown
+    id: type
+    attributes:
+      label: Feature Type
+      options:
+        - New tool
+        - New bundled skill
+        - CLI improvement
+        - Gateway / messaging improvement
+        - Configuration option
+        - Performance / reliability
+        - Developer experience (tests, docs, CI)
+        - Other
+    validations:
+      required: true
+
+  - type: dropdown
+    id: scope
+    attributes:
+      label: Scope
+      description: How big is this change?
+      options:
+        - Small (single file, < 50 lines)
+        - Medium (few files, < 300 lines)
+        - Large (new module or significant refactor)
+
+  - type: checkboxes
+    id: pr-ready
+    attributes:
+      label: Contribution
+      options:
+        - label: I'd like to implement this myself and submit a PR

.github/ISSUE_TEMPLATE/setup_help.yml

@@ -0,0 +1,100 @@
+name: "🔧 Setup / Installation Help"
+description: Having trouble installing or configuring Hermes? Ask here.
+title: "[Setup]: "
+labels: ["setup"]
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Sorry you're having trouble! Please fill out the details below so we can help.
+
+        **Quick checks first:**
+        - Run `hermes doctor` and include the output below
+        - Try `hermes update` to get the latest version
+        - Check the [README troubleshooting section](https://github.com/NousResearch/hermes-agent#troubleshooting)
+        - For general questions, consider the [Nous Research Discord](https://discord.gg/NousResearch) for faster help
+
+  - type: textarea
+    id: description
+    attributes:
+      label: What's Going Wrong?
+      description: Describe what you're trying to do and where it fails.
+      placeholder: |
+        I ran `hermes setup` and selected Nous Portal, but when I try to
+        start the gateway I get...
+    validations:
+      required: true
+
+  - type: textarea
+    id: steps
+    attributes:
+      label: Steps Taken
+      description: What did you do? Include the exact commands you ran.
+      placeholder: |
+        1. Ran the install script: `curl -fsSL ... | bash`
+        2. Ran `hermes setup` and chose "Quick setup"
+        3. Selected OpenRouter, entered API key
+        4. Ran `hermes chat` and got error...
+    validations:
+      required: true
+
+  - type: dropdown
+    id: install-method
+    attributes:
+      label: Installation Method
+      options:
+        - Install script (curl | bash)
+        - Manual clone + pip/uv install
+        - PowerShell installer (Windows)
+        - Docker
+        - Other
+    validations:
+      required: true
+
+  - type: input
+    id: os
+    attributes:
+      label: Operating System
+      placeholder: Ubuntu 24.04 / macOS 15.2 / Windows 11
+    validations:
+      required: true
+
+  - type: input
+    id: python-version
+    attributes:
+      label: Python Version
+      description: Output of `python --version` (or `python3 --version`)
+      placeholder: "3.11.9"
+
+  - type: input
+    id: hermes-version
+    attributes:
+      label: Hermes Version
+      description: Output of `hermes version` (if install got that far)
+      placeholder: "2.1.0"
+
+  - type: textarea
+    id: doctor-output
+    attributes:
+      label: Output of `hermes doctor`
+      description: Run `hermes doctor` and paste the full output. This will be auto-formatted.
+      render: shell
+
+  - type: textarea
+    id: error-output
+    attributes:
+      label: Full Error Output
+      description: Paste the complete error message or traceback. This will be auto-formatted.
+      render: shell
+    validations:
+      required: true
+
+  - type: textarea
+    id: tried
+    attributes:
+      label: What I've Already Tried
+      description: List any fixes or workarounds you've already attempted.
+      placeholder: |
+        - Ran `hermes update`
+        - Tried reinstalling with `pip install -e ".[all]"`
+        - Checked that OPENROUTER_API_KEY is set in ~/.hermes/.env

.github/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,75 @@
+## What does this PR do?
+
+<!-- Describe the change clearly. What problem does it solve? Why is this approach the right one? -->
+
+
+
+## Related Issue
+
+<!-- Link the issue this PR addresses. If no issue exists, consider creating one first. -->
+
+Fixes #
+
+## Type of Change
+
+<!-- Check the one that applies. -->
+
+- [ ] 🐛 Bug fix (non-breaking change that fixes an issue)
+- [ ] ✨ New feature (non-breaking change that adds functionality)
+- [ ] 🔒 Security fix
+- [ ] 📝 Documentation update
+- [ ] ✅ Tests (adding or improving test coverage)
+- [ ] ♻️ Refactor (no behavior change)
+- [ ] 🎯 New skill (bundled or hub)
+
+## Changes Made
+
+<!-- List the specific changes. Include file paths for code changes. -->
+
+- 
+
+## How to Test
+
+<!-- Steps to verify this change works. For bugs: reproduction steps + proof that the fix works. -->
+
+1. 
+2. 
+3. 
+
+## Checklist
+
+<!-- Complete these before requesting review. -->
+
+### Code
+
+- [ ] I've read the [Contributing Guide](https://github.com/NousResearch/hermes-agent/blob/main/CONTRIBUTING.md)
+- [ ] My commit messages follow [Conventional Commits](https://www.conventionalcommits.org/) (`fix(scope):`, `feat(scope):`, etc.)
+- [ ] I searched for [existing PRs](https://github.com/NousResearch/hermes-agent/pulls) to make sure this isn't a duplicate
+- [ ] My PR contains **only** changes related to this fix/feature (no unrelated commits)
+- [ ] I've run `pytest tests/ -q` and all tests pass
+- [ ] I've added tests for my changes (required for bug fixes, strongly encouraged for features)
+- [ ] I've tested on my platform: <!-- e.g. Ubuntu 24.04, macOS 15.2, Windows 11 -->
+
+### Documentation & Housekeeping
+
+<!-- Check all that apply. It's OK to check "N/A" if a category doesn't apply to your change. -->
+
+- [ ] I've updated relevant documentation (README, `docs/`, docstrings) — or N/A
+- [ ] I've updated `cli-config.yaml.example` if I added/changed config keys — or N/A
+- [ ] I've updated `CONTRIBUTING.md` or `AGENTS.md` if I changed architecture or workflows — or N/A
+- [ ] I've considered cross-platform impact (Windows, macOS) per the [compatibility guide](https://github.com/NousResearch/hermes-agent/blob/main/CONTRIBUTING.md#cross-platform-compatibility) — or N/A
+- [ ] I've updated tool descriptions/schemas if I changed tool behavior — or N/A
+
+## For New Skills
+
+<!-- Only fill this out if you're adding a skill. Delete this section otherwise. -->
+
+- [ ] This skill is **broadly useful** to most users (if bundled) — see [Contributing Guide](https://github.com/NousResearch/hermes-agent/blob/main/CONTRIBUTING.md#should-the-skill-be-bundled)
+- [ ] SKILL.md follows the [standard format](https://github.com/NousResearch/hermes-agent/blob/main/CONTRIBUTING.md#skillmd-format) (frontmatter, trigger conditions, steps, pitfalls)
+- [ ] No external dependencies that aren't already available (prefer stdlib, curl, existing Hermes tools)
+- [ ] I've tested the skill end-to-end: `hermes --toolsets skills -q "Use the X skill to do Y"`
+
+## Screenshots / Logs
+
+<!-- If applicable, add screenshots or log output showing the fix/feature in action. -->
+

.github/workflows/deploy-site.yml

@@ -0,0 +1,62 @@
+name: Deploy Site
+
+on:
+  push:
+    branches: [main]
+    paths:
+      - 'website/**'
+      - 'landingpage/**'
+      - '.github/workflows/deploy-site.yml'
+  workflow_dispatch:
+
+permissions:
+  pages: write
+  id-token: write
+
+concurrency:
+  group: pages
+  cancel-in-progress: false
+
+jobs:
+  build-and-deploy:
+    # Only run on the upstream repository, not on forks
+    if: github.repository == 'NousResearch/hermes-agent'
+    runs-on: ubuntu-latest
+    environment:
+      name: github-pages
+      url: ${{ steps.deploy.outputs.page_url }}
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: actions/setup-node@v4
+        with:
+          node-version: 20
+          cache: npm
+          cache-dependency-path: website/package-lock.json
+
+      - name: Install dependencies
+        run: npm ci
+        working-directory: website
+
+      - name: Build Docusaurus
+        run: npm run build
+        working-directory: website
+
+      - name: Stage deployment
+        run: |
+          mkdir -p _site/docs
+          # Landing page at root
+          cp -r landingpage/* _site/
+          # Docusaurus at /docs/
+          cp -r website/build/* _site/docs/
+          # CNAME so GitHub Pages keeps the custom domain between deploys
+          echo "hermes-agent.nousresearch.com" > _site/CNAME
+
+      - name: Upload artifact
+        uses: actions/upload-pages-artifact@v3
+        with:
+          path: _site
+
+      - name: Deploy to GitHub Pages
+        id: deploy
+        uses: actions/deploy-pages@v4

.github/workflows/docker-publish.yml

@@ -0,0 +1,79 @@
+name: Docker Build and Publish
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+  release:
+    types: [published]
+
+concurrency:
+  group: docker-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  build-and-push:
+    # Only run on the upstream repository, not on forks
+    if: github.repository == 'NousResearch/hermes-agent'
+    runs-on: ubuntu-latest
+    timeout-minutes: 30
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+        with:
+          submodules: recursive
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Build image
+        uses: docker/build-push-action@v6
+        with:
+          context: .
+          file: Dockerfile
+          load: true
+          tags: nousresearch/hermes-agent:test
+          cache-from: type=gha
+          cache-to: type=gha,mode=max
+
+      - name: Test image starts
+        run: |
+          docker run --rm \
+            -v /tmp/hermes-test:/opt/data \
+            --entrypoint /opt/hermes/docker/entrypoint.sh \
+            nousresearch/hermes-agent:test --help
+
+      - name: Log in to Docker Hub
+        if: github.event_name == 'push' && github.ref == 'refs/heads/main' || github.event_name == 'release'
+        uses: docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKERHUB_TOKEN }}
+
+      - name: Push image (main branch)
+        if: github.event_name == 'push' && github.ref == 'refs/heads/main'
+        uses: docker/build-push-action@v6
+        with:
+          context: .
+          file: Dockerfile
+          push: true
+          tags: |
+            nousresearch/hermes-agent:latest
+            nousresearch/hermes-agent:${{ github.sha }}
+          cache-from: type=gha
+          cache-to: type=gha,mode=max
+
+      - name: Push image (release)
+        if: github.event_name == 'release'
+        uses: docker/build-push-action@v6
+        with:
+          context: .
+          file: Dockerfile
+          push: true
+          tags: |
+            nousresearch/hermes-agent:latest
+            nousresearch/hermes-agent:${{ github.event.release.tag_name }}
+            nousresearch/hermes-agent:${{ github.sha }}
+          cache-from: type=gha
+          cache-to: type=gha,mode=max

.github/workflows/docs-site-checks.yml

@@ -0,0 +1,39 @@
+name: Docs Site Checks
+
+on:
+  pull_request:
+    paths:
+      - 'website/**'
+      - '.github/workflows/docs-site-checks.yml'
+  workflow_dispatch:
+
+jobs:
+  docs-site-checks:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: actions/setup-node@v4
+        with:
+          node-version: 20
+          cache: npm
+          cache-dependency-path: website/package-lock.json
+
+      - name: Install website dependencies
+        run: npm ci
+        working-directory: website
+
+      - uses: actions/setup-python@v5
+        with:
+          python-version: '3.11'
+
+      - name: Install ascii-guard
+        run: python -m pip install ascii-guard
+
+      - name: Lint docs diagrams
+        run: npm run lint:diagrams
+        working-directory: website
+
+      - name: Build Docusaurus
+        run: npm run build
+        working-directory: website

.github/workflows/nix.yml

@@ -0,0 +1,40 @@
+name: Nix
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    paths:
+      - 'flake.nix'
+      - 'flake.lock'
+      - 'nix/**'
+      - 'pyproject.toml'
+      - 'uv.lock'
+      - 'hermes_cli/**'
+      - 'run_agent.py'
+      - 'acp_adapter/**'
+
+concurrency:
+  group: nix-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  nix:
+    strategy:
+      matrix:
+        os: [ubuntu-latest, macos-latest]
+    runs-on: ${{ matrix.os }}
+    timeout-minutes: 30
+    steps:
+      - uses: actions/checkout@v4
+      - uses: DeterminateSystems/nix-installer-action@main
+      - uses: DeterminateSystems/magic-nix-cache-action@main
+      - name: Check flake
+        if: runner.os == 'Linux'
+        run: nix flake check --print-build-logs
+      - name: Build package
+        if: runner.os == 'Linux'
+        run: nix build --print-build-logs
+      - name: Evaluate flake (macOS)
+        if: runner.os == 'macOS'
+        run: nix flake show --json > /dev/null

.github/workflows/supply-chain-audit.yml

@@ -0,0 +1,192 @@
+name: Supply Chain Audit
+
+on:
+  pull_request:
+    types: [opened, synchronize, reopened]
+
+permissions:
+  pull-requests: write
+  contents: read
+
+jobs:
+  scan:
+    name: Scan PR for supply chain risks
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Scan diff for suspicious patterns
+        id: scan
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          set -euo pipefail
+
+          BASE="${{ github.event.pull_request.base.sha }}"
+          HEAD="${{ github.event.pull_request.head.sha }}"
+
+          # Get the full diff (added lines only)
+          DIFF=$(git diff "$BASE".."$HEAD" -- . ':!uv.lock' ':!*.lock' ':!package-lock.json' ':!yarn.lock' || true)
+
+          FINDINGS=""
+          CRITICAL=false
+
+          # --- .pth files (auto-execute on Python startup) ---
+          PTH_FILES=$(git diff --name-only "$BASE".."$HEAD" | grep '\.pth$' || true)
+          if [ -n "$PTH_FILES" ]; then
+            CRITICAL=true
+            FINDINGS="${FINDINGS}
+          ### 🚨 CRITICAL: .pth file added or modified
+          Python \`.pth\` files in \`site-packages/\` execute automatically when the interpreter starts — no import required. This is the exact mechanism used in the [litellm supply chain attack](https://github.com/BerriAI/litellm/issues/24512).
+
+          **Files:**
+          \`\`\`
+          ${PTH_FILES}
+          \`\`\`
+          "
+          fi
+
+          # --- base64 + exec/eval combo (the litellm attack pattern) ---
+          B64_EXEC_HITS=$(echo "$DIFF" | grep -n '^\+' | grep -iE 'base64\.(b64decode|decodebytes|urlsafe_b64decode)' | grep -iE 'exec\(|eval\(' | head -10 || true)
+          if [ -n "$B64_EXEC_HITS" ]; then
+            CRITICAL=true
+            FINDINGS="${FINDINGS}
+          ### 🚨 CRITICAL: base64 decode + exec/eval combo
+          This is the exact pattern used in the [litellm supply chain attack](https://github.com/BerriAI/litellm/issues/24512) — base64-decoded strings passed to exec/eval to hide credential-stealing payloads.
+
+          **Matches:**
+          \`\`\`
+          ${B64_EXEC_HITS}
+          \`\`\`
+          "
+          fi
+
+          # --- base64 decode/encode (alone — legitimate uses exist) ---
+          B64_HITS=$(echo "$DIFF" | grep -n '^\+' | grep -iE 'base64\.(b64decode|b64encode|decodebytes|encodebytes|urlsafe_b64decode)|atob\(|btoa\(|Buffer\.from\(.*base64' | head -20 || true)
+          if [ -n "$B64_HITS" ]; then
+            FINDINGS="${FINDINGS}
+          ### ⚠️ WARNING: base64 encoding/decoding detected
+          Base64 has legitimate uses (images, JWT, etc.) but is also commonly used to obfuscate malicious payloads. Verify the usage is appropriate.
+
+          **Matches (first 20):**
+          \`\`\`
+          ${B64_HITS}
+          \`\`\`
+          "
+          fi
+
+          # --- exec/eval with string arguments ---
+          EXEC_HITS=$(echo "$DIFF" | grep -n '^\+' | grep -E '(exec|eval)\s*\(' | grep -v '^\+\s*#' | grep -v 'test_\|mock\|assert\|# ' | head -20 || true)
+          if [ -n "$EXEC_HITS" ]; then
+            FINDINGS="${FINDINGS}
+          ### ⚠️ WARNING: exec() or eval() usage
+          Dynamic code execution can hide malicious behavior, especially when combined with base64 or network fetches.
+
+          **Matches (first 20):**
+          \`\`\`
+          ${EXEC_HITS}
+          \`\`\`
+          "
+          fi
+
+          # --- subprocess with encoded/obfuscated commands ---
+          PROC_HITS=$(echo "$DIFF" | grep -n '^\+' | grep -E 'subprocess\.(Popen|call|run)\s*\(' | grep -iE 'base64|decode|encode|\\x|chr\(' | head -10 || true)
+          if [ -n "$PROC_HITS" ]; then
+            CRITICAL=true
+            FINDINGS="${FINDINGS}
+          ### 🚨 CRITICAL: subprocess with encoded/obfuscated command
+          Subprocess calls with encoded arguments are a strong indicator of payload execution.
+
+          **Matches:**
+          \`\`\`
+          ${PROC_HITS}
+          \`\`\`
+          "
+          fi
+
+          # --- Network calls to non-standard domains ---
+          EXFIL_HITS=$(echo "$DIFF" | grep -n '^\+' | grep -iE 'requests\.(post|put)\(|httpx\.(post|put)\(|urllib\.request\.urlopen' | grep -v '^\+\s*#' | grep -v 'test_\|mock\|assert' | head -10 || true)
+          if [ -n "$EXFIL_HITS" ]; then
+            FINDINGS="${FINDINGS}
+          ### ⚠️ WARNING: Outbound network calls (POST/PUT)
+          Outbound POST/PUT requests in new code could be data exfiltration. Verify the destination URLs are legitimate.
+
+          **Matches (first 10):**
+          \`\`\`
+          ${EXFIL_HITS}
+          \`\`\`
+          "
+          fi
+
+          # --- setup.py / setup.cfg install hooks ---
+          SETUP_HITS=$(git diff --name-only "$BASE".."$HEAD" | grep -E '(setup\.py|setup\.cfg|__init__\.pth|sitecustomize\.py|usercustomize\.py)$' || true)
+          if [ -n "$SETUP_HITS" ]; then
+            FINDINGS="${FINDINGS}
+          ### ⚠️ WARNING: Install hook files modified
+          These files can execute code during package installation or interpreter startup.
+
+          **Files:**
+          \`\`\`
+          ${SETUP_HITS}
+          \`\`\`
+          "
+          fi
+
+          # --- Compile/marshal/pickle (code object injection) ---
+          MARSHAL_HITS=$(echo "$DIFF" | grep -n '^\+' | grep -iE 'marshal\.loads|pickle\.loads|compile\(' | grep -v '^\+\s*#' | grep -v 'test_\|re\.compile\|ast\.compile' | head -10 || true)
+          if [ -n "$MARSHAL_HITS" ]; then
+            FINDINGS="${FINDINGS}
+          ### ⚠️ WARNING: marshal/pickle/compile usage
+          These can deserialize or construct executable code objects.
+
+          **Matches:**
+          \`\`\`
+          ${MARSHAL_HITS}
+          \`\`\`
+          "
+          fi
+
+          # --- Output results ---
+          if [ -n "$FINDINGS" ]; then
+            echo "found=true" >> "$GITHUB_OUTPUT"
+            if [ "$CRITICAL" = true ]; then
+              echo "critical=true" >> "$GITHUB_OUTPUT"
+            else
+              echo "critical=false" >> "$GITHUB_OUTPUT"
+            fi
+            # Write findings to a file (multiline env vars are fragile)
+            echo "$FINDINGS" > /tmp/findings.md
+          else
+            echo "found=false" >> "$GITHUB_OUTPUT"
+            echo "critical=false" >> "$GITHUB_OUTPUT"
+          fi
+
+      - name: Post warning comment
+        if: steps.scan.outputs.found == 'true'
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          SEVERITY="⚠️ Supply Chain Risk Detected"
+          if [ "${{ steps.scan.outputs.critical }}" = "true" ]; then
+            SEVERITY="🚨 CRITICAL Supply Chain Risk Detected"
+          fi
+
+          BODY="## ${SEVERITY}
+
+          This PR contains patterns commonly associated with supply chain attacks. This does **not** mean the PR is malicious — but these patterns require careful human review before merging.
+
+          $(cat /tmp/findings.md)
+
+          ---
+          *Automated scan triggered by [supply-chain-audit](/.github/workflows/supply-chain-audit.yml). If this is a false positive, a maintainer can approve after manual review.*"
+
+          gh pr comment "${{ github.event.pull_request.number }}" --body "$BODY"
+
+      - name: Fail on critical findings
+        if: steps.scan.outputs.critical == 'true'
+        run: |
+          echo "::error::CRITICAL supply chain risk patterns detected in this PR. See the PR comment for details."
+          exit 1

.github/workflows/tests.yml

@@ -0,0 +1,42 @@
+name: Tests
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+
+# Cancel in-progress runs for the same PR/branch
+concurrency:
+  group: tests-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    timeout-minutes: 10
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Install uv
+        uses: astral-sh/setup-uv@v5
+
+      - name: Set up Python 3.11
+        run: uv python install 3.11
+
+      - name: Install dependencies
+        run: |
+          uv venv .venv --python 3.11
+          source .venv/bin/activate
+          uv pip install -e ".[all,dev]"
+
+      - name: Run tests
+        run: |
+          source .venv/bin/activate
+          python -m pytest tests/ -q --ignore=tests/integration --tb=short -n auto
+        env:
+          # Ensure tests don't accidentally call real APIs
+          OPENROUTER_API_KEY: ""
+          OPENAI_API_KEY: ""
+          NOUS_API_KEY: ""

.gitignore

@@ -1,32 +1,60 @@
-/venv/
-/_pycache/
-hecate/
-hecate-lib/
-*.pyc*
-__pycache__/
-.venv/
-.vscode/
-.env
-.env.local
-.env.development.local
-.env.test.local
-.env.production.local
-.env.development
-.env.test
-export*
-__pycache__/model_tools.cpython-310.pyc
-__pycache__/web_tools.cpython-310.pyc
-logs/
-data/
-.pytest_cache/
-tmp/
-temp_vision_images/
-hermes-*/*
-examples/
-tests/quick_test_dataset.jsonl
-tests/sample_dataset.jsonl
-run_datagen_kimik2-thinking.sh
-run_datagen_megascience_glm4-6.sh
-run_datagen_sonnet.sh
-source-data/*
-run_datagen_megascience_glm4-6.sh
+/venv/
+/_pycache/
+*.pyc*
+__pycache__/
+.venv/
+.vscode/
+.env
+.env.local
+.env.development.local
+.env.test.local
+.env.production.local
+.env.development
+.env.test
+export*
+__pycache__/model_tools.cpython-310.pyc
+__pycache__/web_tools.cpython-310.pyc
+logs/
+data/
+.pytest_cache/
+tmp/
+temp_vision_images/
+hermes-*/*
+examples/
+tests/quick_test_dataset.jsonl
+tests/sample_dataset.jsonl
+run_datagen_kimik2-thinking.sh
+run_datagen_megascience_glm4-6.sh
+run_datagen_sonnet.sh
+source-data/*
+run_datagen_megascience_glm4-6.sh
+data/*
+node_modules/
+browser-use/
+agent-browser/
+# Private keys
+*.ppk
+*.pem
+privvy*
+images/
+__pycache__/
+hermes_agent.egg-info/
+wandb/
+testlogs
+
+# CLI config (may contain sensitive SSH paths)
+cli-config.yaml
+
+# Skills Hub state (lives in ~/.hermes/skills/.hub/ at runtime, but just in case)
+skills/.hub/
+ignored/
+.worktrees/
+environments/benchmarks/evals/
+
+# Release script temp files
+.release_notes.md
+mini-swe-agent/
+
+# Nix
+.direnv/
+result

.gitmodules

@@ -0,0 +1,3 @@
+[submodule "tinker-atropos"]
+	path = tinker-atropos
+	url = https://github.com/nousresearch/tinker-atropos

.plans/openai-api-server.md

@@ -0,0 +1,291 @@
+# OpenAI-Compatible API Server for Hermes Agent
+
+## Motivation
+
+Every major chat frontend (Open WebUI 126k★, LobeChat 73k★, LibreChat 34k★,
+AnythingLLM 56k★, NextChat 87k★, ChatBox 39k★, Jan 26k★, HF Chat-UI 8k★,
+big-AGI 7k★) connects to backends via the OpenAI-compatible REST API with
+SSE streaming. By exposing this endpoint, hermes-agent becomes instantly
+usable as a backend for all of them — no custom adapters needed.
+
+## What It Enables
+
+```
+┌──────────────────┐
+│  Open WebUI      │──┐
+│  LobeChat        │  │    POST /v1/chat/completions
+│  LibreChat       │  ├──► Authorization: Bearer <key>     ┌─────────────────┐
+│  AnythingLLM     │  │    {"messages": [...]}             │  hermes-agent   │
+│  NextChat        │  │                                    │  gateway        │
+│  Any OAI client  │──┘    ◄── SSE streaming response      │  (API server)   │
+└──────────────────┘                                        └─────────────────┘
+```
+
+A user would:
+1. Set `API_SERVER_ENABLED=true` in `~/.hermes/.env`
+2. Run `hermes gateway` (API server starts alongside Telegram/Discord/etc.)
+3. Point Open WebUI (or any frontend) at `http://localhost:8642/v1`
+4. Chat with hermes-agent through any OpenAI-compatible UI
+
+## Endpoints
+
+| Method | Path | Purpose |
+|--------|------|---------|
+| POST | `/v1/chat/completions` | Chat with the agent (streaming + non-streaming) |
+| GET | `/v1/models` | List available "models" (returns hermes-agent as a model) |
+| GET | `/health` | Health check |
+
+## Architecture
+
+### Option A: Gateway Platform Adapter (recommended)
+
+Create `gateway/platforms/api_server.py` as a new platform adapter that
+extends `BasePlatformAdapter`. This is the cleanest approach because:
+
+- Reuses all gateway infrastructure (session management, auth, context building)
+- Runs in the same async loop as other adapters
+- Gets message handling, interrupt support, and session persistence for free
+- Follows the established pattern (like Telegram, Discord, etc.)
+- Uses `aiohttp.web` (already a dependency) for the HTTP server
+
+The adapter would start an `aiohttp.web.Application` server in `connect()`
+and route incoming HTTP requests through the standard `handle_message()` pipeline.
+
+### Option B: Standalone Component
+
+A separate HTTP server class in `gateway/api_server.py` that creates its own
+AIAgent instances directly. Simpler but duplicates session/auth logic.
+
+**Recommendation: Option A** — fits the existing architecture, less code to
+maintain, gets all gateway features for free.
+
+## Request/Response Format
+
+### Chat Completions (non-streaming)
+
+```
+POST /v1/chat/completions
+Authorization: Bearer hermes-api-key-here
+Content-Type: application/json
+
+{
+  "model": "hermes-agent",
+  "messages": [
+    {"role": "system", "content": "You are a helpful assistant."},
+    {"role": "user", "content": "What files are in the current directory?"}
+  ],
+  "stream": false,
+  "temperature": 0.7
+}
+```
+
+Response:
+```json
+{
+  "id": "chatcmpl-abc123",
+  "object": "chat.completion",
+  "created": 1710000000,
+  "model": "hermes-agent",
+  "choices": [{
+    "index": 0,
+    "message": {
+      "role": "assistant",
+      "content": "Here are the files in the current directory:\n..."
+    },
+    "finish_reason": "stop"
+  }],
+  "usage": {
+    "prompt_tokens": 50,
+    "completion_tokens": 200,
+    "total_tokens": 250
+  }
+}
+```
+
+### Chat Completions (streaming)
+
+Same request with `"stream": true`. Response is SSE:
+
+```
+data: {"id":"chatcmpl-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"role":"assistant"},"finish_reason":null}]}
+
+data: {"id":"chatcmpl-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"Here "},"finish_reason":null}]}
+
+data: {"id":"chatcmpl-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"are "},"finish_reason":null}]}
+
+data: {"id":"chatcmpl-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}
+
+data: [DONE]
+```
+
+### Models List
+
+```
+GET /v1/models
+Authorization: Bearer hermes-api-key-here
+```
+
+Response:
+```json
+{
+  "object": "list",
+  "data": [{
+    "id": "hermes-agent",
+    "object": "model",
+    "created": 1710000000,
+    "owned_by": "hermes-agent"
+  }]
+}
+```
+
+## Key Design Decisions
+
+### 1. Session Management
+
+The OpenAI API is stateless — each request includes the full conversation.
+But hermes-agent sessions have persistent state (memory, skills, tool context).
+
+**Approach: Hybrid**
+- Default: Stateless. Each request is independent. The `messages` array IS
+  the conversation. No session persistence between requests.
+- Opt-in persistent sessions via `X-Session-ID` header. When provided, the
+  server maintains session state across requests (conversation history,
+  memory context, tool state). This enables richer agent behavior.
+- The session ID also enables interrupt support — a subsequent request with
+  the same session ID while one is running triggers an interrupt.
+
+### 2. Streaming
+
+The agent's `run_conversation()` is synchronous and returns the full response.
+For real SSE streaming, we need to emit chunks as they're generated.
+
+**Phase 1 (MVP):** Run agent in a thread, return the complete response as
+a single SSE chunk + `[DONE]`. This works with all frontends — they just see
+a fast single-chunk response. Not true streaming but functional.
+
+**Phase 2:** Add a response callback to AIAgent that emits text chunks as the
+LLM generates them. The API server captures these via a queue and streams them
+as SSE events. This gives real token-by-token streaming.
+
+**Phase 3:** Stream tool execution progress too — emit tool call/result events
+as the agent works, giving frontends visibility into what the agent is doing.
+
+### 3. Tool Transparency
+
+Two modes:
+- **Opaque (default):** Frontends see only the final response. Tool calls
+  happen server-side and are invisible. Best for general-purpose UIs.
+- **Transparent (opt-in via header):** Tool calls are emitted as OpenAI-format
+  tool_call/tool_result messages in the stream. Useful for agent-aware frontends.
+
+### 4. Authentication
+
+- Bearer token via `Authorization: Bearer <key>` header
+- Token configured via `API_SERVER_KEY` env var
+- Optional: allow unauthenticated local-only access (127.0.0.1 bind)
+- Follows the same pattern as other platform adapters
+
+### 5. Model Mapping
+
+Frontends send `"model": "hermes-agent"` (or whatever). The actual LLM model
+used is configured server-side in config.yaml. The API server maps any
+requested model name to the configured hermes-agent model.
+
+Optionally, allow model passthrough: if the frontend sends
+`"model": "anthropic/claude-sonnet-4"`, the agent uses that model. Controlled
+by a config flag.
+
+## Configuration
+
+```yaml
+# In config.yaml
+api_server:
+  enabled: true
+  port: 8642
+  host: "127.0.0.1"        # localhost only by default
+  key: "your-secret-key"   # or via API_SERVER_KEY env var
+  allow_model_override: false  # let clients choose the model
+  max_concurrent: 5         # max simultaneous requests
+```
+
+Environment variables:
+```bash
+API_SERVER_ENABLED=true
+API_SERVER_PORT=8642
+API_SERVER_HOST=127.0.0.1
+API_SERVER_KEY=your-secret-key
+```
+
+## Implementation Plan
+
+### Phase 1: MVP (non-streaming) — PR
+
+1. `gateway/platforms/api_server.py` — new adapter
+   - aiohttp.web server with endpoints:
+     - `POST /v1/chat/completions` — Chat Completions API (universal compat)
+     - `POST /v1/responses` — Responses API (server-side state, tool preservation)
+     - `GET /v1/models` — list available models
+     - `GET /health` — health check
+   - Bearer token auth middleware
+   - Non-streaming responses (run agent, return full result)
+   - Chat Completions: stateless, messages array is the conversation
+   - Responses API: server-side conversation storage via previous_response_id
+     - Store full internal conversation (including tool calls) keyed by response ID
+     - On subsequent requests, reconstruct full context from stored chain
+   - Frontend system prompt layered on top of hermes-agent's core prompt
+
+2. `gateway/config.py` — add `Platform.API_SERVER` enum + config
+
+3. `gateway/run.py` — register adapter in `_create_adapter()`
+
+4. Tests in `tests/gateway/test_api_server.py`
+
+### Phase 2: SSE Streaming
+
+1. Add response streaming to both endpoints
+   - Chat Completions: `choices[0].delta.content` SSE format
+   - Responses API: semantic events (response.output_text.delta, etc.)
+   - Run agent in thread, collect output via callback queue
+   - Handle client disconnect (cancel agent)
+
+2. Add `stream_callback` parameter to `AIAgent.run_conversation()`
+
+### Phase 3: Enhanced Features
+
+1. Tool call transparency mode (opt-in)
+2. Model passthrough/override
+3. Concurrent request limiting
+4. Usage tracking / rate limiting
+5. CORS headers for browser-based frontends
+6. GET /v1/responses/{id} — retrieve stored response
+7. DELETE /v1/responses/{id} — delete stored response
+
+## Files Changed
+
+| File | Change |
+|------|--------|
+| `gateway/platforms/api_server.py` | NEW — main adapter (~300 lines) |
+| `gateway/config.py` | Add Platform.API_SERVER + config (~20 lines) |
+| `gateway/run.py` | Register adapter in _create_adapter() (~10 lines) |
+| `tests/gateway/test_api_server.py` | NEW — tests (~200 lines) |
+| `cli-config.yaml.example` | Add api_server section |
+| `README.md` | Mention API server in platform list |
+
+## Compatibility Matrix
+
+Once implemented, hermes-agent works as a drop-in backend for:
+
+| Frontend | Stars | How to Connect |
+|----------|-------|---------------|
+| Open WebUI | 126k | Settings → Connections → Add OpenAI API, URL: `http://localhost:8642/v1` |
+| NextChat | 87k | BASE_URL env var |
+| LobeChat | 73k | Custom provider endpoint |
+| AnythingLLM | 56k | LLM Provider → Generic OpenAI |
+| Oobabooga | 42k | Already a backend, not a frontend |
+| ChatBox | 39k | API Host setting |
+| LibreChat | 34k | librechat.yaml custom endpoint |
+| Chatbot UI | 29k | Custom API endpoint |
+| Jan | 26k | Remote model config |
+| AionUI | 18k | Custom API endpoint |
+| HF Chat-UI | 8k | OPENAI_BASE_URL env var |
+| big-AGI | 7k | Custom endpoint |

.plans/streaming-support.md

@@ -0,0 +1,705 @@
+# Streaming LLM Response Support for Hermes Agent
+
+## Overview
+
+Add token-by-token streaming of LLM responses across all platforms. When enabled,
+users see the response typing out live instead of waiting for the full generation.
+Streaming is opt-in via config, defaults to off, and all existing non-streaming
+code paths remain intact as the default.
+
+## Design Principles
+
+1. **Feature-flagged**: `streaming.enabled: true` in config.yaml. Off by default.
+   When off, all existing code paths are unchanged — zero risk to current behavior.
+2. **Callback-based**: A simple `stream_callback(text_delta: str)` function injected
+   into AIAgent. The agent doesn't know or care what the consumer does with tokens.
+3. **Graceful degradation**: If the provider doesn't support streaming, or streaming
+   fails for any reason, silently fall back to the non-streaming path.
+4. **Platform-agnostic core**: The streaming mechanism in AIAgent works the same
+   regardless of whether the consumer is CLI, Telegram, Discord, or the API server.
+
+---
+
+## Architecture
+
+```
+                              stream_callback(delta)
+                                    │
+  ┌─────────────┐    ┌─────────────▼──────────────┐
+  │  LLM API    │    │      queue.Queue()          │
+  │  (stream)   │───►│  thread-safe bridge between │
+  │             │    │  agent thread & consumer    │
+  └─────────────┘    └─────────────┬──────────────┘
+                                   │
+                    ┌──────────────┼──────────────┐
+                    │              │              │
+              ┌─────▼─────┐ ┌─────▼─────┐ ┌─────▼─────┐
+              │    CLI     │ │  Gateway  │ │ API Server│
+              │ print to   │ │ edit msg  │ │ SSE event │
+              │ terminal   │ │ on Tg/Dc  │ │ to client │
+              └───────────┘ └───────────┘ └───────────┘
+```
+
+The agent runs in a thread. The callback puts tokens into a thread-safe queue.
+Each consumer reads the queue in its own context (async task, main thread, etc.).
+
+---
+
+## Configuration
+
+### config.yaml
+
+```yaml
+streaming:
+  enabled: false          # Master switch. Default off.
+  # Per-platform overrides (optional):
+  # cli: true             # Override for CLI only
+  # telegram: true        # Override for Telegram only
+  # discord: false        # Keep Discord non-streaming
+  # api_server: true      # Override for API server
+```
+
+### Environment variables
+
+```
+HERMES_STREAMING_ENABLED=true    # Master switch via env
+```
+
+### How the flag is read
+
+- **CLI**: `load_cli_config()` reads `streaming.enabled`, sets env var. AIAgent
+  checks at init time.
+- **Gateway**: `_run_agent()` reads config, decides whether to pass
+  `stream_callback` to the AIAgent constructor.
+- **API server**: For Chat Completions `stream=true` requests, always uses streaming
+  regardless of config (the client is explicitly requesting it). For non-stream
+  requests, uses config.
+
+### Precedence
+
+1. API server: client's `stream` field overrides everything
+2. Per-platform config override (e.g., `streaming.telegram: true`)
+3. Master `streaming.enabled` flag
+4. Default: off
+
+---
+
+## Implementation Plan
+
+### Phase 1: Core streaming infrastructure in AIAgent
+
+**File: run_agent.py**
+
+#### 1a. Add stream_callback parameter to __init__ (~5 lines)
+
+```python
+def __init__(self, ..., stream_callback: callable = None, ...):
+    self.stream_callback = stream_callback
+```
+
+No other init changes. The callback is optional — when None, everything
+works exactly as before.
+
+#### 1b. Add _run_streaming_chat_completion() method (~65 lines)
+
+New method for Chat Completions API streaming:
+
+```python
+def _run_streaming_chat_completion(self, api_kwargs: dict):
+    """Stream a chat completion, emitting text tokens via stream_callback.
+    
+    Returns a fake response object compatible with the non-streaming code path.
+    Falls back to non-streaming on any error.
+    """
+    stream_kwargs = dict(api_kwargs)
+    stream_kwargs["stream"] = True
+    stream_kwargs["stream_options"] = {"include_usage": True}
+    
+    accumulated_content = []
+    accumulated_tool_calls = {}  # index -> {id, name, arguments}
+    final_usage = None
+    
+    try:
+        stream = self.client.chat.completions.create(**stream_kwargs)
+        
+        for chunk in stream:
+            if not chunk.choices:
+                # Usage-only chunk (final)
+                if chunk.usage:
+                    final_usage = chunk.usage
+                continue
+            
+            delta = chunk.choices[0].delta
+            
+            # Text content — emit via callback
+            if delta.content:
+                accumulated_content.append(delta.content)
+                if self.stream_callback:
+                    try:
+                        self.stream_callback(delta.content)
+                    except Exception:
+                        pass
+            
+            # Tool call deltas — accumulate silently
+            if delta.tool_calls:
+                for tc_delta in delta.tool_calls:
+                    idx = tc_delta.index
+                    if idx not in accumulated_tool_calls:
+                        accumulated_tool_calls[idx] = {
+                            "id": tc_delta.id or "",
+                            "name": "", "arguments": ""
+                        }
+                    if tc_delta.function:
+                        if tc_delta.function.name:
+                            accumulated_tool_calls[idx]["name"] = tc_delta.function.name
+                        if tc_delta.function.arguments:
+                            accumulated_tool_calls[idx]["arguments"] += tc_delta.function.arguments
+        
+        # Build fake response compatible with existing code
+        tool_calls = []
+        for idx in sorted(accumulated_tool_calls):
+            tc = accumulated_tool_calls[idx]
+            if tc["name"]:
+                tool_calls.append(SimpleNamespace(
+                    id=tc["id"], type="function",
+                    function=SimpleNamespace(name=tc["name"], arguments=tc["arguments"]),
+                ))
+        
+        return SimpleNamespace(
+            choices=[SimpleNamespace(
+                message=SimpleNamespace(
+                    content="".join(accumulated_content) or "",
+                    tool_calls=tool_calls or None,
+                    role="assistant",
+                ),
+                finish_reason="tool_calls" if tool_calls else "stop",
+            )],
+            usage=final_usage,
+            model=self.model,
+        )
+    
+    except Exception as e:
+        logger.debug("Streaming failed, falling back to non-streaming: %s", e)
+        return self.client.chat.completions.create(**api_kwargs)
+```
+
+#### 1c. Modify _run_codex_stream() for Responses API (~10 lines)
+
+The method already iterates the stream. Add callback emission:
+
+```python
+def _run_codex_stream(self, api_kwargs: dict):
+    with self.client.responses.stream(**api_kwargs) as stream:
+        for event in stream:
+            # Emit text deltas if streaming callback is set
+            if self.stream_callback and hasattr(event, 'type'):
+                if event.type == 'response.output_text.delta':
+                    try:
+                        self.stream_callback(event.delta)
+                    except Exception:
+                        pass
+        return stream.get_final_response()
+```
+
+#### 1d. Modify _interruptible_api_call() (~5 lines)
+
+Add the streaming branch:
+
+```python
+def _call():
+    try:
+        if self.api_mode == "codex_responses":
+            result["response"] = self._run_codex_stream(api_kwargs)
+        elif self.stream_callback is not None:
+            result["response"] = self._run_streaming_chat_completion(api_kwargs)
+        else:
+            result["response"] = self.client.chat.completions.create(**api_kwargs)
+    except Exception as e:
+        result["error"] = e
+```
+
+#### 1e. Signal end-of-stream to consumers (~5 lines)
+
+After the API call returns, signal the callback that streaming is done
+so consumers can finalize (remove cursor, close SSE, etc.):
+
+```python
+# In run_conversation(), after _interruptible_api_call returns:
+if self.stream_callback:
+    try:
+        self.stream_callback(None)  # None = end of stream signal
+    except Exception:
+        pass
+```
+
+Consumers check: `if delta is None: finalize()`
+
+**Tests for Phase 1:** (~150 lines)
+- Test _run_streaming_chat_completion with mocked stream
+- Test fallback to non-streaming on error
+- Test tool_call accumulation during streaming
+- Test stream_callback receives correct deltas
+- Test None signal at end of stream
+- Test streaming disabled when callback is None
+
+---
+
+### Phase 2: Gateway consumers (Telegram, Discord, etc.)
+
+**File: gateway/run.py**
+
+#### 2a. Read streaming config (~15 lines)
+
+In `_run_agent()`, before creating the AIAgent:
+
+```python
+# Read streaming config
+_streaming_enabled = False
+try:
+    # Check per-platform override first
+    platform_key = source.platform.value if source.platform else ""
+    _stream_cfg = {}  # loaded from config.yaml streaming section
+    if _stream_cfg.get(platform_key) is not None:
+        _streaming_enabled = bool(_stream_cfg[platform_key])
+    else:
+        _streaming_enabled = bool(_stream_cfg.get("enabled", False))
+except Exception:
+    pass
+# Env var override
+if os.getenv("HERMES_STREAMING_ENABLED", "").lower() in ("true", "1", "yes"):
+    _streaming_enabled = True
+```
+
+#### 2b. Set up queue + callback (~15 lines)
+
+```python
+_stream_q = None
+_stream_done = None
+_stream_msg_id = [None]  # mutable ref for the async task
+
+if _streaming_enabled:
+    import queue as _q
+    _stream_q = _q.Queue()
+    _stream_done = threading.Event()
+    
+    def _on_token(delta):
+        if delta is None:
+            _stream_done.set()
+        else:
+            _stream_q.put(delta)
+```
+
+Pass `stream_callback=_on_token` to the AIAgent constructor.
+
+#### 2c. Telegram/Discord stream preview task (~50 lines)
+
+```python
+async def stream_preview():
+    """Progressively edit a message with streaming tokens."""
+    if not _stream_q:
+        return
+    adapter = self.adapters.get(source.platform)
+    if not adapter:
+        return
+    
+    accumulated = []
+    token_count = 0
+    last_edit = 0.0
+    MIN_TOKENS = 20          # Don't show until enough context
+    EDIT_INTERVAL = 1.5      # Respect Telegram rate limits
+    
+    try:
+        while not _stream_done.is_set():
+            try:
+                chunk = _stream_q.get(timeout=0.1)
+                accumulated.append(chunk)
+                token_count += 1
+            except queue.Empty:
+                continue
+            
+            now = time.monotonic()
+            if token_count >= MIN_TOKENS and (now - last_edit) >= EDIT_INTERVAL:
+                preview = "".join(accumulated) + " ▌"
+                if _stream_msg_id[0] is None:
+                    r = await adapter.send(
+                        chat_id=source.chat_id,
+                        content=preview,
+                        metadata=_thread_metadata,
+                    )
+                    if r.success and r.message_id:
+                        _stream_msg_id[0] = r.message_id
+                else:
+                    await adapter.edit_message(
+                        chat_id=source.chat_id,
+                        message_id=_stream_msg_id[0],
+                        content=preview,
+                    )
+                last_edit = now
+        
+        # Drain remaining tokens
+        while not _stream_q.empty():
+            accumulated.append(_stream_q.get_nowait())
+        
+        # Final edit — remove cursor, show complete text
+        if _stream_msg_id[0] and accumulated:
+            await adapter.edit_message(
+                chat_id=source.chat_id,
+                message_id=_stream_msg_id[0],
+                content="".join(accumulated),
+            )
+    
+    except asyncio.CancelledError:
+        # Clean up on cancel
+        if _stream_msg_id[0] and accumulated:
+            try:
+                await adapter.edit_message(
+                    chat_id=source.chat_id,
+                    message_id=_stream_msg_id[0],
+                    content="".join(accumulated),
+                )
+            except Exception:
+                pass
+    except Exception as e:
+        logger.debug("stream_preview error: %s", e)
+```
+
+#### 2d. Skip final send if already streamed (~10 lines)
+
+In `_process_message_background()` (base.py), after getting the response,
+if streaming was active and `_stream_msg_id[0]` is set, the final response
+was already delivered via progressive edits. Skip the normal `self.send()`
+call to avoid duplicating the message.
+
+This is the most delicate integration point — we need to communicate from
+the gateway's `_run_agent` back to the base adapter's response sender that
+the response was already delivered. Options:
+
+- **Option A**: Return a special marker in the result dict:
+  `result["_streamed_msg_id"] = _stream_msg_id[0]`
+  The base adapter checks this and skips `send()`.
+  
+- **Option B**: Edit the already-sent message with the final response
+  (which may differ slightly from accumulated tokens due to think-block
+  stripping, etc.) and don't send a new one.
+
+- **Option C**: The stream preview task handles the FULL final response
+  (including any post-processing), and the handler returns None to skip
+  the normal send path.
+
+Recommended: **Option A** — cleanest separation. The result dict already
+carries metadata; adding one more field is low-risk.
+
+**Platform-specific considerations:**
+
+| Platform | Edit support | Rate limits | Streaming approach |
+|----------|-------------|-------------|-------------------|
+| Telegram | ✅ edit_message_text | ~20 edits/min | Edit every 1.5s |
+| Discord | ✅ message.edit | 5 edits/5s per message | Edit every 1.2s |
+| Slack | ✅ chat.update | Tier 3 (~50/min) | Edit every 1.5s |
+| WhatsApp | ❌ no edit support | N/A | Skip streaming, use normal path |
+| HomeAssistant | ❌ no edit | N/A | Skip streaming |
+| API Server | ✅ SSE native | No limit | Real SSE events |
+
+WhatsApp and HomeAssistant fall back to non-streaming automatically because
+they don't support message editing.
+
+**Tests for Phase 2:** (~100 lines)
+- Test stream_preview sends/edits correctly
+- Test skip-final-send when streaming delivered
+- Test WhatsApp/HA graceful fallback
+- Test streaming disabled per-platform config
+- Test thread_id metadata forwarded in stream messages
+
+---
+
+### Phase 3: CLI streaming
+
+**File: cli.py**
+
+#### 3a. Set up callback in the CLI chat loop (~20 lines)
+
+In `_chat_once()` or wherever the agent is invoked:
+
+```python
+if streaming_enabled:
+    _stream_q = queue.Queue()
+    _stream_done = threading.Event()
+    
+    def _cli_stream_callback(delta):
+        if delta is None:
+            _stream_done.set()
+        else:
+            _stream_q.put(delta)
+    
+    agent.stream_callback = _cli_stream_callback
+```
+
+#### 3b. Token display thread/task (~30 lines)
+
+Start a thread that reads the queue and prints tokens:
+
+```python
+def _stream_display():
+    """Print tokens to terminal as they arrive."""
+    first_token = True
+    while not _stream_done.is_set():
+        try:
+            delta = _stream_q.get(timeout=0.1)
+        except queue.Empty:
+            continue
+        if first_token:
+            # Print response box top border
+            _cprint(f"\n{top}")
+            first_token = False
+        sys.stdout.write(delta)
+        sys.stdout.flush()
+    # Drain remaining
+    while not _stream_q.empty():
+        sys.stdout.write(_stream_q.get_nowait())
+    sys.stdout.flush()
+    # Print bottom border
+    _cprint(f"\n\n{bot}")
+```
+
+**Integration challenge: prompt_toolkit**
+
+The CLI uses prompt_toolkit which controls the terminal. Writing directly
+to stdout while prompt_toolkit is active can cause display corruption.
+The existing KawaiiSpinner already solves this by using prompt_toolkit's
+`patch_stdout` context. The streaming display would need to do the same.
+
+Alternative: use `_cprint()` for each token chunk (routes through
+prompt_toolkit's renderer). But this might be slow for individual tokens.
+
+Recommended approach: accumulate tokens in small batches (e.g., every 50ms)
+and `_cprint()` the batch. This balances display responsiveness with
+prompt_toolkit compatibility.
+
+**Tests for Phase 3:** (~50 lines)
+- Test CLI streaming callback setup
+- Test response box borders with streaming
+- Test fallback when streaming disabled
+
+---
+
+### Phase 4: API Server real streaming
+
+**File: gateway/platforms/api_server.py**
+
+Replace the pseudo-streaming `_write_sse_chat_completion()` with real
+token-by-token SSE when the agent supports it.
+
+#### 4a. Wire streaming callback for stream=true requests (~20 lines)
+
+```python
+if stream:
+    _stream_q = queue.Queue()
+    
+    def _api_stream_callback(delta):
+        _stream_q.put(delta)  # None = done
+    
+    # Pass callback to _run_agent
+    result, usage = await self._run_agent(
+        ..., stream_callback=_api_stream_callback,
+    )
+```
+
+#### 4b. Real SSE writer (~40 lines)
+
+```python
+async def _write_real_sse(self, request, completion_id, model, stream_q):
+    response = web.StreamResponse(
+        headers={"Content-Type": "text/event-stream", "Cache-Control": "no-cache"},
+    )
+    await response.prepare(request)
+    
+    # Role chunk
+    await response.write(...)
+    
+    # Stream content chunks as they arrive
+    while True:
+        try:
+            delta = await asyncio.get_event_loop().run_in_executor(
+                None, lambda: stream_q.get(timeout=0.1)
+            )
+        except queue.Empty:
+            continue
+        
+        if delta is None:  # End of stream
+            break
+        
+        chunk = {"id": completion_id, "object": "chat.completion.chunk", ...
+                 "choices": [{"delta": {"content": delta}, ...}]}
+        await response.write(f"data: {json.dumps(chunk)}\n\n".encode())
+    
+    # Finish + [DONE]
+    await response.write(...)
+    await response.write(b"data: [DONE]\n\n")
+    return response
+```
+
+**Challenge: concurrent execution**
+
+The agent runs in a thread executor. SSE writing happens in the async event
+loop. The queue bridges them. But `_run_agent()` currently awaits the full
+result before returning. For real streaming, we need to start the agent in
+the background and stream tokens while it runs:
+
+```python
+# Start agent in background
+agent_task = asyncio.create_task(self._run_agent_async(...))
+
+# Stream tokens while agent runs
+await self._write_real_sse(request, ..., stream_q)
+
+# Agent is done by now (stream_q received None)
+result, usage = await agent_task
+```
+
+This requires splitting `_run_agent` into an async version that doesn't
+block waiting for the result, or running it in a separate task.
+
+**Responses API SSE format:**
+
+For `/v1/responses` with `stream=true`, the SSE events are different:
+
+```
+event: response.output_text.delta
+data: {"type":"response.output_text.delta","delta":"Hello"}
+
+event: response.completed  
+data: {"type":"response.completed","response":{...}}
+```
+
+This needs a separate SSE writer that emits Responses API format events.
+
+**Tests for Phase 4:** (~80 lines)
+- Test real SSE streaming with mocked agent
+- Test SSE event format (Chat Completions vs Responses)
+- Test client disconnect during streaming
+- Test fallback to pseudo-streaming when callback not available
+
+---
+
+## Integration Issues & Edge Cases
+
+### 1. Tool calls during streaming
+
+When the model returns tool calls instead of text, no text tokens are emitted.
+The stream_callback is simply never called with text. After tools execute, the
+next API call may produce the final text response — streaming picks up again.
+
+The stream preview task needs to handle this: if no tokens arrive during a
+tool-call round, don't send/edit any message. The tool progress messages
+continue working as before.
+
+### 2. Duplicate messages
+
+The biggest risk: the agent sends the final response normally (via the
+existing send path) AND the stream preview already showed it. The user
+sees the response twice.
+
+Prevention: when streaming is active and tokens were delivered, the final
+response send must be suppressed. The `result["_streamed_msg_id"]` marker
+tells the base adapter to skip its normal send.
+
+### 3. Response post-processing
+
+The final response may differ from the accumulated streamed tokens:
+- Think block stripping (`<think>...</think>` removed)
+- Trailing whitespace cleanup
+- Tool result media tag appending
+
+The stream preview shows raw tokens. The final edit should use the
+post-processed version. This means the final edit (removing the cursor)
+should use the post-processed `final_response`, not just the accumulated
+stream text.
+
+### 4. Context compression during streaming
+
+If the agent triggers context compression mid-conversation, the streaming
+tokens from BEFORE compression are from a different context than those
+after. This isn't a problem in practice — compression happens between
+API calls, not during streaming.
+
+### 5. Interrupt during streaming
+
+User sends a new message while streaming → interrupt. The stream is killed
+(HTTP connection closed), accumulated tokens are shown as-is (no cursor),
+and the interrupt message is processed normally. This is already handled by
+`_interruptible_api_call` closing the client.
+
+### 6. Multi-model / fallback
+
+If the primary model fails and the agent falls back to a different model,
+streaming state resets. The fallback call may or may not support streaming.
+The graceful fallback in `_run_streaming_chat_completion` handles this.
+
+### 7. Rate limiting on edits
+
+Telegram: ~20 edits/minute (~1 every 3 seconds to be safe)
+Discord: 5 edits per 5 seconds per message
+Slack: ~50 API calls/minute
+
+The 1.5s edit interval is conservative enough for all platforms. If we get
+429 rate limit errors on edits, just skip that edit cycle and try next time.
+
+---
+
+## Files Changed Summary
+
+| File | Phase | Changes |
+|------|-------|---------|
+| `run_agent.py` | 1 | +stream_callback param, +_run_streaming_chat_completion(), modify _run_codex_stream(), modify _interruptible_api_call() |
+| `gateway/run.py` | 2 | +streaming config reader, +queue/callback setup, +stream_preview task, +skip-final-send logic |
+| `gateway/platforms/base.py` | 2 | +check for _streamed_msg_id in response handler |
+| `cli.py` | 3 | +streaming setup, +token display, +response box integration |
+| `gateway/platforms/api_server.py` | 4 | +real SSE writer, +streaming callback wiring |
+| `hermes_cli/config.py` | 1 | +streaming config defaults |
+| `cli-config.yaml.example` | 1 | +streaming section |
+| `tests/test_streaming.py` | 1-4 | NEW — ~380 lines of tests |
+
+**Total new code**: ~500 lines across all phases
+**Total test code**: ~380 lines
+
+---
+
+## Rollout Plan
+
+1. **Phase 1** (core): Merge to main. Streaming disabled by default.
+   Zero impact on existing behavior. Can be tested with env var.
+
+2. **Phase 2** (gateway): Merge to main. Test on Telegram manually.
+   Enable per-platform: `streaming.telegram: true` in config.
+
+3. **Phase 3** (CLI): Merge to main. Test in terminal.
+   Enable: `streaming.cli: true` or `streaming.enabled: true`.
+
+4. **Phase 4** (API server): Merge to main. Test with Open WebUI.
+   Auto-enabled when client sends `stream: true`.
+
+Each phase is independently mergeable and testable. Streaming stays
+off by default throughout. Once all phases are stable, consider
+changing the default to enabled.
+
+---
+
+## Config Reference (final state)
+
+```yaml
+# config.yaml
+streaming:
+  enabled: false          # Master switch (default: off)
+  cli: true               # Per-platform override
+  telegram: true
+  discord: true
+  slack: true
+  api_server: true        # API server always streams when client requests it
+  edit_interval: 1.5      # Seconds between message edits (default: 1.5)
+  min_tokens: 20          # Tokens before first display (default: 20)
+```
+
+```bash
+# Environment variable override
+HERMES_STREAMING_ENABLED=true
+```

AGENTS.md

@@ -0,0 +1,469 @@
+# Hermes Agent - Development Guide
+
+Instructions for AI coding assistants and developers working on the hermes-agent codebase.
+
+## Development Environment
+
+```bash
+source venv/bin/activate  # ALWAYS activate before running Python
+```
+
+## Project Structure
+
+```
+hermes-agent/
+├── run_agent.py          # AIAgent class — core conversation loop
+├── model_tools.py        # Tool orchestration, _discover_tools(), handle_function_call()
+├── toolsets.py           # Toolset definitions, _HERMES_CORE_TOOLS list
+├── cli.py                # HermesCLI class — interactive CLI orchestrator
+├── hermes_state.py       # SessionDB — SQLite session store (FTS5 search)
+├── agent/                # Agent internals
+│   ├── prompt_builder.py     # System prompt assembly
+│   ├── context_compressor.py # Auto context compression
+│   ├── prompt_caching.py     # Anthropic prompt caching
+│   ├── auxiliary_client.py   # Auxiliary LLM client (vision, summarization)
+│   ├── model_metadata.py     # Model context lengths, token estimation
+│   ├── models_dev.py         # models.dev registry integration (provider-aware context)
+│   ├── display.py            # KawaiiSpinner, tool preview formatting
+│   ├── skill_commands.py     # Skill slash commands (shared CLI/gateway)
+│   └── trajectory.py         # Trajectory saving helpers
+├── hermes_cli/           # CLI subcommands and setup
+│   ├── main.py           # Entry point — all `hermes` subcommands
+│   ├── config.py         # DEFAULT_CONFIG, OPTIONAL_ENV_VARS, migration
+│   ├── commands.py       # Slash command definitions + SlashCommandCompleter
+│   ├── callbacks.py      # Terminal callbacks (clarify, sudo, approval)
+│   ├── setup.py          # Interactive setup wizard
+│   ├── skin_engine.py    # Skin/theme engine — CLI visual customization
+│   ├── skills_config.py  # `hermes skills` — enable/disable skills per platform
+│   ├── tools_config.py   # `hermes tools` — enable/disable tools per platform
+│   ├── skills_hub.py     # `/skills` slash command (search, browse, install)
+│   ├── models.py         # Model catalog, provider model lists
+│   ├── model_switch.py   # Shared /model switch pipeline (CLI + gateway)
+│   └── auth.py           # Provider credential resolution
+├── tools/                # Tool implementations (one file per tool)
+│   ├── registry.py       # Central tool registry (schemas, handlers, dispatch)
+│   ├── approval.py       # Dangerous command detection
+│   ├── terminal_tool.py  # Terminal orchestration
+│   ├── process_registry.py # Background process management
+│   ├── file_tools.py     # File read/write/search/patch
+│   ├── web_tools.py      # Web search/extract (Parallel + Firecrawl)
+│   ├── browser_tool.py   # Browserbase browser automation
+│   ├── code_execution_tool.py # execute_code sandbox
+│   ├── delegate_tool.py  # Subagent delegation
+│   ├── mcp_tool.py       # MCP client (~1050 lines)
+│   └── environments/     # Terminal backends (local, docker, ssh, modal, daytona, singularity)
+├── gateway/              # Messaging platform gateway
+│   ├── run.py            # Main loop, slash commands, message dispatch
+│   ├── session.py        # SessionStore — conversation persistence
+│   └── platforms/        # Adapters: telegram, discord, slack, whatsapp, homeassistant, signal
+├── acp_adapter/          # ACP server (VS Code / Zed / JetBrains integration)
+├── cron/                 # Scheduler (jobs.py, scheduler.py)
+├── environments/         # RL training environments (Atropos)
+├── tests/                # Pytest suite (~3000 tests)
+└── batch_runner.py       # Parallel batch processing
+```
+
+**User config:** `~/.hermes/config.yaml` (settings), `~/.hermes/.env` (API keys)
+
+## File Dependency Chain
+
+```
+tools/registry.py  (no deps — imported by all tool files)
+       ↑
+tools/*.py  (each calls registry.register() at import time)
+       ↑
+model_tools.py  (imports tools/registry + triggers tool discovery)
+       ↑
+run_agent.py, cli.py, batch_runner.py, environments/
+```
+
+---
+
+## AIAgent Class (run_agent.py)
+
+```python
+class AIAgent:
+    def __init__(self,
+        model: str = "anthropic/claude-opus-4.6",
+        max_iterations: int = 90,
+        enabled_toolsets: list = None,
+        disabled_toolsets: list = None,
+        quiet_mode: bool = False,
+        save_trajectories: bool = False,
+        platform: str = None,           # "cli", "telegram", etc.
+        session_id: str = None,
+        skip_context_files: bool = False,
+        skip_memory: bool = False,
+        # ... plus provider, api_mode, callbacks, routing params
+    ): ...
+
+    def chat(self, message: str) -> str:
+        """Simple interface — returns final response string."""
+
+    def run_conversation(self, user_message: str, system_message: str = None,
+                         conversation_history: list = None, task_id: str = None) -> dict:
+        """Full interface — returns dict with final_response + messages."""
+```
+
+### Agent Loop
+
+The core loop is inside `run_conversation()` — entirely synchronous:
+
+```python
+while api_call_count < self.max_iterations and self.iteration_budget.remaining > 0:
+    response = client.chat.completions.create(model=model, messages=messages, tools=tool_schemas)
+    if response.tool_calls:
+        for tool_call in response.tool_calls:
+            result = handle_function_call(tool_call.name, tool_call.args, task_id)
+            messages.append(tool_result_message(result))
+        api_call_count += 1
+    else:
+        return response.content
+```
+
+Messages follow OpenAI format: `{"role": "system/user/assistant/tool", ...}`. Reasoning content is stored in `assistant_msg["reasoning"]`.
+
+---
+
+## CLI Architecture (cli.py)
+
+- **Rich** for banner/panels, **prompt_toolkit** for input with autocomplete
+- **KawaiiSpinner** (`agent/display.py`) — animated faces during API calls, `┊` activity feed for tool results
+- `load_cli_config()` in cli.py merges hardcoded defaults + user config YAML
+- **Skin engine** (`hermes_cli/skin_engine.py`) — data-driven CLI theming; initialized from `display.skin` config key at startup; skins customize banner colors, spinner faces/verbs/wings, tool prefix, response box, branding text
+- `process_command()` is a method on `HermesCLI` — dispatches on canonical command name resolved via `resolve_command()` from the central registry
+- Skill slash commands: `agent/skill_commands.py` scans `~/.hermes/skills/`, injects as **user message** (not system prompt) to preserve prompt caching
+
+### Slash Command Registry (`hermes_cli/commands.py`)
+
+All slash commands are defined in a central `COMMAND_REGISTRY` list of `CommandDef` objects. Every downstream consumer derives from this registry automatically:
+
+- **CLI** — `process_command()` resolves aliases via `resolve_command()`, dispatches on canonical name
+- **Gateway** — `GATEWAY_KNOWN_COMMANDS` frozenset for hook emission, `resolve_command()` for dispatch
+- **Gateway help** — `gateway_help_lines()` generates `/help` output
+- **Telegram** — `telegram_bot_commands()` generates the BotCommand menu
+- **Slack** — `slack_subcommand_map()` generates `/hermes` subcommand routing
+- **Autocomplete** — `COMMANDS` flat dict feeds `SlashCommandCompleter`
+- **CLI help** — `COMMANDS_BY_CATEGORY` dict feeds `show_help()`
+
+### Adding a Slash Command
+
+1. Add a `CommandDef` entry to `COMMAND_REGISTRY` in `hermes_cli/commands.py`:
+```python
+CommandDef("mycommand", "Description of what it does", "Session",
+           aliases=("mc",), args_hint="[arg]"),
+```
+2. Add handler in `HermesCLI.process_command()` in `cli.py`:
+```python
+elif canonical == "mycommand":
+    self._handle_mycommand(cmd_original)
+```
+3. If the command is available in the gateway, add a handler in `gateway/run.py`:
+```python
+if canonical == "mycommand":
+    return await self._handle_mycommand(event)
+```
+4. For persistent settings, use `save_config_value()` in `cli.py`
+
+**CommandDef fields:**
+- `name` — canonical name without slash (e.g. `"background"`)
+- `description` — human-readable description
+- `category` — one of `"Session"`, `"Configuration"`, `"Tools & Skills"`, `"Info"`, `"Exit"`
+- `aliases` — tuple of alternative names (e.g. `("bg",)`)
+- `args_hint` — argument placeholder shown in help (e.g. `"<prompt>"`, `"[name]"`)
+- `cli_only` — only available in the interactive CLI
+- `gateway_only` — only available in messaging platforms
+- `gateway_config_gate` — config dotpath (e.g. `"display.tool_progress_command"`); when set on a `cli_only` command, the command becomes available in the gateway if the config value is truthy. `GATEWAY_KNOWN_COMMANDS` always includes config-gated commands so the gateway can dispatch them; help/menus only show them when the gate is open.
+
+**Adding an alias** requires only adding it to the `aliases` tuple on the existing `CommandDef`. No other file changes needed — dispatch, help text, Telegram menu, Slack mapping, and autocomplete all update automatically.
+
+---
+
+## Adding New Tools
+
+Requires changes in **3 files**:
+
+**1. Create `tools/your_tool.py`:**
+```python
+import json, os
+from tools.registry import registry
+
+def check_requirements() -> bool:
+    return bool(os.getenv("EXAMPLE_API_KEY"))
+
+def example_tool(param: str, task_id: str = None) -> str:
+    return json.dumps({"success": True, "data": "..."})
+
+registry.register(
+    name="example_tool",
+    toolset="example",
+    schema={"name": "example_tool", "description": "...", "parameters": {...}},
+    handler=lambda args, **kw: example_tool(param=args.get("param", ""), task_id=kw.get("task_id")),
+    check_fn=check_requirements,
+    requires_env=["EXAMPLE_API_KEY"],
+)
+```
+
+**2. Add import** in `model_tools.py` `_discover_tools()` list.
+
+**3. Add to `toolsets.py`** — either `_HERMES_CORE_TOOLS` (all platforms) or a new toolset.
+
+The registry handles schema collection, dispatch, availability checking, and error wrapping. All handlers MUST return a JSON string.
+
+**Path references in tool schemas**: If the schema description mentions file paths (e.g. default output directories), use `display_hermes_home()` to make them profile-aware. The schema is generated at import time, which is after `_apply_profile_override()` sets `HERMES_HOME`.
+
+**State files**: If a tool stores persistent state (caches, logs, checkpoints), use `get_hermes_home()` for the base directory — never `Path.home() / ".hermes"`. This ensures each profile gets its own state.
+
+**Agent-level tools** (todo, memory): intercepted by `run_agent.py` before `handle_function_call()`. See `todo_tool.py` for the pattern.
+
+---
+
+## Adding Configuration
+
+### config.yaml options:
+1. Add to `DEFAULT_CONFIG` in `hermes_cli/config.py`
+2. Bump `_config_version` (currently 5) to trigger migration for existing users
+
+### .env variables:
+1. Add to `OPTIONAL_ENV_VARS` in `hermes_cli/config.py` with metadata:
+```python
+"NEW_API_KEY": {
+    "description": "What it's for",
+    "prompt": "Display name",
+    "url": "https://...",
+    "password": True,
+    "category": "tool",  # provider, tool, messaging, setting
+},
+```
+
+### Config loaders (two separate systems):
+
+| Loader | Used by | Location |
+|--------|---------|----------|
+| `load_cli_config()` | CLI mode | `cli.py` |
+| `load_config()` | `hermes tools`, `hermes setup` | `hermes_cli/config.py` |
+| Direct YAML load | Gateway | `gateway/run.py` |
+
+---
+
+## Skin/Theme System
+
+The skin engine (`hermes_cli/skin_engine.py`) provides data-driven CLI visual customization. Skins are **pure data** — no code changes needed to add a new skin.
+
+### Architecture
+
+```
+hermes_cli/skin_engine.py    # SkinConfig dataclass, built-in skins, YAML loader
+~/.hermes/skins/*.yaml       # User-installed custom skins (drop-in)
+```
+
+- `init_skin_from_config()` — called at CLI startup, reads `display.skin` from config
+- `get_active_skin()` — returns cached `SkinConfig` for the current skin
+- `set_active_skin(name)` — switches skin at runtime (used by `/skin` command)
+- `load_skin(name)` — loads from user skins first, then built-ins, then falls back to default
+- Missing skin values inherit from the `default` skin automatically
+
+### What skins customize
+
+| Element | Skin Key | Used By |
+|---------|----------|---------|
+| Banner panel border | `colors.banner_border` | `banner.py` |
+| Banner panel title | `colors.banner_title` | `banner.py` |
+| Banner section headers | `colors.banner_accent` | `banner.py` |
+| Banner dim text | `colors.banner_dim` | `banner.py` |
+| Banner body text | `colors.banner_text` | `banner.py` |
+| Response box border | `colors.response_border` | `cli.py` |
+| Spinner faces (waiting) | `spinner.waiting_faces` | `display.py` |
+| Spinner faces (thinking) | `spinner.thinking_faces` | `display.py` |
+| Spinner verbs | `spinner.thinking_verbs` | `display.py` |
+| Spinner wings (optional) | `spinner.wings` | `display.py` |
+| Tool output prefix | `tool_prefix` | `display.py` |
+| Per-tool emojis | `tool_emojis` | `display.py` → `get_tool_emoji()` |
+| Agent name | `branding.agent_name` | `banner.py`, `cli.py` |
+| Welcome message | `branding.welcome` | `cli.py` |
+| Response box label | `branding.response_label` | `cli.py` |
+| Prompt symbol | `branding.prompt_symbol` | `cli.py` |
+
+### Built-in skins
+
+- `default` — Classic Hermes gold/kawaii (the current look)
+- `ares` — Crimson/bronze war-god theme with custom spinner wings
+- `mono` — Clean grayscale monochrome
+- `slate` — Cool blue developer-focused theme
+
+### Adding a built-in skin
+
+Add to `_BUILTIN_SKINS` dict in `hermes_cli/skin_engine.py`:
+
+```python
+"mytheme": {
+    "name": "mytheme",
+    "description": "Short description",
+    "colors": { ... },
+    "spinner": { ... },
+    "branding": { ... },
+    "tool_prefix": "┊",
+},
+```
+
+### User skins (YAML)
+
+Users create `~/.hermes/skins/<name>.yaml`:
+
+```yaml
+name: cyberpunk
+description: Neon-soaked terminal theme
+
+colors:
+  banner_border: "#FF00FF"
+  banner_title: "#00FFFF"
+  banner_accent: "#FF1493"
+
+spinner:
+  thinking_verbs: ["jacking in", "decrypting", "uploading"]
+  wings:
+    - ["⟨⚡", "⚡⟩"]
+
+branding:
+  agent_name: "Cyber Agent"
+  response_label: " ⚡ Cyber "
+
+tool_prefix: "▏"
+```
+
+Activate with `/skin cyberpunk` or `display.skin: cyberpunk` in config.yaml.
+
+---
+
+## Important Policies
+### Prompt Caching Must Not Break
+
+Hermes-Agent ensures caching remains valid throughout a conversation. **Do NOT implement changes that would:**
+- Alter past context mid-conversation
+- Change toolsets mid-conversation
+- Reload memories or rebuild system prompts mid-conversation
+
+Cache-breaking forces dramatically higher costs. The ONLY time we alter context is during context compression.
+
+### Working Directory Behavior
+- **CLI**: Uses current directory (`.` → `os.getcwd()`)
+- **Messaging**: Uses `MESSAGING_CWD` env var (default: home directory)
+
+### Background Process Notifications (Gateway)
+
+When `terminal(background=true, check_interval=...)` is used, the gateway runs a watcher that
+pushes status updates to the user's chat. Control verbosity with `display.background_process_notifications`
+in config.yaml (or `HERMES_BACKGROUND_NOTIFICATIONS` env var):
+
+- `all` — running-output updates + final message (default)
+- `result` — only the final completion message
+- `error` — only the final message when exit code != 0
+- `off` — no watcher messages at all
+
+---
+
+## Profiles: Multi-Instance Support
+
+Hermes supports **profiles** — multiple fully isolated instances, each with its own
+`HERMES_HOME` directory (config, API keys, memory, sessions, skills, gateway, etc.).
+
+The core mechanism: `_apply_profile_override()` in `hermes_cli/main.py` sets
+`HERMES_HOME` before any module imports. All 119+ references to `get_hermes_home()`
+automatically scope to the active profile.
+
+### Rules for profile-safe code
+
+1. **Use `get_hermes_home()` for all HERMES_HOME paths.** Import from `hermes_constants`.
+   NEVER hardcode `~/.hermes` or `Path.home() / ".hermes"` in code that reads/writes state.
+   ```python
+   # GOOD
+   from hermes_constants import get_hermes_home
+   config_path = get_hermes_home() / "config.yaml"
+
+   # BAD — breaks profiles
+   config_path = Path.home() / ".hermes" / "config.yaml"
+   ```
+
+2. **Use `display_hermes_home()` for user-facing messages.** Import from `hermes_constants`.
+   This returns `~/.hermes` for default or `~/.hermes/profiles/<name>` for profiles.
+   ```python
+   # GOOD
+   from hermes_constants import display_hermes_home
+   print(f"Config saved to {display_hermes_home()}/config.yaml")
+
+   # BAD — shows wrong path for profiles
+   print("Config saved to ~/.hermes/config.yaml")
+   ```
+
+3. **Module-level constants are fine** — they cache `get_hermes_home()` at import time,
+   which is AFTER `_apply_profile_override()` sets the env var. Just use `get_hermes_home()`,
+   not `Path.home() / ".hermes"`.
+
+4. **Tests that mock `Path.home()` must also set `HERMES_HOME`** — since code now uses
+   `get_hermes_home()` (reads env var), not `Path.home() / ".hermes"`:
+   ```python
+   with patch.object(Path, "home", return_value=tmp_path), \
+        patch.dict(os.environ, {"HERMES_HOME": str(tmp_path / ".hermes")}):
+       ...
+   ```
+
+5. **Gateway platform adapters should use token locks** — if the adapter connects with
+   a unique credential (bot token, API key), call `acquire_scoped_lock()` from
+   `gateway.status` in the `connect()`/`start()` method and `release_scoped_lock()` in
+   `disconnect()`/`stop()`. This prevents two profiles from using the same credential.
+   See `gateway/platforms/telegram.py` for the canonical pattern.
+
+6. **Profile operations are HOME-anchored, not HERMES_HOME-anchored** — `_get_profiles_root()`
+   returns `Path.home() / ".hermes" / "profiles"`, NOT `get_hermes_home() / "profiles"`.
+   This is intentional — it lets `hermes -p coder profile list` see all profiles regardless
+   of which one is active.
+
+## Known Pitfalls
+
+### DO NOT hardcode `~/.hermes` paths
+Use `get_hermes_home()` from `hermes_constants` for code paths. Use `display_hermes_home()`
+for user-facing print/log messages. Hardcoding `~/.hermes` breaks profiles — each profile
+has its own `HERMES_HOME` directory. This was the source of 5 bugs fixed in PR #3575.
+
+### DO NOT use `simple_term_menu` for interactive menus
+Rendering bugs in tmux/iTerm2 — ghosting on scroll. Use `curses` (stdlib) instead. See `hermes_cli/tools_config.py` for the pattern.
+
+### DO NOT use `\033[K` (ANSI erase-to-EOL) in spinner/display code
+Leaks as literal `?[K` text under `prompt_toolkit`'s `patch_stdout`. Use space-padding: `f"\r{line}{' ' * pad}"`.
+
+### `_last_resolved_tool_names` is a process-global in `model_tools.py`
+`_run_single_child()` in `delegate_tool.py` saves and restores this global around subagent execution. If you add new code that reads this global, be aware it may be temporarily stale during child agent runs.
+
+### DO NOT hardcode cross-tool references in schema descriptions
+Tool schema descriptions must not mention tools from other toolsets by name (e.g., `browser_navigate` saying "prefer web_search"). Those tools may be unavailable (missing API keys, disabled toolset), causing the model to hallucinate calls to non-existent tools. If a cross-reference is needed, add it dynamically in `get_tool_definitions()` in `model_tools.py` — see the `browser_navigate` / `execute_code` post-processing blocks for the pattern.
+
+### Tests must not write to `~/.hermes/`
+The `_isolate_hermes_home` autouse fixture in `tests/conftest.py` redirects `HERMES_HOME` to a temp dir. Never hardcode `~/.hermes/` paths in tests.
+
+**Profile tests**: When testing profile features, also mock `Path.home()` so that
+`_get_profiles_root()` and `_get_default_hermes_home()` resolve within the temp dir.
+Use the pattern from `tests/hermes_cli/test_profiles.py`:
+```python
+@pytest.fixture
+def profile_env(tmp_path, monkeypatch):
+    home = tmp_path / ".hermes"
+    home.mkdir()
+    monkeypatch.setattr(Path, "home", lambda: tmp_path)
+    monkeypatch.setenv("HERMES_HOME", str(home))
+    return home
+```
+
+---
+
+## Testing
+
+```bash
+source venv/bin/activate
+python -m pytest tests/ -q          # Full suite (~3000 tests, ~3 min)
+python -m pytest tests/test_model_tools.py -q   # Toolset resolution
+python -m pytest tests/test_cli_init.py -q       # CLI config loading
+python -m pytest tests/gateway/ -q               # Gateway tests
+python -m pytest tests/tools/ -q                 # Tool-level tests
+```
+
+Always run the full suite before pushing changes.

CONTRIBUTING.md

@@ -0,0 +1,660 @@
+# Contributing to Hermes Agent
+
+Thank you for contributing to Hermes Agent! This guide covers everything you need: setting up your dev environment, understanding the architecture, deciding what to build, and getting your PR merged.
+
+---
+
+## Contribution Priorities
+
+We value contributions in this order:
+
+1. **Bug fixes** — crashes, incorrect behavior, data loss. Always top priority.
+2. **Cross-platform compatibility** — Windows, macOS, different Linux distros, different terminal emulators. We want Hermes to work everywhere.
+3. **Security hardening** — shell injection, prompt injection, path traversal, privilege escalation. See [Security](#security-considerations).
+4. **Performance and robustness** — retry logic, error handling, graceful degradation.
+5. **New skills** — but only broadly useful ones. See [Should it be a Skill or a Tool?](#should-it-be-a-skill-or-a-tool)
+6. **New tools** — rarely needed. Most capabilities should be skills. See below.
+7. **Documentation** — fixes, clarifications, new examples.
+
+---
+
+## Should it be a Skill or a Tool?
+
+This is the most common question for new contributors. The answer is almost always **skill**.
+
+### Make it a Skill when:
+
+- The capability can be expressed as instructions + shell commands + existing tools
+- It wraps an external CLI or API that the agent can call via `terminal` or `web_extract`
+- It doesn't need custom Python integration or API key management baked into the agent
+- Examples: arXiv search, git workflows, Docker management, PDF processing, email via CLI tools
+
+### Make it a Tool when:
+
+- It requires end-to-end integration with API keys, auth flows, or multi-component configuration managed by the agent harness
+- It needs custom processing logic that must execute precisely every time (not "best effort" from LLM interpretation)
+- It handles binary data, streaming, or real-time events that can't go through the terminal
+- Examples: browser automation (Browserbase session management), TTS (audio encoding + platform delivery), vision analysis (base64 image handling)
+
+### Should the Skill be bundled?
+
+Bundled skills (in `skills/`) ship with every Hermes install. They should be **broadly useful to most users**:
+
+- Document handling, web research, common dev workflows, system administration
+- Used regularly by a wide range of people
+
+If your skill is official and useful but not universally needed (e.g., a paid service integration, a heavyweight dependency), put it in **`optional-skills/`** — it ships with the repo but isn't activated by default. Users can discover it via `hermes skills browse` (labeled "official") and install it with `hermes skills install` (no third-party warning, builtin trust).
+
+If your skill is specialized, community-contributed, or niche, it's better suited for a **Skills Hub** — upload it to a skills registry and share it in the [Nous Research Discord](https://discord.gg/NousResearch). Users can install it with `hermes skills install`.
+
+---
+
+## Development Setup
+
+### Prerequisites
+
+| Requirement | Notes |
+|-------------|-------|
+| **Git** | With `--recurse-submodules` support |
+| **Python 3.11+** | uv will install it if missing |
+| **uv** | Fast Python package manager ([install](https://docs.astral.sh/uv/)) |
+| **Node.js 18+** | Optional — needed for browser tools and WhatsApp bridge |
+
+### Clone and install
+
+```bash
+git clone --recurse-submodules https://github.com/NousResearch/hermes-agent.git
+cd hermes-agent
+
+# Create venv with Python 3.11
+uv venv venv --python 3.11
+export VIRTUAL_ENV="$(pwd)/venv"
+
+# Install with all extras (messaging, cron, CLI menus, dev tools)
+uv pip install -e ".[all,dev]"
+
+# Optional: RL training submodule
+# git submodule update --init tinker-atropos && uv pip install -e "./tinker-atropos"
+
+# Optional: browser tools
+npm install
+```
+
+### Configure for development
+
+```bash
+mkdir -p ~/.hermes/{cron,sessions,logs,memories,skills}
+cp cli-config.yaml.example ~/.hermes/config.yaml
+touch ~/.hermes/.env
+
+# Add at minimum an LLM provider key:
+echo 'OPENROUTER_API_KEY=sk-or-v1-your-key' >> ~/.hermes/.env
+```
+
+### Run
+
+```bash
+# Symlink for global access
+mkdir -p ~/.local/bin
+ln -sf "$(pwd)/venv/bin/hermes" ~/.local/bin/hermes
+
+# Verify
+hermes doctor
+hermes chat -q "Hello"
+```
+
+### Run tests
+
+```bash
+pytest tests/ -v
+```
+
+---
+
+## Project Structure
+
+```
+hermes-agent/
+├── run_agent.py              # AIAgent class — core conversation loop, tool dispatch, session persistence
+├── cli.py                    # HermesCLI class — interactive TUI, prompt_toolkit integration
+├── model_tools.py            # Tool orchestration (thin layer over tools/registry.py)
+├── toolsets.py               # Tool groupings and presets (hermes-cli, hermes-telegram, etc.)
+├── hermes_state.py           # SQLite session database with FTS5 full-text search, session titles
+├── batch_runner.py           # Parallel batch processing for trajectory generation
+│
+├── agent/                    # Agent internals (extracted modules)
+│   ├── prompt_builder.py         # System prompt assembly (identity, skills, context files, memory)
+│   ├── context_compressor.py     # Auto-summarization when approaching context limits
+│   ├── auxiliary_client.py       # Resolves auxiliary OpenAI clients (summarization, vision)
+│   ├── display.py                # KawaiiSpinner, tool progress formatting
+│   ├── model_metadata.py         # Model context lengths, token estimation
+│   └── trajectory.py             # Trajectory saving helpers
+│
+├── hermes_cli/               # CLI command implementations
+│   ├── main.py                   # Entry point, argument parsing, command dispatch
+│   ├── config.py                 # Config management, migration, env var definitions
+│   ├── setup.py                  # Interactive setup wizard
+│   ├── auth.py                   # Provider resolution, OAuth, Nous Portal
+│   ├── models.py                 # OpenRouter model selection lists
+│   ├── banner.py                 # Welcome banner, ASCII art
+│   ├── commands.py               # Central slash command registry (CommandDef), autocomplete, gateway helpers
+│   ├── callbacks.py              # Interactive callbacks (clarify, sudo, approval)
+│   ├── doctor.py                 # Diagnostics
+│   ├── skills_hub.py             # Skills Hub CLI + /skills slash command
+│   └── skin_engine.py            # Skin/theme engine — data-driven CLI visual customization
+│
+├── tools/                    # Tool implementations (self-registering)
+│   ├── registry.py               # Central tool registry (schemas, handlers, dispatch)
+│   ├── approval.py               # Dangerous command detection + per-session approval
+│   ├── terminal_tool.py          # Terminal orchestration (sudo, env lifecycle, backends)
+│   ├── file_operations.py        # read_file, write_file, search, patch, etc.
+│   ├── web_tools.py              # web_search, web_extract (Parallel/Firecrawl + Gemini summarization)
+│   ├── vision_tools.py           # Image analysis via multimodal models
+│   ├── delegate_tool.py          # Subagent spawning and parallel task execution
+│   ├── code_execution_tool.py    # Sandboxed Python with RPC tool access
+│   ├── session_search_tool.py    # Search past conversations with FTS5 + summarization
+│   ├── cronjob_tools.py          # Scheduled task management
+│   ├── skill_tools.py            # Skill search, load, manage
+│   └── environments/             # Terminal execution backends
+│       ├── base.py                   # BaseEnvironment ABC
+│       ├── local.py, docker.py, ssh.py, singularity.py, modal.py, daytona.py
+│
+├── gateway/                  # Messaging gateway
+│   ├── run.py                    # GatewayRunner — platform lifecycle, message routing, cron
+│   ├── config.py                 # Platform configuration resolution
+│   ├── session.py                # Session store, context prompts, reset policies
+│   └── platforms/                # Platform adapters
+│       ├── telegram.py, discord_adapter.py, slack.py, whatsapp.py
+│
+├── scripts/                  # Installer and bridge scripts
+│   ├── install.sh                # Linux/macOS installer
+│   ├── install.ps1               # Windows PowerShell installer
+│   └── whatsapp-bridge/          # Node.js WhatsApp bridge (Baileys)
+│
+├── skills/                   # Bundled skills (copied to ~/.hermes/skills/ on install)
+├── optional-skills/          # Official optional skills (discoverable via hub, not activated by default)
+├── environments/             # RL training environments (Atropos integration)
+├── tests/                    # Test suite
+├── website/                  # Documentation site (hermes-agent.nousresearch.com)
+│
+├── cli-config.yaml.example   # Example configuration (copied to ~/.hermes/config.yaml)
+└── AGENTS.md                 # Development guide for AI coding assistants
+```
+
+### User configuration (stored in `~/.hermes/`)
+
+| Path | Purpose |
+|------|---------|
+| `~/.hermes/config.yaml` | Settings (model, terminal, toolsets, compression, etc.) |
+| `~/.hermes/.env` | API keys and secrets |
+| `~/.hermes/auth.json` | OAuth credentials (Nous Portal) |
+| `~/.hermes/skills/` | All active skills (bundled + hub-installed + agent-created) |
+| `~/.hermes/memories/` | Persistent memory (MEMORY.md, USER.md) |
+| `~/.hermes/state.db` | SQLite session database |
+| `~/.hermes/sessions/` | JSON session logs |
+| `~/.hermes/cron/` | Scheduled job data |
+| `~/.hermes/whatsapp/session/` | WhatsApp bridge credentials |
+
+---
+
+## Architecture Overview
+
+### Core Loop
+
+```
+User message → AIAgent._run_agent_loop()
+  ├── Build system prompt (prompt_builder.py)
+  ├── Build API kwargs (model, messages, tools, reasoning config)
+  ├── Call LLM (OpenAI-compatible API)
+  ├── If tool_calls in response:
+  │     ├── Execute each tool via registry dispatch
+  │     ├── Add tool results to conversation
+  │     └── Loop back to LLM call
+  ├── If text response:
+  │     ├── Persist session to DB
+  │     └── Return final_response
+  └── Context compression if approaching token limit
+```
+
+### Key Design Patterns
+
+- **Self-registering tools**: Each tool file calls `registry.register()` at import time. `model_tools.py` triggers discovery by importing all tool modules.
+- **Toolset grouping**: Tools are grouped into toolsets (`web`, `terminal`, `file`, `browser`, etc.) that can be enabled/disabled per platform.
+- **Session persistence**: All conversations are stored in SQLite (`hermes_state.py`) with full-text search and unique session titles. JSON logs go to `~/.hermes/sessions/`.
+- **Ephemeral injection**: System prompts and prefill messages are injected at API call time, never persisted to the database or logs.
+- **Provider abstraction**: The agent works with any OpenAI-compatible API. Provider resolution happens at init time (Nous Portal OAuth, OpenRouter API key, or custom endpoint).
+- **Provider routing**: When using OpenRouter, `provider_routing` in config.yaml controls provider selection (sort by throughput/latency/price, allow/ignore specific providers, data retention policies). These are injected as `extra_body.provider` in API requests.
+
+---
+
+## Code Style
+
+- **PEP 8** with practical exceptions (we don't enforce strict line length)
+- **Comments**: Only when explaining non-obvious intent, trade-offs, or API quirks. Don't narrate what the code does — `# increment counter` adds nothing
+- **Error handling**: Catch specific exceptions. Log with `logger.warning()`/`logger.error()` — use `exc_info=True` for unexpected errors so stack traces appear in logs
+- **Cross-platform**: Never assume Unix. See [Cross-Platform Compatibility](#cross-platform-compatibility)
+
+---
+
+## Adding a New Tool
+
+Before writing a tool, ask: [should this be a skill instead?](#should-it-be-a-skill-or-a-tool)
+
+Tools self-register with the central registry. Each tool file co-locates its schema, handler, and registration:
+
+```python
+"""my_tool — Brief description of what this tool does."""
+
+import json
+from tools.registry import registry
+
+
+def my_tool(param1: str, param2: int = 10, **kwargs) -> str:
+    """Handler. Returns a string result (often JSON)."""
+    result = do_work(param1, param2)
+    return json.dumps(result)
+
+
+MY_TOOL_SCHEMA = {
+    "type": "function",
+    "function": {
+        "name": "my_tool",
+        "description": "What this tool does and when the agent should use it.",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "param1": {"type": "string", "description": "What param1 is"},
+                "param2": {"type": "integer", "description": "What param2 is", "default": 10},
+            },
+            "required": ["param1"],
+        },
+    },
+}
+
+
+def _check_requirements() -> bool:
+    """Return True if this tool's dependencies are available."""
+    return True
+
+
+registry.register(
+    name="my_tool",
+    toolset="my_toolset",
+    schema=MY_TOOL_SCHEMA,
+    handler=lambda args, **kw: my_tool(**args, **kw),
+    check_fn=_check_requirements,
+)
+```
+
+Then add the import to `model_tools.py` in the `_modules` list:
+
+```python
+_modules = [
+    # ... existing modules ...
+    "tools.my_tool",
+]
+```
+
+If it's a new toolset, add it to `toolsets.py` and to the relevant platform presets.
+
+---
+
+## Adding a Skill
+
+Bundled skills live in `skills/` organized by category. Official optional skills use the same structure in `optional-skills/`:
+
+```
+skills/
+├── research/
+│   └── arxiv/
+│       ├── SKILL.md              # Required: main instructions
+│       └── scripts/              # Optional: helper scripts
+│           └── search_arxiv.py
+├── productivity/
+│   └── ocr-and-documents/
+│       ├── SKILL.md
+│       ├── scripts/
+│       └── references/
+└── ...
+```
+
+### SKILL.md format
+
+```markdown
+---
+name: my-skill
+description: Brief description (shown in skill search results)
+version: 1.0.0
+author: Your Name
+license: MIT
+platforms: [macos, linux]          # Optional — restrict to specific OS platforms
+                                   #   Valid: macos, linux, windows
+                                   #   Omit to load on all platforms (default)
+required_environment_variables:    # Optional — secure setup-on-load metadata
+  - name: MY_API_KEY
+    prompt: API key
+    help: Where to get it
+    required_for: full functionality
+prerequisites:                     # Optional legacy runtime requirements
+  env_vars: [MY_API_KEY]           #   Backward-compatible alias for required env vars
+  commands: [curl, jq]             #   Advisory only; does not hide the skill
+metadata:
+  hermes:
+    tags: [Category, Subcategory, Keywords]
+    related_skills: [other-skill-name]
+    fallback_for_toolsets: [web]       # Optional — show only when toolset is unavailable
+    requires_toolsets: [terminal]      # Optional — show only when toolset is available
+---
+
+# Skill Title
+
+Brief intro.
+
+## When to Use
+Trigger conditions — when should the agent load this skill?
+
+## Quick Reference
+Table of common commands or API calls.
+
+## Procedure
+Step-by-step instructions the agent follows.
+
+## Pitfalls
+Known failure modes and how to handle them.
+
+## Verification
+How the agent confirms it worked.
+```
+
+### Platform-specific skills
+
+Skills can declare which OS platforms they support via the `platforms` frontmatter field. Skills with this field are automatically hidden from the system prompt, `skills_list()`, and slash commands on incompatible platforms.
+
+```yaml
+platforms: [macos]            # macOS only (e.g., iMessage, Apple Reminders)
+platforms: [macos, linux]     # macOS and Linux
+platforms: [windows]          # Windows only
+```
+
+If the field is omitted or empty, the skill loads on all platforms (backward compatible). See `skills/apple/` for examples of macOS-only skills.
+
+### Conditional skill activation
+
+Skills can declare conditions that control when they appear in the system prompt, based on which tools and toolsets are available in the current session. This is primarily used for **fallback skills** — alternatives that should only be shown when a primary tool is unavailable.
+
+Four fields are supported under `metadata.hermes`:
+
+```yaml
+metadata:
+  hermes:
+    fallback_for_toolsets: [web]      # Show ONLY when these toolsets are unavailable
+    requires_toolsets: [terminal]     # Show ONLY when these toolsets are available
+    fallback_for_tools: [web_search]  # Show ONLY when these specific tools are unavailable
+    requires_tools: [terminal]        # Show ONLY when these specific tools are available
+```
+
+**Semantics:**
+- `fallback_for_*`: The skill is a backup. It is **hidden** when the listed tools/toolsets are available, and **shown** when they are unavailable. Use this for free alternatives to premium tools.
+- `requires_*`: The skill needs certain tools to function. It is **hidden** when the listed tools/toolsets are unavailable. Use this for skills that depend on specific capabilities (e.g., a skill that only makes sense with terminal access).
+- If both are specified, both conditions must be satisfied for the skill to appear.
+- If neither is specified, the skill is always shown (backward compatible).
+
+**Examples:**
+
+```yaml
+# DuckDuckGo search — shown when Firecrawl (web toolset) is unavailable
+metadata:
+  hermes:
+    fallback_for_toolsets: [web]
+
+# Smart home skill — only useful when terminal is available
+metadata:
+  hermes:
+    requires_toolsets: [terminal]
+
+# Local browser fallback — shown when Browserbase is unavailable
+metadata:
+  hermes:
+    fallback_for_toolsets: [browser]
+```
+
+The filtering happens at prompt build time in `agent/prompt_builder.py`. The `build_skills_system_prompt()` function receives the set of available tools and toolsets from the agent and uses `_skill_should_show()` to evaluate each skill's conditions.
+
+### Skill setup metadata
+
+Skills can declare secure setup-on-load metadata via the `required_environment_variables` frontmatter field. Missing values do not hide the skill from discovery; they trigger a CLI-only secure prompt when the skill is actually loaded.
+
+```yaml
+required_environment_variables:
+  - name: TENOR_API_KEY
+    prompt: Tenor API key
+    help: Get a key from https://developers.google.com/tenor
+    required_for: full functionality
+```
+
+The user may skip setup and keep loading the skill. Hermes only exposes metadata (`stored_as`, `skipped`, `validated`) to the model — never the secret value.
+
+Legacy `prerequisites.env_vars` remains supported and is normalized into the new representation.
+
+```yaml
+prerequisites:
+  env_vars: [TENOR_API_KEY]       # Legacy alias for required_environment_variables
+  commands: [curl, jq]            # Advisory CLI checks
+```
+
+Gateway and messaging sessions never collect secrets in-band; they instruct the user to run `hermes setup` or update `~/.hermes/.env` locally.
+
+**When to declare required environment variables:**
+- The skill uses an API key or token that should be collected securely at load time
+- The skill can still be useful if the user skips setup, but may degrade gracefully
+
+**When to declare command prerequisites:**
+- The skill relies on a CLI tool that may not be installed (e.g., `himalaya`, `openhue`, `ddgs`)
+- Treat command checks as guidance, not discovery-time hiding
+
+See `skills/gifs/gif-search/` and `skills/email/himalaya/` for examples.
+
+### Skill guidelines
+
+- **No external dependencies unless absolutely necessary.** Prefer stdlib Python, curl, and existing Hermes tools (`web_extract`, `terminal`, `read_file`).
+- **Progressive disclosure.** Put the most common workflow first. Edge cases and advanced usage go at the bottom.
+- **Include helper scripts** for XML/JSON parsing or complex logic — don't expect the LLM to write parsers inline every time.
+- **Test it.** Run `hermes --toolsets skills -q "Use the X skill to do Y"` and verify the agent follows the instructions correctly.
+
+---
+
+## Adding a Skin / Theme
+
+Hermes uses a data-driven skin system — no code changes needed to add a new skin.
+
+**Option A: User skin (YAML file)**
+
+Create `~/.hermes/skins/<name>.yaml`:
+
+```yaml
+name: mytheme
+description: Short description of the theme
+
+colors:
+  banner_border: "#HEX"     # Panel border color
+  banner_title: "#HEX"      # Panel title color
+  banner_accent: "#HEX"     # Section header color
+  banner_dim: "#HEX"        # Muted/dim text color
+  banner_text: "#HEX"       # Body text color
+  response_border: "#HEX"   # Response box border
+
+spinner:
+  waiting_faces: ["(⚔)", "(⛨)"]
+  thinking_faces: ["(⚔)", "(⌁)"]
+  thinking_verbs: ["forging", "plotting"]
+  wings:                     # Optional left/right decorations
+    - ["⟪⚔", "⚔⟫"]
+
+branding:
+  agent_name: "My Agent"
+  welcome: "Welcome message"
+  response_label: " ⚔ Agent "
+  prompt_symbol: "⚔ ❯ "
+
+tool_prefix: "╎"             # Tool output line prefix
+```
+
+All fields are optional — missing values inherit from the default skin.
+
+**Option B: Built-in skin**
+
+Add to `_BUILTIN_SKINS` dict in `hermes_cli/skin_engine.py`. Use the same schema as above but as a Python dict. Built-in skins ship with the package and are always available.
+
+**Activating:**
+- CLI: `/skin mytheme` or set `display.skin: mytheme` in config.yaml
+- Config: `display: { skin: mytheme }`
+
+See `hermes_cli/skin_engine.py` for the full schema and existing skins as examples.
+
+---
+
+## Cross-Platform Compatibility
+
+Hermes runs on Linux, macOS, and Windows. When writing code that touches the OS:
+
+### Critical rules
+
+1. **`termios` and `fcntl` are Unix-only.** Always catch both `ImportError` and `NotImplementedError`:
+   ```python
+   try:
+       from simple_term_menu import TerminalMenu
+       menu = TerminalMenu(options)
+       idx = menu.show()
+   except (ImportError, NotImplementedError):
+       # Fallback: numbered menu for Windows
+       for i, opt in enumerate(options):
+           print(f"  {i+1}. {opt}")
+       idx = int(input("Choice: ")) - 1
+   ```
+
+2. **File encoding.** Windows may save `.env` files in `cp1252`. Always handle encoding errors:
+   ```python
+   try:
+       load_dotenv(env_path)
+   except UnicodeDecodeError:
+       load_dotenv(env_path, encoding="latin-1")
+   ```
+
+3. **Process management.** `os.setsid()`, `os.killpg()`, and signal handling differ on Windows. Use platform checks:
+   ```python
+   import platform
+   if platform.system() != "Windows":
+       kwargs["preexec_fn"] = os.setsid
+   ```
+
+4. **Path separators.** Use `pathlib.Path` instead of string concatenation with `/`.
+
+5. **Shell commands in installers.** If you change `scripts/install.sh`, check if the equivalent change is needed in `scripts/install.ps1`.
+
+---
+
+## Security Considerations
+
+Hermes has terminal access. Security matters.
+
+### Existing protections
+
+| Layer | Implementation |
+|-------|---------------|
+| **Sudo password piping** | Uses `shlex.quote()` to prevent shell injection |
+| **Dangerous command detection** | Regex patterns in `tools/approval.py` with user approval flow |
+| **Cron prompt injection** | Scanner in `tools/cronjob_tools.py` blocks instruction-override patterns |
+| **Write deny list** | Protected paths (`~/.ssh/authorized_keys`, `/etc/shadow`) resolved via `os.path.realpath()` to prevent symlink bypass |
+| **Skills guard** | Security scanner for hub-installed skills (`tools/skills_guard.py`) |
+| **Code execution sandbox** | `execute_code` child process runs with API keys stripped from environment |
+| **Container hardening** | Docker: all capabilities dropped, no privilege escalation, PID limits, size-limited tmpfs |
+
+### When contributing security-sensitive code
+
+- **Always use `shlex.quote()`** when interpolating user input into shell commands
+- **Resolve symlinks** with `os.path.realpath()` before path-based access control checks
+- **Don't log secrets.** API keys, tokens, and passwords should never appear in log output
+- **Catch broad exceptions** around tool execution so a single failure doesn't crash the agent loop
+- **Test on all platforms** if your change touches file paths, process management, or shell commands
+
+If your PR affects security, note it explicitly in the description.
+
+---
+
+## Pull Request Process
+
+### Branch naming
+
+```
+fix/description        # Bug fixes
+feat/description       # New features
+docs/description       # Documentation
+test/description       # Tests
+refactor/description   # Code restructuring
+```
+
+### Before submitting
+
+1. **Run tests**: `pytest tests/ -v`
+2. **Test manually**: Run `hermes` and exercise the code path you changed
+3. **Check cross-platform impact**: If you touch file I/O, process management, or terminal handling, consider Windows and macOS
+4. **Keep PRs focused**: One logical change per PR. Don't mix a bug fix with a refactor with a new feature.
+
+### PR description
+
+Include:
+- **What** changed and **why**
+- **How to test** it (reproduction steps for bugs, usage examples for features)
+- **What platforms** you tested on
+- Reference any related issues
+
+### Commit messages
+
+We use [Conventional Commits](https://www.conventionalcommits.org/):
+
+```
+<type>(<scope>): <description>
+```
+
+| Type | Use for |
+|------|---------|
+| `fix` | Bug fixes |
+| `feat` | New features |
+| `docs` | Documentation |
+| `test` | Tests |
+| `refactor` | Code restructuring (no behavior change) |
+| `chore` | Build, CI, dependency updates |
+
+Scopes: `cli`, `gateway`, `tools`, `skills`, `agent`, `install`, `whatsapp`, `security`, etc.
+
+Examples:
+```
+fix(cli): prevent crash in save_config_value when model is a string
+feat(gateway): add WhatsApp multi-user session isolation
+fix(security): prevent shell injection in sudo password piping
+test(tools): add unit tests for file_operations
+```
+
+---
+
+## Reporting Issues
+
+- Use [GitHub Issues](https://github.com/NousResearch/hermes-agent/issues)
+- Include: OS, Python version, Hermes version (`hermes version`), full error traceback
+- Include steps to reproduce
+- Check existing issues before creating duplicates
+- For security vulnerabilities, please report privately
+
+---
+
+## Community
+
+- **Discord**: [discord.gg/NousResearch](https://discord.gg/NousResearch) — for questions, showcasing projects, and sharing skills
+- **GitHub Discussions**: For design proposals and architecture discussions
+- **Skills Hub**: Upload specialized skills to a registry and share them with the community
+
+---
+
+## License
+
+By contributing, you agree that your contributions will be licensed under the [MIT License](LICENSE).

Dockerfile

@@ -0,0 +1,25 @@
+FROM debian:13.4
+
+# Install system dependencies in one layer, clear APT cache
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends \
+        build-essential nodejs npm python3 python3-pip ripgrep ffmpeg gcc python3-dev libffi-dev && \
+    rm -rf /var/lib/apt/lists/*
+
+COPY . /opt/hermes
+WORKDIR /opt/hermes
+
+# Install Python and Node dependencies in one layer, no cache
+RUN pip install --no-cache-dir -e ".[all]" --break-system-packages && \
+    npm install --prefer-offline --no-audit && \
+    npx playwright install --with-deps chromium --only-shell && \
+    cd /opt/hermes/scripts/whatsapp-bridge && \
+    npm install --prefer-offline --no-audit && \
+    npm cache clean --force
+
+WORKDIR /opt/hermes
+RUN chmod +x /opt/hermes/docker/entrypoint.sh
+
+ENV HERMES_HOME=/opt/data
+VOLUME [ "/opt/data" ]
+ENTRYPOINT [ "/opt/hermes/docker/entrypoint.sh" ]

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 Nous Research
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

MANIFEST.in

@@ -0,0 +1,4 @@
+graft skills
+graft optional-skills
+global-exclude __pycache__
+global-exclude *.py[cod]

README.md

@@ -1,243 +1,175 @@
-# Hermes Agent
+<p align="center">
+  <img src="assets/banner.png" alt="Hermes Agent" width="100%">
+</p>
 
-An AI agent with advanced tool-calling capabilities, featuring a flexible toolsets system for organizing and managing tools.
+# Hermes Agent ☤
 
-## Features
+<p align="center">
+  <a href="https://hermes-agent.nousresearch.com/docs/"><img src="https://img.shields.io/badge/Docs-hermes--agent.nousresearch.com-FFD700?style=for-the-badge" alt="Documentation"></a>
+  <a href="https://discord.gg/NousResearch"><img src="https://img.shields.io/badge/Discord-5865F2?style=for-the-badge&logo=discord&logoColor=white" alt="Discord"></a>
+  <a href="https://github.com/NousResearch/hermes-agent/blob/main/LICENSE"><img src="https://img.shields.io/badge/License-MIT-green?style=for-the-badge" alt="License: MIT"></a>
+  <a href="https://nousresearch.com"><img src="https://img.shields.io/badge/Built%20by-Nous%20Research-blueviolet?style=for-the-badge" alt="Built by Nous Research"></a>
+</p>
 
-- **Web Tools**: Search, extract content, and crawl websites
-- **Terminal Tools**: Execute commands with interactive session support
-- **Vision Tools**: Analyze images from URLs
-- **Reasoning Tools**: Advanced multi-model reasoning (Mixture of Agents)
-- **Creative Tools**: Generate images from text prompts
-- **Toolsets System**: Organize tools into logical groups for different scenarios
-- **Batch Processing**: Process datasets in parallel with checkpointing and statistics tracking
-- **Ephemeral System Prompts**: Guide model behavior without polluting training datasets
+**The self-improving AI agent built by [Nous Research](https://nousresearch.com).** It's the only agent with a built-in learning loop — it creates skills from experience, improves them during use, nudges itself to persist knowledge, searches its own past conversations, and builds a deepening model of who you are across sessions. Run it on a $5 VPS, a GPU cluster, or serverless infrastructure that costs nearly nothing when idle. It's not tied to your laptop — talk to it from Telegram while it works on a cloud VM.
 
-## Setup
+Use any model you want — [Nous Portal](https://portal.nousresearch.com), [OpenRouter](https://openrouter.ai) (200+ models), [z.ai/GLM](https://z.ai), [Kimi/Moonshot](https://platform.moonshot.ai), [MiniMax](https://www.minimax.io), OpenAI, or your own endpoint. Switch with `hermes model` — no code changes, no lock-in.
 
-### 1. Install Dependencies
-```bash
-# Create and activate virtual environment (recommended)
-python3 -m venv venv
-source venv/bin/activate  # On Windows: venv\Scripts\activate
+<table>
+<tr><td><b>A real terminal interface</b></td><td>Full TUI with multiline editing, slash-command autocomplete, conversation history, interrupt-and-redirect, and streaming tool output.</td></tr>
+<tr><td><b>Lives where you do</b></td><td>Telegram, Discord, Slack, WhatsApp, Signal, and CLI — all from a single gateway process. Voice memo transcription, cross-platform conversation continuity.</td></tr>
+<tr><td><b>A closed learning loop</b></td><td>Agent-curated memory with periodic nudges. Autonomous skill creation after complex tasks. Skills self-improve during use. FTS5 session search with LLM summarization for cross-session recall. <a href="https://github.com/plastic-labs/honcho">Honcho</a> dialectic user modeling. Compatible with the <a href="https://agentskills.io">agentskills.io</a> open standard.</td></tr>
+<tr><td><b>Scheduled automations</b></td><td>Built-in cron scheduler with delivery to any platform. Daily reports, nightly backups, weekly audits — all in natural language, running unattended.</td></tr>
+<tr><td><b>Delegates and parallelizes</b></td><td>Spawn isolated subagents for parallel workstreams. Write Python scripts that call tools via RPC, collapsing multi-step pipelines into zero-context-cost turns.</td></tr>
+<tr><td><b>Runs anywhere, not just your laptop</b></td><td>Six terminal backends — local, Docker, SSH, Daytona, Singularity, and Modal. Daytona and Modal offer serverless persistence — your agent's environment hibernates when idle and wakes on demand, costing nearly nothing between sessions. Run it on a $5 VPS or a GPU cluster.</td></tr>
+<tr><td><b>Research-ready</b></td><td>Batch trajectory generation, Atropos RL environments, trajectory compression for training the next generation of tool-calling models.</td></tr>
+</table>
 
-# Install required packages
-pip install -r requirements.txt
+---
 
-# Install Hecate for terminal tools
-git clone git@github.com:NousResearch/hecate.git
-cd hecate
-pip install -e .
-cd ..
-```
-
-### 2. Configure Environment Variables
-```bash
-# Copy the example environment file
-cp .env.example .env
-
-# Edit .env and add your API keys
-nano .env  # or use your preferred editor
-```
-
-**Required API Keys:**
-- `ANTHROPIC_API_KEY` - Main agent model (get at: https://console.anthropic.com/)
-- `FIRECRAWL_API_KEY` - Web tools (get at: https://firecrawl.dev/)
-- `NOUS_API_KEY` - Vision & reasoning tools (get at: https://inference-api.nousresearch.com/)
-- `MORPH_API_KEY` - Terminal tools (get at: https://morph.so/)
-- `FAL_KEY` - Image generation (get at: https://fal.ai/)
-- `OPENAI_API_KEY` - Optional, for some Hecate features
-
-See `.env.example` for all available configuration options including debug settings and terminal tool configuration.
-
-## Toolsets System
-
-The agent uses a toolsets system for organizing and managing tools. All tools must be part of a toolset to be accessible - individual tool selection is not supported. This ensures consistent and logical grouping of capabilities.
-
-### Key Concepts
-
-- **Toolsets**: Logical groups of tools for specific use cases (e.g., "research", "development", "debugging")
-- **Composition**: Toolsets can include other toolsets for powerful combinations
-- **Custom Toolsets**: Create your own toolsets at runtime or by editing `toolsets.py`
-- **Toolset-Only Access**: Tools are only accessible through toolsets, not individually
-
-### Available Toolsets
-
-See `toolsets.py` for the complete list of predefined toolsets including:
-- Basic toolsets (web, terminal, vision, creative, reasoning)
-- Composite toolsets (research, development, analysis, etc.)
-- Scenario-specific toolsets (debugging, documentation, API testing, etc.)
-- Special toolsets (safe mode without terminal, minimal, offline)
-
-### Using Toolsets
+## Quick Install
 
 ```bash
-# Use a predefined toolset
-python run_agent.py --enabled_toolsets=research --query "Find latest AI papers"
-
-# Combine multiple toolsets
-python run_agent.py --enabled_toolsets=web,vision --query "Analyze this website"
-
-# Enable all toolsets explicitly (same as omitting the flag)
-python run_agent.py --enabled_toolsets=all --query "Do web research and run commands if helpful"
-
-# Safe mode (no terminal access)
-python run_agent.py --enabled_toolsets=safe --query "Help without running commands"
-
-# List all available toolsets and tools
-python run_agent.py --list_tools
+curl -fsSL https://raw.githubusercontent.com/NousResearch/hermes-agent/main/scripts/install.sh | bash
 ```
 
-For detailed documentation on toolsets, see `TOOLSETS_README.md`.
+Works on Linux, macOS, and WSL2. The installer handles everything — Python, Node.js, dependencies, and the `hermes` command. No prerequisites except git.
 
-## Basic Usage
+> **Windows:** Native Windows is not supported. Please install [WSL2](https://learn.microsoft.com/en-us/windows/wsl/install) and run the command above.
 
-### Default (all tools enabled)
-```bash
-python run_agent.py \
-  --query "search up the latest docs on jit in python 3.13 and write me basic example that's not in their docs. profile its perf" \
-  --max_turns 20 \
-  --model claude-sonnet-4-20250514 \
-  --base_url https://api.anthropic.com/v1/ \
-  --api_key $ANTHROPIC_API_KEY
-```
-
-### With specific toolset
-```bash
-python run_agent.py \
-  --query "Debug this Python error" \
-  --enabled_toolsets=debugging \
-  --model claude-sonnet-4-20250514 \
-  --api_key $ANTHROPIC_API_KEY
-```
-
-### Python API
-```python
-from run_agent import AIAgent
-
-# Use a specific toolset
-agent = AIAgent(
-    model="claude-opus-4-20250514",
-    enabled_toolsets=["research"]
-)
-response = agent.chat("Find information about quantum computing")
-
-# Create custom toolset at runtime
-from toolsets import create_custom_toolset
-
-create_custom_toolset(
-    name="my_tools",
-    description="My custom toolkit",
-    tools=["web_search"],
-    includes=["terminal", "vision"]
-)
-
-agent = AIAgent(enabled_toolsets=["my_tools"])
-```
-
-## Batch Processing
-
-Process multiple prompts from a dataset in parallel with automatic checkpointing and statistics tracking:
+After installation:
 
 ```bash
-# Basic batch processing
-python batch_runner.py \
-  --dataset_file=prompts.jsonl \
-  --batch_size=20 \
-  --run_name=my_run
-
-# With specific distribution
-python batch_runner.py \
-  --dataset_file=prompts.jsonl \
-  --batch_size=20 \
-  --run_name=image_run \
-  --distribution=image_gen \
-  --num_workers=4
+source ~/.bashrc    # reload shell (or: source ~/.zshrc)
+hermes              # start chatting!
 ```
 
-**Key Features:**
-- Parallel processing with configurable workers
-- Toolset distributions for varied data generation
-- Automatic checkpointing and resume capability
-- Combined output in `data/<run_name>/trajectories.jsonl`
-- Tool usage statistics and success rates
+---
 
-**Quick Start:** See [QUICKSTART_BATCH.md](QUICKSTART_BATCH.md) for a 5-minute getting started guide.  
-**Full Documentation:** See [BATCH_PROCESSING.md](BATCH_PROCESSING.md) for comprehensive documentation.
+## Getting Started
 
-### Ephemeral System Prompts
-
-The ephemeral system prompt feature allows you to guide the model's behavior during batch processing **without** saving that prompt to the training dataset trajectories. This is useful for:
-
-- Guiding model behavior during data collection
-- Adding task-specific instructions 
-- Keeping saved trajectories clean and focused on tool-calling format
-
-**Example:**
 ```bash
-python batch_runner.py \
-  --dataset_file=prompts.jsonl \
-  --batch_size=10 \
-  --run_name=my_run \
-  --ephemeral_system_prompt="You are a helpful assistant focused on image generation."
+hermes              # Interactive CLI — start a conversation
+hermes model        # Choose your LLM provider and model
+hermes tools        # Configure which tools are enabled
+hermes config set   # Set individual config values
+hermes gateway      # Start the messaging gateway (Telegram, Discord, etc.)
+hermes setup        # Run the full setup wizard (configures everything at once)
+hermes claw migrate # Migrate from OpenClaw (if coming from OpenClaw)
+hermes update       # Update to the latest version
+hermes doctor       # Diagnose any issues
 ```
 
-The ephemeral prompt will influence the model's behavior during execution, but **only the standard tool-calling system prompt** will be saved in the trajectory files.
+📖 **[Full documentation →](https://hermes-agent.nousresearch.com/docs/)**
 
-**Documentation:** See [docs/ephemeral_system_prompt.md](docs/ephemeral_system_prompt.md) for complete details.
+## CLI vs Messaging Quick Reference
 
-## Command Line Arguments
+Hermes has two entry points: start the terminal UI with `hermes`, or run the gateway and talk to it from Telegram, Discord, Slack, WhatsApp, Signal, or Email. Once you're in a conversation, many slash commands are shared across both interfaces.
 
-**Single Agent (`run_agent.py`):**
-- `--query`: The question or task for the agent
-- `--model`: Model to use (default: claude-opus-4-20250514)
-- `--api_key`: API key for authentication
-- `--base_url`: API endpoint URL
-- `--max_turns`: Maximum number of tool-calling iterations
-- `--enabled_toolsets`: Comma-separated list of toolsets to enable. Use `all` (or `*`) to enable everything. If omitted, all toolsets are enabled by default.
-- `--disabled_toolsets`: Comma-separated list of toolsets to disable
-- `--list_tools`: List all available toolsets and tools
-- `--save_trajectories`: Save conversation trajectories to JSONL files
+| Action | CLI | Messaging platforms |
+|---------|-----|---------------------|
+| Start chatting | `hermes` | Run `hermes gateway setup` + `hermes gateway start`, then send the bot a message |
+| Start fresh conversation | `/new` or `/reset` | `/new` or `/reset` |
+| Change model | `/model [provider:model]` | `/model [provider:model]` |
+| Set a personality | `/personality [name]` | `/personality [name]` |
+| Retry or undo the last turn | `/retry`, `/undo` | `/retry`, `/undo` |
+| Compress context / check usage | `/compress`, `/usage`, `/insights [--days N]` | `/compress`, `/usage`, `/insights [days]` |
+| Browse skills | `/skills` or `/<skill-name>` | `/skills` or `/<skill-name>` |
+| Interrupt current work | `Ctrl+C` or send a new message | `/stop` or send a new message |
+| Platform-specific status | `/platforms` | `/status`, `/sethome` |
 
-**Batch Processing (`batch_runner.py`):**
-- `--dataset_file`: Path to JSONL file with prompts
-- `--batch_size`: Number of prompts per batch
-- `--run_name`: Name for this run (for output/checkpointing)
-- `--distribution`: Toolset distribution to use (default: "default")
-- `--num_workers`: Number of parallel workers (default: 4)
-- `--resume`: Resume from checkpoint if interrupted
-- `--ephemeral_system_prompt`: System prompt used during execution but NOT saved to trajectories
-- `--list_distributions`: List available toolset distributions
+For the full command lists, see the [CLI guide](https://hermes-agent.nousresearch.com/docs/user-guide/cli) and the [Messaging Gateway guide](https://hermes-agent.nousresearch.com/docs/user-guide/messaging).
 
-## Environment Variables
-
-All environment variables can be configured in the `.env` file (copy from `.env.example`).
-
-**Core API Keys:**
-- `ANTHROPIC_API_KEY`: Main agent model
-- `FIRECRAWL_API_KEY`: Web tools (search, extract, crawl)
-- `NOUS_API_KEY`: Vision and reasoning tools
-- `MORPH_API_KEY`: Terminal tools
-- `FAL_KEY`: Image generation tools
-- `OPENAI_API_KEY`: Optional, for some Hecate features
-
-**Configuration Options:**
-- `HECATE_VM_LIFETIME_SECONDS`: VM lifetime (default: 300)
-- `HECATE_DEFAULT_SNAPSHOT_ID`: Default snapshot (default: snapshot_p5294qxt)
-- `WEB_TOOLS_DEBUG`, `VISION_TOOLS_DEBUG`, `MOA_TOOLS_DEBUG`, `IMAGE_TOOLS_DEBUG`: Enable debug logging
+---
 
 ## Documentation
 
-**Single Agent Usage:**
-- `TOOLSETS_README.md`: Comprehensive guide to the toolsets system
-- `toolsets.py`: View and modify available toolsets
-- `model_tools.py`: Core tool definitions and handlers
+All documentation lives at **[hermes-agent.nousresearch.com/docs](https://hermes-agent.nousresearch.com/docs/)**:
 
-**Batch Processing:**
-- `QUICKSTART_BATCH.md`: 5-minute quick start guide
-- `BATCH_PROCESSING.md`: Complete batch processing documentation
-- `toolset_distributions.py`: Toolset distributions for data generation
+| Section | What's Covered |
+|---------|---------------|
+| [Quickstart](https://hermes-agent.nousresearch.com/docs/getting-started/quickstart) | Install → setup → first conversation in 2 minutes |
+| [CLI Usage](https://hermes-agent.nousresearch.com/docs/user-guide/cli) | Commands, keybindings, personalities, sessions |
+| [Configuration](https://hermes-agent.nousresearch.com/docs/user-guide/configuration) | Config file, providers, models, all options |
+| [Messaging Gateway](https://hermes-agent.nousresearch.com/docs/user-guide/messaging) | Telegram, Discord, Slack, WhatsApp, Signal, Home Assistant |
+| [Security](https://hermes-agent.nousresearch.com/docs/user-guide/security) | Command approval, DM pairing, container isolation |
+| [Tools & Toolsets](https://hermes-agent.nousresearch.com/docs/user-guide/features/tools) | 40+ tools, toolset system, terminal backends |
+| [Skills System](https://hermes-agent.nousresearch.com/docs/user-guide/features/skills) | Procedural memory, Skills Hub, creating skills |
+| [Memory](https://hermes-agent.nousresearch.com/docs/user-guide/features/memory) | Persistent memory, user profiles, best practices |
+| [MCP Integration](https://hermes-agent.nousresearch.com/docs/user-guide/features/mcp) | Connect any MCP server for extended capabilities |
+| [Cron Scheduling](https://hermes-agent.nousresearch.com/docs/user-guide/features/cron) | Scheduled tasks with platform delivery |
+| [Context Files](https://hermes-agent.nousresearch.com/docs/user-guide/features/context-files) | Project context that shapes every conversation |
+| [Architecture](https://hermes-agent.nousresearch.com/docs/developer-guide/architecture) | Project structure, agent loop, key classes |
+| [Contributing](https://hermes-agent.nousresearch.com/docs/developer-guide/contributing) | Development setup, PR process, code style |
+| [CLI Reference](https://hermes-agent.nousresearch.com/docs/reference/cli-commands) | All commands and flags |
+| [Environment Variables](https://hermes-agent.nousresearch.com/docs/reference/environment-variables) | Complete env var reference |
 
-## Examples
+---
 
-See `TOOLSETS_README.md` for extensive examples of using different toolsets for various scenarios.
+## Migrating from OpenClaw
+
+If you're coming from OpenClaw, Hermes can automatically import your settings, memories, skills, and API keys.
+
+**During first-time setup:** The setup wizard (`hermes setup`) automatically detects `~/.openclaw` and offers to migrate before configuration begins.
+
+**Anytime after install:**
+
+```bash
+hermes claw migrate              # Interactive migration (full preset)
+hermes claw migrate --dry-run    # Preview what would be migrated
+hermes claw migrate --preset user-data   # Migrate without secrets
+hermes claw migrate --overwrite  # Overwrite existing conflicts
+```
+
+What gets imported:
+- **SOUL.md** — persona file
+- **Memories** — MEMORY.md and USER.md entries
+- **Skills** — user-created skills → `~/.hermes/skills/openclaw-imports/`
+- **Command allowlist** — approval patterns
+- **Messaging settings** — platform configs, allowed users, working directory
+- **API keys** — allowlisted secrets (Telegram, OpenRouter, OpenAI, Anthropic, ElevenLabs)
+- **TTS assets** — workspace audio files
+- **Workspace instructions** — AGENTS.md (with `--workspace-target`)
+
+See `hermes claw migrate --help` for all options, or use the `openclaw-migration` skill for an interactive agent-guided migration with dry-run previews.
+
+---
+
+## Contributing
+
+We welcome contributions! See the [Contributing Guide](https://hermes-agent.nousresearch.com/docs/developer-guide/contributing) for development setup, code style, and PR process.
+
+Quick start for contributors:
+
+```bash
+git clone https://github.com/NousResearch/hermes-agent.git
+cd hermes-agent
+curl -LsSf https://astral.sh/uv/install.sh | sh
+uv venv venv --python 3.11
+source venv/bin/activate
+uv pip install -e ".[all,dev]"
+python -m pytest tests/ -q
+```
+
+> **RL Training (optional):** To work on the RL/Tinker-Atropos integration:
+> ```bash
+> git submodule update --init tinker-atropos
+> uv pip install -e "./tinker-atropos"
+> ```
+
+---
+
+## Community
+
+- 💬 [Discord](https://discord.gg/NousResearch)
+- 📚 [Skills Hub](https://agentskills.io)
+- 🐛 [Issues](https://github.com/NousResearch/hermes-agent/issues)
+- 💡 [Discussions](https://github.com/NousResearch/hermes-agent/discussions)
+
+---
+
+## License
+
+MIT — see [LICENSE](LICENSE).
+
+Built by [Nous Research](https://nousresearch.com).

RELEASE_v0.2.0.md

@@ -0,0 +1,383 @@
+# Hermes Agent v0.2.0 (v2026.3.12)
+
+**Release Date:** March 12, 2026
+
+> First tagged release since v0.1.0 (the initial pre-public foundation). In just over two weeks, Hermes Agent went from a small internal project to a full-featured AI agent platform — thanks to an explosion of community contributions. This release covers **216 merged pull requests** from **63 contributors**, resolving **119 issues**.
+
+---
+
+## ✨ Highlights
+
+- **Multi-Platform Messaging Gateway** — Telegram, Discord, Slack, WhatsApp, Signal, Email (IMAP/SMTP), and Home Assistant platforms with unified session management, media attachments, and per-platform tool configuration.
+
+- **MCP (Model Context Protocol) Client** — Native MCP support with stdio and HTTP transports, reconnection, resource/prompt discovery, and sampling (server-initiated LLM requests). ([#291](https://github.com/NousResearch/hermes-agent/pull/291) — @0xbyt4, [#301](https://github.com/NousResearch/hermes-agent/pull/301), [#753](https://github.com/NousResearch/hermes-agent/pull/753))
+
+- **Skills Ecosystem** — 70+ bundled and optional skills across 15+ categories with a Skills Hub for community discovery, per-platform enable/disable, conditional activation based on tool availability, and prerequisite validation. ([#743](https://github.com/NousResearch/hermes-agent/pull/743) — @teyrebaz33, [#785](https://github.com/NousResearch/hermes-agent/pull/785) — @teyrebaz33)
+
+- **Centralized Provider Router** — Unified `call_llm()`/`async_call_llm()` API replaces scattered provider logic across vision, summarization, compression, and trajectory saving. All auxiliary consumers route through a single code path with automatic credential resolution. ([#1003](https://github.com/NousResearch/hermes-agent/pull/1003))
+
+- **ACP Server** — VS Code, Zed, and JetBrains editor integration via the Agent Communication Protocol standard. ([#949](https://github.com/NousResearch/hermes-agent/pull/949))
+
+- **CLI Skin/Theme Engine** — Data-driven visual customization: banners, spinners, colors, branding. 7 built-in skins + custom YAML skins.
+
+- **Git Worktree Isolation** — `hermes -w` launches isolated agent sessions in git worktrees for safe parallel work on the same repo. ([#654](https://github.com/NousResearch/hermes-agent/pull/654))
+
+- **Filesystem Checkpoints & Rollback** — Automatic snapshots before destructive operations with `/rollback` to restore. ([#824](https://github.com/NousResearch/hermes-agent/pull/824))
+
+- **3,289 Tests** — From near-zero test coverage to a comprehensive test suite covering agent, gateway, tools, cron, and CLI.
+
+---
+
+## 🏗️ Core Agent & Architecture
+
+### Provider & Model Support
+- Centralized provider router with `resolve_provider_client()` + `call_llm()` API ([#1003](https://github.com/NousResearch/hermes-agent/pull/1003))
+- Nous Portal as first-class provider in setup ([#644](https://github.com/NousResearch/hermes-agent/issues/644))
+- OpenAI Codex (Responses API) with ChatGPT subscription support ([#43](https://github.com/NousResearch/hermes-agent/pull/43)) — @grp06
+- Codex OAuth vision support + multimodal content adapter
+- Validate `/model` against live API instead of hardcoded lists
+- Self-hosted Firecrawl support ([#460](https://github.com/NousResearch/hermes-agent/pull/460)) — @caentzminger
+- Kimi Code API support ([#635](https://github.com/NousResearch/hermes-agent/pull/635)) — @christomitov
+- MiniMax model ID update ([#473](https://github.com/NousResearch/hermes-agent/pull/473)) — @tars90percent
+- OpenRouter provider routing configuration (provider_preferences)
+- Nous credential refresh on 401 errors ([#571](https://github.com/NousResearch/hermes-agent/pull/571), [#269](https://github.com/NousResearch/hermes-agent/pull/269)) — @rewbs
+- z.ai/GLM, Kimi/Moonshot, MiniMax, Azure OpenAI as first-class providers
+- Unified `/model` and `/provider` into single view
+
+### Agent Loop & Conversation
+- Simple fallback model for provider resilience ([#740](https://github.com/NousResearch/hermes-agent/pull/740))
+- Shared iteration budget across parent + subagent delegation
+- Iteration budget pressure via tool result injection
+- Configurable subagent provider/model with full credential resolution
+- Handle 413 payload-too-large via compression instead of aborting ([#153](https://github.com/NousResearch/hermes-agent/pull/153)) — @tekelala
+- Retry with rebuilt payload after compression ([#616](https://github.com/NousResearch/hermes-agent/pull/616)) — @tripledoublev
+- Auto-compress pathologically large gateway sessions ([#628](https://github.com/NousResearch/hermes-agent/issues/628))
+- Tool call repair middleware — auto-lowercase and invalid tool handler
+- Reasoning effort configuration and `/reasoning` command ([#921](https://github.com/NousResearch/hermes-agent/pull/921))
+- Detect and block file re-read/search loops after context compression ([#705](https://github.com/NousResearch/hermes-agent/pull/705)) — @0xbyt4
+
+### Session & Memory
+- Session naming with unique titles, auto-lineage, rich listing, and resume by name ([#720](https://github.com/NousResearch/hermes-agent/pull/720))
+- Interactive session browser with search filtering ([#733](https://github.com/NousResearch/hermes-agent/pull/733))
+- Display previous messages when resuming a session ([#734](https://github.com/NousResearch/hermes-agent/pull/734))
+- Honcho AI-native cross-session user modeling ([#38](https://github.com/NousResearch/hermes-agent/pull/38)) — @erosika
+- Proactive async memory flush on session expiry
+- Smart context length probing with persistent caching + banner display
+- `/resume` command for switching to named sessions in gateway
+- Session reset policy for messaging platforms
+
+---
+
+## 📱 Messaging Platforms (Gateway)
+
+### Telegram
+- Native file attachments: send_document + send_video
+- Document file processing for PDF, text, and Office files — @tekelala
+- Forum topic session isolation ([#766](https://github.com/NousResearch/hermes-agent/pull/766)) — @spanishflu-est1918
+- Browser screenshot sharing via MEDIA: protocol ([#657](https://github.com/NousResearch/hermes-agent/pull/657))
+- Location support for find-nearby skill
+- TTS voice message accumulation fix ([#176](https://github.com/NousResearch/hermes-agent/pull/176)) — @Bartok9
+- Improved error handling and logging ([#763](https://github.com/NousResearch/hermes-agent/pull/763)) — @aydnOktay
+- Italic regex newline fix + 43 format tests ([#204](https://github.com/NousResearch/hermes-agent/pull/204)) — @0xbyt4
+
+### Discord
+- Channel topic included in session context ([#248](https://github.com/NousResearch/hermes-agent/pull/248)) — @Bartok9
+- DISCORD_ALLOW_BOTS config for bot message filtering ([#758](https://github.com/NousResearch/hermes-agent/pull/758))
+- Document and video support ([#784](https://github.com/NousResearch/hermes-agent/pull/784))
+- Improved error handling and logging ([#761](https://github.com/NousResearch/hermes-agent/pull/761)) — @aydnOktay
+
+### Slack
+- App_mention 404 fix + document/video support ([#784](https://github.com/NousResearch/hermes-agent/pull/784))
+- Structured logging replacing print statements — @aydnOktay
+
+### WhatsApp
+- Native media sending — images, videos, documents ([#292](https://github.com/NousResearch/hermes-agent/pull/292)) — @satelerd
+- Multi-user session isolation ([#75](https://github.com/NousResearch/hermes-agent/pull/75)) — @satelerd
+- Cross-platform port cleanup replacing Linux-only fuser ([#433](https://github.com/NousResearch/hermes-agent/pull/433)) — @Farukest
+- DM interrupt key mismatch fix ([#350](https://github.com/NousResearch/hermes-agent/pull/350)) — @Farukest
+
+### Signal
+- Full Signal messenger gateway via signal-cli-rest-api ([#405](https://github.com/NousResearch/hermes-agent/issues/405))
+- Media URL support in message events ([#871](https://github.com/NousResearch/hermes-agent/pull/871))
+
+### Email (IMAP/SMTP)
+- New email gateway platform — @0xbyt4
+
+### Home Assistant
+- REST tools + WebSocket gateway integration ([#184](https://github.com/NousResearch/hermes-agent/pull/184)) — @0xbyt4
+- Service discovery and enhanced setup
+- Toolset mapping fix ([#538](https://github.com/NousResearch/hermes-agent/pull/538)) — @Himess
+
+### Gateway Core
+- Expose subagent tool calls and thinking to users ([#186](https://github.com/NousResearch/hermes-agent/pull/186)) — @cutepawss
+- Configurable background process watcher notifications ([#840](https://github.com/NousResearch/hermes-agent/pull/840))
+- `edit_message()` for Telegram/Discord/Slack with fallback
+- `/compress`, `/usage`, `/update` slash commands
+- Eliminated 3x SQLite message duplication in gateway sessions ([#873](https://github.com/NousResearch/hermes-agent/pull/873))
+- Stabilize system prompt across gateway turns for cache hits ([#754](https://github.com/NousResearch/hermes-agent/pull/754))
+- MCP server shutdown on gateway exit ([#796](https://github.com/NousResearch/hermes-agent/pull/796)) — @0xbyt4
+- Pass session_db to AIAgent, fixing session_search error ([#108](https://github.com/NousResearch/hermes-agent/pull/108)) — @Bartok9
+- Persist transcript changes in /retry, /undo; fix /reset attribute ([#217](https://github.com/NousResearch/hermes-agent/pull/217)) — @Farukest
+- UTF-8 encoding fix preventing Windows crashes ([#369](https://github.com/NousResearch/hermes-agent/pull/369)) — @ch3ronsa
+
+---
+
+## 🖥️ CLI & User Experience
+
+### Interactive CLI
+- Data-driven skin/theme engine — 7 built-in skins (default, ares, mono, slate, poseidon, sisyphus, charizard) + custom YAML skins
+- `/personality` command with custom personality + disable support ([#773](https://github.com/NousResearch/hermes-agent/pull/773)) — @teyrebaz33
+- User-defined quick commands that bypass the agent loop ([#746](https://github.com/NousResearch/hermes-agent/pull/746)) — @teyrebaz33
+- `/reasoning` command for effort level and display toggle ([#921](https://github.com/NousResearch/hermes-agent/pull/921))
+- `/verbose` slash command to toggle debug at runtime ([#94](https://github.com/NousResearch/hermes-agent/pull/94)) — @cesareth
+- `/insights` command — usage analytics, cost estimation & activity patterns ([#552](https://github.com/NousResearch/hermes-agent/pull/552))
+- `/background` command for managing background processes
+- `/help` formatting with command categories
+- Bell-on-complete — terminal bell when agent finishes ([#738](https://github.com/NousResearch/hermes-agent/pull/738))
+- Up/down arrow history navigation
+- Clipboard image paste (Alt+V / Ctrl+V)
+- Loading indicators for slow slash commands ([#882](https://github.com/NousResearch/hermes-agent/pull/882))
+- Spinner flickering fix under patch_stdout ([#91](https://github.com/NousResearch/hermes-agent/pull/91)) — @0xbyt4
+- `--quiet/-Q` flag for programmatic single-query mode
+- `--fuck-it-ship-it` flag to bypass all approval prompts ([#724](https://github.com/NousResearch/hermes-agent/pull/724)) — @dmahan93
+- Tools summary flag ([#767](https://github.com/NousResearch/hermes-agent/pull/767)) — @luisv-1
+- Terminal blinking fix on SSH ([#284](https://github.com/NousResearch/hermes-agent/pull/284)) — @ygd58
+- Multi-line paste detection fix ([#84](https://github.com/NousResearch/hermes-agent/pull/84)) — @0xbyt4
+
+### Setup & Configuration
+- Modular setup wizard with section subcommands and tool-first UX
+- Container resource configuration prompts
+- Backend validation for required binaries
+- Config migration system (currently v7)
+- API keys properly routed to .env instead of config.yaml ([#469](https://github.com/NousResearch/hermes-agent/pull/469)) — @ygd58
+- Atomic write for .env to prevent API key loss on crash ([#954](https://github.com/NousResearch/hermes-agent/pull/954))
+- `hermes tools` — per-platform tool enable/disable with curses UI
+- `hermes doctor` for health checks across all configured providers
+- `hermes update` with auto-restart for gateway service
+- Show update-available notice in CLI banner
+- Multiple named custom providers
+- Shell config detection improvement for PATH setup ([#317](https://github.com/NousResearch/hermes-agent/pull/317)) — @mehmetkr-31
+- Consistent HERMES_HOME and .env path resolution ([#51](https://github.com/NousResearch/hermes-agent/pull/51), [#48](https://github.com/NousResearch/hermes-agent/pull/48)) — @deankerr
+- Docker backend fix on macOS + subagent auth for Nous Portal ([#46](https://github.com/NousResearch/hermes-agent/pull/46)) — @rsavitt
+
+---
+
+## 🔧 Tool System
+
+### MCP (Model Context Protocol)
+- Native MCP client with stdio + HTTP transports ([#291](https://github.com/NousResearch/hermes-agent/pull/291) — @0xbyt4, [#301](https://github.com/NousResearch/hermes-agent/pull/301))
+- Sampling support — server-initiated LLM requests ([#753](https://github.com/NousResearch/hermes-agent/pull/753))
+- Resource and prompt discovery
+- Automatic reconnection and security hardening
+- Banner integration, `/reload-mcp` command
+- `hermes tools` UI integration
+
+### Browser
+- Local browser backend — zero-cost headless Chromium (no Browserbase needed)
+- Console/errors tool, annotated screenshots, auto-recording, dogfood QA skill ([#745](https://github.com/NousResearch/hermes-agent/pull/745))
+- Screenshot sharing via MEDIA: on all messaging platforms ([#657](https://github.com/NousResearch/hermes-agent/pull/657))
+
+### Terminal & Execution
+- `execute_code` sandbox with json_parse, shell_quote, retry helpers
+- Docker: custom volume mounts ([#158](https://github.com/NousResearch/hermes-agent/pull/158)) — @Indelwin
+- Daytona cloud sandbox backend ([#451](https://github.com/NousResearch/hermes-agent/pull/451)) — @rovle
+- SSH backend fix ([#59](https://github.com/NousResearch/hermes-agent/pull/59)) — @deankerr
+- Shell noise filtering and login shell execution for environment consistency
+- Head+tail truncation for execute_code stdout overflow
+- Configurable background process notification modes
+
+### File Operations
+- Filesystem checkpoints and `/rollback` command ([#824](https://github.com/NousResearch/hermes-agent/pull/824))
+- Structured tool result hints (next-action guidance) for patch and search_files ([#722](https://github.com/NousResearch/hermes-agent/issues/722))
+- Docker volumes passed to sandbox container config ([#687](https://github.com/NousResearch/hermes-agent/pull/687)) — @manuelschipper
+
+---
+
+## 🧩 Skills Ecosystem
+
+### Skills System
+- Per-platform skill enable/disable ([#743](https://github.com/NousResearch/hermes-agent/pull/743)) — @teyrebaz33
+- Conditional skill activation based on tool availability ([#785](https://github.com/NousResearch/hermes-agent/pull/785)) — @teyrebaz33
+- Skill prerequisites — hide skills with unmet dependencies ([#659](https://github.com/NousResearch/hermes-agent/pull/659)) — @kshitijk4poor
+- Optional skills — shipped but not activated by default
+- `hermes skills browse` — paginated hub browsing
+- Skills sub-category organization
+- Platform-conditional skill loading
+- Atomic skill file writes ([#551](https://github.com/NousResearch/hermes-agent/pull/551)) — @aydnOktay
+- Skills sync data loss prevention ([#563](https://github.com/NousResearch/hermes-agent/pull/563)) — @0xbyt4
+- Dynamic skill slash commands for CLI and gateway
+
+### New Skills (selected)
+- **ASCII Art** — pyfiglet (571 fonts), cowsay, image-to-ascii ([#209](https://github.com/NousResearch/hermes-agent/pull/209)) — @0xbyt4
+- **ASCII Video** — Full production pipeline ([#854](https://github.com/NousResearch/hermes-agent/pull/854)) — @SHL0MS
+- **DuckDuckGo Search** — Firecrawl fallback ([#267](https://github.com/NousResearch/hermes-agent/pull/267)) — @gamedevCloudy; DDGS API expansion ([#598](https://github.com/NousResearch/hermes-agent/pull/598)) — @areu01or00
+- **Solana Blockchain** — Wallet balances, USD pricing, token names ([#212](https://github.com/NousResearch/hermes-agent/pull/212)) — @gizdusum
+- **AgentMail** — Agent-owned email inboxes ([#330](https://github.com/NousResearch/hermes-agent/pull/330)) — @teyrebaz33
+- **Polymarket** — Prediction market data (read-only) ([#629](https://github.com/NousResearch/hermes-agent/pull/629))
+- **OpenClaw Migration** — Official migration tool ([#570](https://github.com/NousResearch/hermes-agent/pull/570)) — @unmodeled-tyler
+- **Domain Intelligence** — Passive recon: subdomains, SSL, WHOIS, DNS ([#136](https://github.com/NousResearch/hermes-agent/pull/136)) — @FurkanL0
+- **Superpowers** — Software development skills ([#137](https://github.com/NousResearch/hermes-agent/pull/137)) — @kaos35
+- **Hermes-Atropos** — RL environment development skill ([#815](https://github.com/NousResearch/hermes-agent/pull/815))
+- Plus: arXiv search, OCR/documents, Excalidraw diagrams, YouTube transcripts, GIF search, Pokémon player, Minecraft modpack server, OpenHue (Philips Hue), Google Workspace, Notion, PowerPoint, Obsidian, find-nearby, and 40+ MLOps skills
+
+---
+
+## 🔒 Security & Reliability
+
+### Security Hardening
+- Path traversal fix in skill_view — prevented reading arbitrary files ([#220](https://github.com/NousResearch/hermes-agent/issues/220)) — @Farukest
+- Shell injection prevention in sudo password piping ([#65](https://github.com/NousResearch/hermes-agent/pull/65)) — @leonsgithub
+- Dangerous command detection: multiline bypass fix ([#233](https://github.com/NousResearch/hermes-agent/pull/233)) — @Farukest; tee/process substitution patterns ([#280](https://github.com/NousResearch/hermes-agent/pull/280)) — @dogiladeveloper
+- Symlink boundary check fix in skills_guard ([#386](https://github.com/NousResearch/hermes-agent/pull/386)) — @Farukest
+- Symlink bypass fix in write deny list on macOS ([#61](https://github.com/NousResearch/hermes-agent/pull/61)) — @0xbyt4
+- Multi-word prompt injection bypass prevention ([#192](https://github.com/NousResearch/hermes-agent/pull/192)) — @0xbyt4
+- Cron prompt injection scanner bypass fix ([#63](https://github.com/NousResearch/hermes-agent/pull/63)) — @0xbyt4
+- Enforce 0600/0700 file permissions on sensitive files ([#757](https://github.com/NousResearch/hermes-agent/pull/757))
+- .env file permissions restricted to owner-only ([#529](https://github.com/NousResearch/hermes-agent/pull/529)) — @Himess
+- `--force` flag properly blocked from overriding dangerous verdicts ([#388](https://github.com/NousResearch/hermes-agent/pull/388)) — @Farukest
+- FTS5 query sanitization + DB connection leak fix ([#565](https://github.com/NousResearch/hermes-agent/pull/565)) — @0xbyt4
+- Expand secret redaction patterns + config toggle to disable
+- In-memory permanent allowlist to prevent data leak ([#600](https://github.com/NousResearch/hermes-agent/pull/600)) — @alireza78a
+
+### Atomic Writes (data loss prevention)
+- sessions.json ([#611](https://github.com/NousResearch/hermes-agent/pull/611)) — @alireza78a
+- Cron jobs ([#146](https://github.com/NousResearch/hermes-agent/pull/146)) — @alireza78a
+- .env config ([#954](https://github.com/NousResearch/hermes-agent/pull/954))
+- Process checkpoints ([#298](https://github.com/NousResearch/hermes-agent/pull/298)) — @aydnOktay
+- Batch runner ([#297](https://github.com/NousResearch/hermes-agent/pull/297)) — @aydnOktay
+- Skill files ([#551](https://github.com/NousResearch/hermes-agent/pull/551)) — @aydnOktay
+
+### Reliability
+- Guard all print() against OSError for systemd/headless environments ([#963](https://github.com/NousResearch/hermes-agent/pull/963))
+- Reset all retry counters at start of run_conversation ([#607](https://github.com/NousResearch/hermes-agent/pull/607)) — @0xbyt4
+- Return deny on approval callback timeout instead of None ([#603](https://github.com/NousResearch/hermes-agent/pull/603)) — @0xbyt4
+- Fix None message content crashes across codebase ([#277](https://github.com/NousResearch/hermes-agent/pull/277))
+- Fix context overrun crash with local LLM backends ([#403](https://github.com/NousResearch/hermes-agent/pull/403)) — @ch3ronsa
+- Prevent `_flush_sentinel` from leaking to external APIs ([#227](https://github.com/NousResearch/hermes-agent/pull/227)) — @Farukest
+- Prevent conversation_history mutation in callers ([#229](https://github.com/NousResearch/hermes-agent/pull/229)) — @Farukest
+- Fix systemd restart loop ([#614](https://github.com/NousResearch/hermes-agent/pull/614)) — @voidborne-d
+- Close file handles and sockets to prevent fd leaks ([#568](https://github.com/NousResearch/hermes-agent/pull/568) — @alireza78a, [#296](https://github.com/NousResearch/hermes-agent/pull/296) — @alireza78a, [#709](https://github.com/NousResearch/hermes-agent/pull/709) — @memosr)
+- Prevent data loss in clipboard PNG conversion ([#602](https://github.com/NousResearch/hermes-agent/pull/602)) — @0xbyt4
+- Eliminate shell noise from terminal output ([#293](https://github.com/NousResearch/hermes-agent/pull/293)) — @0xbyt4
+- Timezone-aware now() for prompt, cron, and execute_code ([#309](https://github.com/NousResearch/hermes-agent/pull/309)) — @areu01or00
+
+### Windows Compatibility
+- Guard POSIX-only process functions ([#219](https://github.com/NousResearch/hermes-agent/pull/219)) — @Farukest
+- Windows native support via Git Bash + ZIP-based update fallback
+- pywinpty for PTY support ([#457](https://github.com/NousResearch/hermes-agent/pull/457)) — @shitcoinsherpa
+- Explicit UTF-8 encoding on all config/data file I/O ([#458](https://github.com/NousResearch/hermes-agent/pull/458)) — @shitcoinsherpa
+- Windows-compatible path handling ([#354](https://github.com/NousResearch/hermes-agent/pull/354), [#390](https://github.com/NousResearch/hermes-agent/pull/390)) — @Farukest
+- Regex-based search output parsing for drive-letter paths ([#533](https://github.com/NousResearch/hermes-agent/pull/533)) — @Himess
+- Auth store file lock for Windows ([#455](https://github.com/NousResearch/hermes-agent/pull/455)) — @shitcoinsherpa
+
+---
+
+## 🐛 Notable Bug Fixes
+
+- Fix DeepSeek V3 tool call parser silently dropping multi-line JSON arguments ([#444](https://github.com/NousResearch/hermes-agent/pull/444)) — @PercyDikec
+- Fix gateway transcript losing 1 message per turn due to offset mismatch ([#395](https://github.com/NousResearch/hermes-agent/pull/395)) — @PercyDikec
+- Fix /retry command silently discarding the agent's final response ([#441](https://github.com/NousResearch/hermes-agent/pull/441)) — @PercyDikec
+- Fix max-iterations retry returning empty string after think-block stripping ([#438](https://github.com/NousResearch/hermes-agent/pull/438)) — @PercyDikec
+- Fix max-iterations retry using hardcoded max_tokens ([#436](https://github.com/NousResearch/hermes-agent/pull/436)) — @Farukest
+- Fix Codex status dict key mismatch ([#448](https://github.com/NousResearch/hermes-agent/pull/448)) and visibility filter ([#446](https://github.com/NousResearch/hermes-agent/pull/446)) — @PercyDikec
+- Strip \<think\> blocks from final user-facing responses ([#174](https://github.com/NousResearch/hermes-agent/pull/174)) — @Bartok9
+- Fix \<think\> block regex stripping visible content when model discusses tags literally ([#786](https://github.com/NousResearch/hermes-agent/issues/786))
+- Fix Mistral 422 errors from leftover finish_reason in assistant messages ([#253](https://github.com/NousResearch/hermes-agent/pull/253)) — @Sertug17
+- Fix OPENROUTER_API_KEY resolution order across all code paths ([#295](https://github.com/NousResearch/hermes-agent/pull/295)) — @0xbyt4
+- Fix OPENAI_BASE_URL API key priority ([#420](https://github.com/NousResearch/hermes-agent/pull/420)) — @manuelschipper
+- Fix Anthropic "prompt is too long" 400 error not detected as context length error ([#813](https://github.com/NousResearch/hermes-agent/issues/813))
+- Fix SQLite session transcript accumulating duplicate messages — 3-4x token inflation ([#860](https://github.com/NousResearch/hermes-agent/issues/860))
+- Fix setup wizard skipping API key prompts on first install ([#748](https://github.com/NousResearch/hermes-agent/pull/748))
+- Fix setup wizard showing OpenRouter model list for Nous Portal ([#575](https://github.com/NousResearch/hermes-agent/pull/575)) — @PercyDikec
+- Fix provider selection not persisting when switching via hermes model ([#881](https://github.com/NousResearch/hermes-agent/pull/881))
+- Fix Docker backend failing when docker not in PATH on macOS ([#889](https://github.com/NousResearch/hermes-agent/pull/889))
+- Fix ClawHub Skills Hub adapter for API endpoint changes ([#286](https://github.com/NousResearch/hermes-agent/pull/286)) — @BP602
+- Fix Honcho auto-enable when API key is present ([#243](https://github.com/NousResearch/hermes-agent/pull/243)) — @Bartok9
+- Fix duplicate 'skills' subparser crash on Python 3.11+ ([#898](https://github.com/NousResearch/hermes-agent/issues/898))
+- Fix memory tool entry parsing when content contains section sign ([#162](https://github.com/NousResearch/hermes-agent/pull/162)) — @aydnOktay
+- Fix piped install silently aborting when interactive prompts fail ([#72](https://github.com/NousResearch/hermes-agent/pull/72)) — @cutepawss
+- Fix false positives in recursive delete detection ([#68](https://github.com/NousResearch/hermes-agent/pull/68)) — @cutepawss
+- Fix Ruff lint warnings across codebase ([#608](https://github.com/NousResearch/hermes-agent/pull/608)) — @JackTheGit
+- Fix Anthropic native base URL fail-fast ([#173](https://github.com/NousResearch/hermes-agent/pull/173)) — @adavyas
+- Fix install.sh creating ~/.hermes before moving Node.js directory ([#53](https://github.com/NousResearch/hermes-agent/pull/53)) — @JoshuaMart
+- Fix SystemExit traceback during atexit cleanup on Ctrl+C ([#55](https://github.com/NousResearch/hermes-agent/pull/55)) — @bierlingm
+- Restore missing MIT license file ([#620](https://github.com/NousResearch/hermes-agent/pull/620)) — @stablegenius49
+
+---
+
+## 🧪 Testing
+
+- **3,289 tests** across agent, gateway, tools, cron, and CLI
+- Parallelized test suite with pytest-xdist ([#802](https://github.com/NousResearch/hermes-agent/pull/802)) — @OutThisLife
+- Unit tests batch 1: 8 core modules ([#60](https://github.com/NousResearch/hermes-agent/pull/60)) — @0xbyt4
+- Unit tests batch 2: 8 more modules ([#62](https://github.com/NousResearch/hermes-agent/pull/62)) — @0xbyt4
+- Unit tests batch 3: 8 untested modules ([#191](https://github.com/NousResearch/hermes-agent/pull/191)) — @0xbyt4
+- Unit tests batch 4: 5 security/logic-critical modules ([#193](https://github.com/NousResearch/hermes-agent/pull/193)) — @0xbyt4
+- AIAgent (run_agent.py) unit tests ([#67](https://github.com/NousResearch/hermes-agent/pull/67)) — @0xbyt4
+- Trajectory compressor tests ([#203](https://github.com/NousResearch/hermes-agent/pull/203)) — @0xbyt4
+- Clarify tool tests ([#121](https://github.com/NousResearch/hermes-agent/pull/121)) — @Bartok9
+- Telegram format tests — 43 tests for italic/bold/code rendering ([#204](https://github.com/NousResearch/hermes-agent/pull/204)) — @0xbyt4
+- Vision tools type hints + 42 tests ([#792](https://github.com/NousResearch/hermes-agent/pull/792))
+- Compressor tool-call boundary regression tests ([#648](https://github.com/NousResearch/hermes-agent/pull/648)) — @intertwine
+- Test structure reorganization ([#34](https://github.com/NousResearch/hermes-agent/pull/34)) — @0xbyt4
+- Shell noise elimination + fix 36 test failures ([#293](https://github.com/NousResearch/hermes-agent/pull/293)) — @0xbyt4
+
+---
+
+## 🔬 RL & Evaluation Environments
+
+- WebResearchEnv — Multi-step web research RL environment ([#434](https://github.com/NousResearch/hermes-agent/pull/434)) — @jackx707
+- Modal sandbox concurrency limits to avoid deadlocks ([#621](https://github.com/NousResearch/hermes-agent/pull/621)) — @voteblake
+- Hermes-atropos-environments bundled skill ([#815](https://github.com/NousResearch/hermes-agent/pull/815))
+- Local vLLM instance support for evaluation — @dmahan93
+- YC-Bench long-horizon agent benchmark environment
+- OpenThoughts-TBLite evaluation environment and scripts
+
+---
+
+## 📚 Documentation
+
+- Full documentation website (Docusaurus) with 37+ pages
+- Comprehensive platform setup guides for Telegram, Discord, Slack, WhatsApp, Signal, Email
+- AGENTS.md — development guide for AI coding assistants
+- CONTRIBUTING.md ([#117](https://github.com/NousResearch/hermes-agent/pull/117)) — @Bartok9
+- Slash commands reference ([#142](https://github.com/NousResearch/hermes-agent/pull/142)) — @Bartok9
+- Comprehensive AGENTS.md accuracy audit ([#732](https://github.com/NousResearch/hermes-agent/pull/732))
+- Skin/theme system documentation
+- MCP documentation and examples
+- Docs accuracy audit — 35+ corrections
+- Documentation typo fixes ([#825](https://github.com/NousResearch/hermes-agent/pull/825), [#439](https://github.com/NousResearch/hermes-agent/pull/439)) — @JackTheGit
+- CLI config precedence and terminology standardization ([#166](https://github.com/NousResearch/hermes-agent/pull/166), [#167](https://github.com/NousResearch/hermes-agent/pull/167), [#168](https://github.com/NousResearch/hermes-agent/pull/168)) — @Jr-kenny
+- Telegram token regex documentation ([#713](https://github.com/NousResearch/hermes-agent/pull/713)) — @VolodymyrBg
+
+---
+
+## 👥 Contributors
+
+Thank you to the 63 contributors who made this release possible! In just over two weeks, the Hermes Agent community came together to ship an extraordinary amount of work.
+
+### Core
+- **@teknium1** — 43 PRs: Project lead, core architecture, provider router, sessions, skills, CLI, documentation
+
+### Top Community Contributors
+- **@0xbyt4** — 40 PRs: MCP client, Home Assistant, security fixes (symlink, prompt injection, cron), extensive test coverage (6 batches), ascii-art skill, shell noise elimination, skills sync, Telegram formatting, and dozens more
+- **@Farukest** — 16 PRs: Security hardening (path traversal, dangerous command detection, symlink boundary), Windows compatibility (POSIX guards, path handling), WhatsApp fixes, max-iterations retry, gateway fixes
+- **@aydnOktay** — 11 PRs: Atomic writes (process checkpoints, batch runner, skill files), error handling improvements across Telegram, Discord, code execution, transcription, TTS, and skills
+- **@Bartok9** — 9 PRs: CONTRIBUTING.md, slash commands reference, Discord channel topics, think-block stripping, TTS fix, Honcho fix, session count fix, clarify tests
+- **@PercyDikec** — 7 PRs: DeepSeek V3 parser fix, /retry response discard, gateway transcript offset, Codex status/visibility, max-iterations retry, setup wizard fix
+- **@teyrebaz33** — 5 PRs: Skills enable/disable system, quick commands, personality customization, conditional skill activation
+- **@alireza78a** — 5 PRs: Atomic writes (cron, sessions), fd leak prevention, security allowlist, code execution socket cleanup
+- **@shitcoinsherpa** — 3 PRs: Windows support (pywinpty, UTF-8 encoding, auth store lock)
+- **@Himess** — 3 PRs: Cron/HomeAssistant/Daytona fix, Windows drive-letter parsing, .env permissions
+- **@satelerd** — 2 PRs: WhatsApp native media, multi-user session isolation
+- **@rovle** — 1 PR: Daytona cloud sandbox backend (4 commits)
+- **@erosika** — 1 PR: Honcho AI-native memory integration
+- **@dmahan93** — 1 PR: --fuck-it-ship-it flag + RL environment work
+- **@SHL0MS** — 1 PR: ASCII video skill
+
+### All Contributors
+@0xbyt4, @BP602, @Bartok9, @Farukest, @FurkanL0, @Himess, @Indelwin, @JackTheGit, @JoshuaMart, @Jr-kenny, @OutThisLife, @PercyDikec, @SHL0MS, @Sertug17, @VencentSoliman, @VolodymyrBg, @adavyas, @alireza78a, @areu01or00, @aydnOktay, @batuhankocyigit, @bierlingm, @caentzminger, @cesareth, @ch3ronsa, @christomitov, @cutepawss, @deankerr, @dmahan93, @dogiladeveloper, @dragonkhoi, @erosika, @gamedevCloudy, @gizdusum, @grp06, @intertwine, @jackx707, @jdblackstar, @johnh4098, @kaos35, @kshitijk4poor, @leonsgithub, @luisv-1, @manuelschipper, @mehmetkr-31, @memosr, @PeterFile, @rewbs, @rovle, @rsavitt, @satelerd, @spanishflu-est1918, @stablegenius49, @tars90percent, @tekelala, @teknium1, @teyrebaz33, @tripledoublev, @unmodeled-tyler, @voidborne-d, @voteblake, @ygd58
+
+---
+
+**Full Changelog**: [v0.1.0...v2026.3.12](https://github.com/NousResearch/hermes-agent/compare/v0.1.0...v2026.3.12)

RELEASE_v0.3.0.md

@@ -0,0 +1,377 @@
+# Hermes Agent v0.3.0 (v2026.3.17)
+
+**Release Date:** March 17, 2026
+
+> The streaming, plugins, and provider release — unified real-time token delivery, first-class plugin architecture, rebuilt provider system with Vercel AI Gateway, native Anthropic provider, smart approvals, live Chrome CDP browser connect, ACP IDE integration, Honcho memory, voice mode, persistent shell, and 50+ bug fixes across every platform.
+
+---
+
+## ✨ Highlights
+
+- **Unified Streaming Infrastructure** — Real-time token-by-token delivery in CLI and all gateway platforms. Responses stream as they're generated instead of arriving as a block. ([#1538](https://github.com/NousResearch/hermes-agent/pull/1538))
+
+- **First-Class Plugin Architecture** — Drop Python files into `~/.hermes/plugins/` to extend Hermes with custom tools, commands, and hooks. No forking required. ([#1544](https://github.com/NousResearch/hermes-agent/pull/1544), [#1555](https://github.com/NousResearch/hermes-agent/pull/1555))
+
+- **Native Anthropic Provider** — Direct Anthropic API calls with Claude Code credential auto-discovery, OAuth PKCE flows, and native prompt caching. No OpenRouter middleman needed. ([#1097](https://github.com/NousResearch/hermes-agent/pull/1097))
+
+- **Smart Approvals + /stop Command** — Codex-inspired approval system that learns which commands are safe and remembers your preferences. `/stop` kills the current agent run immediately. ([#1543](https://github.com/NousResearch/hermes-agent/pull/1543))
+
+- **Honcho Memory Integration** — Async memory writes, configurable recall modes, session title integration, and multi-user isolation in gateway mode. By @erosika. ([#736](https://github.com/NousResearch/hermes-agent/pull/736))
+
+- **Voice Mode** — Push-to-talk in CLI, voice notes in Telegram/Discord, Discord voice channel support, and local Whisper transcription via faster-whisper. ([#1299](https://github.com/NousResearch/hermes-agent/pull/1299), [#1185](https://github.com/NousResearch/hermes-agent/pull/1185), [#1429](https://github.com/NousResearch/hermes-agent/pull/1429))
+
+- **Concurrent Tool Execution** — Multiple independent tool calls now run in parallel via ThreadPoolExecutor, significantly reducing latency for multi-tool turns. ([#1152](https://github.com/NousResearch/hermes-agent/pull/1152))
+
+- **PII Redaction** — When `privacy.redact_pii` is enabled, personally identifiable information is automatically scrubbed before sending context to LLM providers. ([#1542](https://github.com/NousResearch/hermes-agent/pull/1542))
+
+- **`/browser connect` via CDP** — Attach browser tools to a live Chrome instance through Chrome DevTools Protocol. Debug, inspect, and interact with pages you already have open. ([#1549](https://github.com/NousResearch/hermes-agent/pull/1549))
+
+- **Vercel AI Gateway Provider** — Route Hermes through Vercel's AI Gateway for access to their model catalog and infrastructure. ([#1628](https://github.com/NousResearch/hermes-agent/pull/1628))
+
+- **Centralized Provider Router** — Rebuilt provider system with `call_llm` API, unified `/model` command, auto-detect provider on model switch, and direct endpoint overrides for auxiliary/delegation clients. ([#1003](https://github.com/NousResearch/hermes-agent/pull/1003), [#1506](https://github.com/NousResearch/hermes-agent/pull/1506), [#1375](https://github.com/NousResearch/hermes-agent/pull/1375))
+
+- **ACP Server (IDE Integration)** — VS Code, Zed, and JetBrains can now connect to Hermes as an agent backend, with full slash command support. ([#1254](https://github.com/NousResearch/hermes-agent/pull/1254), [#1532](https://github.com/NousResearch/hermes-agent/pull/1532))
+
+- **Persistent Shell Mode** — Local and SSH terminal backends can maintain shell state across tool calls — cd, env vars, and aliases persist. By @alt-glitch. ([#1067](https://github.com/NousResearch/hermes-agent/pull/1067), [#1483](https://github.com/NousResearch/hermes-agent/pull/1483))
+
+- **Agentic On-Policy Distillation (OPD)** — New RL training environment for distilling agent policies, expanding the Atropos training ecosystem. ([#1149](https://github.com/NousResearch/hermes-agent/pull/1149))
+
+---
+
+## 🏗️ Core Agent & Architecture
+
+### Provider & Model Support
+- **Centralized provider router** with `call_llm` API and unified `/model` command — switch models and providers seamlessly ([#1003](https://github.com/NousResearch/hermes-agent/pull/1003))
+- **Vercel AI Gateway** provider support ([#1628](https://github.com/NousResearch/hermes-agent/pull/1628))
+- **Auto-detect provider** when switching models via `/model` ([#1506](https://github.com/NousResearch/hermes-agent/pull/1506))
+- **Direct endpoint overrides** for auxiliary and delegation clients — point vision/subagent calls at specific endpoints ([#1375](https://github.com/NousResearch/hermes-agent/pull/1375))
+- **Native Anthropic auxiliary vision** — use Claude's native vision API instead of routing through OpenAI-compatible endpoints ([#1377](https://github.com/NousResearch/hermes-agent/pull/1377))
+- Anthropic OAuth flow improvements — auto-run `claude setup-token`, reauthentication, PKCE state persistence, identity fingerprinting ([#1132](https://github.com/NousResearch/hermes-agent/pull/1132), [#1360](https://github.com/NousResearch/hermes-agent/pull/1360), [#1396](https://github.com/NousResearch/hermes-agent/pull/1396), [#1597](https://github.com/NousResearch/hermes-agent/pull/1597))
+- Fix adaptive thinking without `budget_tokens` for Claude 4.6 models — by @ASRagab ([#1128](https://github.com/NousResearch/hermes-agent/pull/1128))
+- Fix Anthropic cache markers through adapter — by @brandtcormorant ([#1216](https://github.com/NousResearch/hermes-agent/pull/1216))
+- Retry Anthropic 429/529 errors and surface details to users — by @0xbyt4 ([#1585](https://github.com/NousResearch/hermes-agent/pull/1585))
+- Fix Anthropic adapter max_tokens, fallback crash, proxy base_url — by @0xbyt4 ([#1121](https://github.com/NousResearch/hermes-agent/pull/1121))
+- Fix DeepSeek V3 parser dropping multiple parallel tool calls — by @mr-emmett-one ([#1365](https://github.com/NousResearch/hermes-agent/pull/1365), [#1300](https://github.com/NousResearch/hermes-agent/pull/1300))
+- Accept unlisted models with warning instead of rejecting ([#1047](https://github.com/NousResearch/hermes-agent/pull/1047), [#1102](https://github.com/NousResearch/hermes-agent/pull/1102))
+- Skip reasoning params for unsupported OpenRouter models ([#1485](https://github.com/NousResearch/hermes-agent/pull/1485))
+- MiniMax Anthropic API compatibility fix ([#1623](https://github.com/NousResearch/hermes-agent/pull/1623))
+- Custom endpoint `/models` verification and `/v1` base URL suggestion ([#1480](https://github.com/NousResearch/hermes-agent/pull/1480))
+- Resolve delegation providers from `custom_providers` config ([#1328](https://github.com/NousResearch/hermes-agent/pull/1328))
+- Kimi model additions and User-Agent fix ([#1039](https://github.com/NousResearch/hermes-agent/pull/1039))
+- Strip `call_id`/`response_item_id` for Mistral compatibility ([#1058](https://github.com/NousResearch/hermes-agent/pull/1058))
+
+### Agent Loop & Conversation
+- **Anthropic Context Editing API** support ([#1147](https://github.com/NousResearch/hermes-agent/pull/1147))
+- Improved context compaction handoff summaries — compressor now preserves more actionable state ([#1273](https://github.com/NousResearch/hermes-agent/pull/1273))
+- Sync session_id after mid-run context compression ([#1160](https://github.com/NousResearch/hermes-agent/pull/1160))
+- Session hygiene threshold tuned to 50% for more proactive compression ([#1096](https://github.com/NousResearch/hermes-agent/pull/1096), [#1161](https://github.com/NousResearch/hermes-agent/pull/1161))
+- Include session ID in system prompt via `--pass-session-id` flag ([#1040](https://github.com/NousResearch/hermes-agent/pull/1040))
+- Prevent closed OpenAI client reuse across retries ([#1391](https://github.com/NousResearch/hermes-agent/pull/1391))
+- Sanitize chat payloads and provider precedence ([#1253](https://github.com/NousResearch/hermes-agent/pull/1253))
+- Handle dict tool call arguments from Codex and local backends ([#1393](https://github.com/NousResearch/hermes-agent/pull/1393), [#1440](https://github.com/NousResearch/hermes-agent/pull/1440))
+
+### Memory & Sessions
+- **Improve memory prioritization** — user preferences and corrections weighted above procedural knowledge ([#1548](https://github.com/NousResearch/hermes-agent/pull/1548))
+- Tighter memory and session recall guidance in system prompts ([#1329](https://github.com/NousResearch/hermes-agent/pull/1329))
+- Persist CLI token counts to session DB for `/insights` ([#1498](https://github.com/NousResearch/hermes-agent/pull/1498))
+- Keep Honcho recall out of the cached system prefix ([#1201](https://github.com/NousResearch/hermes-agent/pull/1201))
+- Correct `seed_ai_identity` to use `session.add_messages()` ([#1475](https://github.com/NousResearch/hermes-agent/pull/1475))
+- Isolate Honcho session routing for multi-user gateway ([#1500](https://github.com/NousResearch/hermes-agent/pull/1500))
+
+---
+
+## 📱 Messaging Platforms (Gateway)
+
+### Gateway Core
+- **System gateway service mode** — run as a system-level systemd service, not just user-level ([#1371](https://github.com/NousResearch/hermes-agent/pull/1371))
+- **Gateway install scope prompts** — choose user vs system scope during setup ([#1374](https://github.com/NousResearch/hermes-agent/pull/1374))
+- **Reasoning hot reload** — change reasoning settings without restarting the gateway ([#1275](https://github.com/NousResearch/hermes-agent/pull/1275))
+- Default group sessions to per-user isolation — no more shared state across users in group chats ([#1495](https://github.com/NousResearch/hermes-agent/pull/1495), [#1417](https://github.com/NousResearch/hermes-agent/pull/1417))
+- Harden gateway restart recovery ([#1310](https://github.com/NousResearch/hermes-agent/pull/1310))
+- Cancel active runs during shutdown ([#1427](https://github.com/NousResearch/hermes-agent/pull/1427))
+- SSL certificate auto-detection for NixOS and non-standard systems ([#1494](https://github.com/NousResearch/hermes-agent/pull/1494))
+- Auto-detect D-Bus session bus for `systemctl --user` on headless servers ([#1601](https://github.com/NousResearch/hermes-agent/pull/1601))
+- Auto-enable systemd linger during gateway install on headless servers ([#1334](https://github.com/NousResearch/hermes-agent/pull/1334))
+- Fall back to module entrypoint when `hermes` is not on PATH ([#1355](https://github.com/NousResearch/hermes-agent/pull/1355))
+- Fix dual gateways on macOS launchd after `hermes update` ([#1567](https://github.com/NousResearch/hermes-agent/pull/1567))
+- Remove recursive ExecStop from systemd units ([#1530](https://github.com/NousResearch/hermes-agent/pull/1530))
+- Prevent logging handler accumulation in gateway mode ([#1251](https://github.com/NousResearch/hermes-agent/pull/1251))
+- Restart on retryable startup failures — by @jplew ([#1517](https://github.com/NousResearch/hermes-agent/pull/1517))
+- Backfill model on gateway sessions after agent runs ([#1306](https://github.com/NousResearch/hermes-agent/pull/1306))
+- PID-based gateway kill and deferred config write ([#1499](https://github.com/NousResearch/hermes-agent/pull/1499))
+
+### Telegram
+- Buffer media groups to prevent self-interruption from photo bursts ([#1341](https://github.com/NousResearch/hermes-agent/pull/1341), [#1422](https://github.com/NousResearch/hermes-agent/pull/1422))
+- Retry on transient TLS failures during connect and send ([#1535](https://github.com/NousResearch/hermes-agent/pull/1535))
+- Harden polling conflict handling ([#1339](https://github.com/NousResearch/hermes-agent/pull/1339))
+- Escape chunk indicators and inline code in MarkdownV2 ([#1478](https://github.com/NousResearch/hermes-agent/pull/1478), [#1626](https://github.com/NousResearch/hermes-agent/pull/1626))
+- Check updater/app state before disconnect ([#1389](https://github.com/NousResearch/hermes-agent/pull/1389))
+
+### Discord
+- `/thread` command with `auto_thread` config and media metadata fixes ([#1178](https://github.com/NousResearch/hermes-agent/pull/1178))
+- Auto-thread on @mention, skip mention text in bot threads ([#1438](https://github.com/NousResearch/hermes-agent/pull/1438))
+- Retry without reply reference for system messages ([#1385](https://github.com/NousResearch/hermes-agent/pull/1385))
+- Preserve native document and video attachment support ([#1392](https://github.com/NousResearch/hermes-agent/pull/1392))
+- Defer discord adapter annotations to avoid optional import crashes ([#1314](https://github.com/NousResearch/hermes-agent/pull/1314))
+
+### Slack
+- Thread handling overhaul — progress messages, responses, and session isolation all respect threads ([#1103](https://github.com/NousResearch/hermes-agent/pull/1103))
+- Formatting, reactions, user resolution, and command improvements ([#1106](https://github.com/NousResearch/hermes-agent/pull/1106))
+- Fix MAX_MESSAGE_LENGTH 3900 → 39000 ([#1117](https://github.com/NousResearch/hermes-agent/pull/1117))
+- File upload fallback preserves thread context — by @0xbyt4 ([#1122](https://github.com/NousResearch/hermes-agent/pull/1122))
+- Improve setup guidance ([#1387](https://github.com/NousResearch/hermes-agent/pull/1387))
+
+### Email
+- Fix IMAP UID tracking and SMTP TLS verification ([#1305](https://github.com/NousResearch/hermes-agent/pull/1305))
+- Add `skip_attachments` option via config.yaml ([#1536](https://github.com/NousResearch/hermes-agent/pull/1536))
+
+### Home Assistant
+- Event filtering closed by default ([#1169](https://github.com/NousResearch/hermes-agent/pull/1169))
+
+---
+
+## 🖥️ CLI & User Experience
+
+### Interactive CLI
+- **Persistent CLI status bar** — always-visible model, provider, and token counts ([#1522](https://github.com/NousResearch/hermes-agent/pull/1522))
+- **File path autocomplete** in the input prompt ([#1545](https://github.com/NousResearch/hermes-agent/pull/1545))
+- **`/plan` command** — generate implementation plans from specs ([#1372](https://github.com/NousResearch/hermes-agent/pull/1372), [#1381](https://github.com/NousResearch/hermes-agent/pull/1381))
+- **Major `/rollback` improvements** — richer checkpoint history, clearer UX ([#1505](https://github.com/NousResearch/hermes-agent/pull/1505))
+- **Preload CLI skills on launch** — skills are ready before the first prompt ([#1359](https://github.com/NousResearch/hermes-agent/pull/1359))
+- **Centralized slash command registry** — all commands defined once, consumed everywhere ([#1603](https://github.com/NousResearch/hermes-agent/pull/1603))
+- `/bg` alias for `/background` ([#1590](https://github.com/NousResearch/hermes-agent/pull/1590))
+- Prefix matching for slash commands — `/mod` resolves to `/model` ([#1320](https://github.com/NousResearch/hermes-agent/pull/1320))
+- `/new`, `/reset`, `/clear` now start genuinely fresh sessions ([#1237](https://github.com/NousResearch/hermes-agent/pull/1237))
+- Accept session ID prefixes for session actions ([#1425](https://github.com/NousResearch/hermes-agent/pull/1425))
+- TUI prompt and accent output now respect active skin ([#1282](https://github.com/NousResearch/hermes-agent/pull/1282))
+- Centralize tool emoji metadata in registry + skin integration ([#1484](https://github.com/NousResearch/hermes-agent/pull/1484))
+- "View full command" option added to dangerous command approval — by @teknium1 based on design by community ([#887](https://github.com/NousResearch/hermes-agent/pull/887))
+- Non-blocking startup update check and banner deduplication ([#1386](https://github.com/NousResearch/hermes-agent/pull/1386))
+- `/reasoning` command output ordering and inline think extraction fixes ([#1031](https://github.com/NousResearch/hermes-agent/pull/1031))
+- Verbose mode shows full untruncated output ([#1472](https://github.com/NousResearch/hermes-agent/pull/1472))
+- Fix `/status` to report live state and tokens ([#1476](https://github.com/NousResearch/hermes-agent/pull/1476))
+- Seed a default global SOUL.md ([#1311](https://github.com/NousResearch/hermes-agent/pull/1311))
+
+### Setup & Configuration
+- **OpenClaw migration** during first-time setup — by @kshitijk4poor ([#981](https://github.com/NousResearch/hermes-agent/pull/981))
+- `hermes claw migrate` command + migration docs ([#1059](https://github.com/NousResearch/hermes-agent/pull/1059))
+- Smart vision setup that respects the user's chosen provider ([#1323](https://github.com/NousResearch/hermes-agent/pull/1323))
+- Handle headless setup flows end-to-end ([#1274](https://github.com/NousResearch/hermes-agent/pull/1274))
+- Prefer curses over `simple_term_menu` in setup.py ([#1487](https://github.com/NousResearch/hermes-agent/pull/1487))
+- Show effective model and provider in `/status` ([#1284](https://github.com/NousResearch/hermes-agent/pull/1284))
+- Config set examples use placeholder syntax ([#1322](https://github.com/NousResearch/hermes-agent/pull/1322))
+- Reload .env over stale shell overrides ([#1434](https://github.com/NousResearch/hermes-agent/pull/1434))
+- Fix is_coding_plan NameError crash — by @0xbyt4 ([#1123](https://github.com/NousResearch/hermes-agent/pull/1123))
+- Add missing packages to setuptools config — by @alt-glitch ([#912](https://github.com/NousResearch/hermes-agent/pull/912))
+- Installer: clarify why sudo is needed at every prompt ([#1602](https://github.com/NousResearch/hermes-agent/pull/1602))
+
+---
+
+## 🔧 Tool System
+
+### Terminal & Execution
+- **Persistent shell mode** for local and SSH backends — maintain shell state across tool calls — by @alt-glitch ([#1067](https://github.com/NousResearch/hermes-agent/pull/1067), [#1483](https://github.com/NousResearch/hermes-agent/pull/1483))
+- **Tirith pre-exec command scanning** — security layer that analyzes commands before execution ([#1256](https://github.com/NousResearch/hermes-agent/pull/1256))
+- Strip Hermes provider env vars from all subprocess environments ([#1157](https://github.com/NousResearch/hermes-agent/pull/1157), [#1172](https://github.com/NousResearch/hermes-agent/pull/1172), [#1399](https://github.com/NousResearch/hermes-agent/pull/1399), [#1419](https://github.com/NousResearch/hermes-agent/pull/1419)) — initial fix by @eren-karakus0
+- SSH preflight check ([#1486](https://github.com/NousResearch/hermes-agent/pull/1486))
+- Docker backend: make cwd workspace mount explicit opt-in ([#1534](https://github.com/NousResearch/hermes-agent/pull/1534))
+- Add project root to PYTHONPATH in execute_code sandbox ([#1383](https://github.com/NousResearch/hermes-agent/pull/1383))
+- Eliminate execute_code progress spam on gateway platforms ([#1098](https://github.com/NousResearch/hermes-agent/pull/1098))
+- Clearer docker backend preflight errors ([#1276](https://github.com/NousResearch/hermes-agent/pull/1276))
+
+### Browser
+- **`/browser connect`** — attach browser tools to a live Chrome instance via CDP ([#1549](https://github.com/NousResearch/hermes-agent/pull/1549))
+- Improve browser cleanup, local browser PATH setup, and screenshot recovery ([#1333](https://github.com/NousResearch/hermes-agent/pull/1333))
+
+### MCP
+- **Selective tool loading** with utility policies — filter which MCP tools are available ([#1302](https://github.com/NousResearch/hermes-agent/pull/1302))
+- Auto-reload MCP tools when `mcp_servers` config changes without restart ([#1474](https://github.com/NousResearch/hermes-agent/pull/1474))
+- Resolve npx stdio connection failures ([#1291](https://github.com/NousResearch/hermes-agent/pull/1291))
+- Preserve MCP toolsets when saving platform tool config ([#1421](https://github.com/NousResearch/hermes-agent/pull/1421))
+
+### Vision
+- Unify vision backend gating ([#1367](https://github.com/NousResearch/hermes-agent/pull/1367))
+- Surface actual error reason instead of generic message ([#1338](https://github.com/NousResearch/hermes-agent/pull/1338))
+- Make Claude image handling work end-to-end ([#1408](https://github.com/NousResearch/hermes-agent/pull/1408))
+
+### Cron
+- **Compress cron management into one tool** — single `cronjob` tool replaces multiple commands ([#1343](https://github.com/NousResearch/hermes-agent/pull/1343))
+- Suppress duplicate cron sends to auto-delivery targets ([#1357](https://github.com/NousResearch/hermes-agent/pull/1357))
+- Persist cron sessions to SQLite ([#1255](https://github.com/NousResearch/hermes-agent/pull/1255))
+- Per-job runtime overrides (provider, model, base_url) ([#1398](https://github.com/NousResearch/hermes-agent/pull/1398))
+- Atomic write in `save_job_output` to prevent data loss on crash ([#1173](https://github.com/NousResearch/hermes-agent/pull/1173))
+- Preserve thread context for `deliver=origin` ([#1437](https://github.com/NousResearch/hermes-agent/pull/1437))
+
+### Patch Tool
+- Avoid corrupting pipe chars in V4A patch apply ([#1286](https://github.com/NousResearch/hermes-agent/pull/1286))
+- Permissive `block_anchor` thresholds and unicode normalization ([#1539](https://github.com/NousResearch/hermes-agent/pull/1539))
+
+### Delegation
+- Add observability metadata to subagent results (model, tokens, duration, tool trace) ([#1175](https://github.com/NousResearch/hermes-agent/pull/1175))
+
+---
+
+## 🧩 Skills Ecosystem
+
+### Skills System
+- **Integrate skills.sh** as a hub source alongside ClawHub ([#1303](https://github.com/NousResearch/hermes-agent/pull/1303))
+- Secure skill env setup on load ([#1153](https://github.com/NousResearch/hermes-agent/pull/1153))
+- Honor policy table for dangerous verdicts ([#1330](https://github.com/NousResearch/hermes-agent/pull/1330))
+- Harden ClawHub skill search exact matches ([#1400](https://github.com/NousResearch/hermes-agent/pull/1400))
+- Fix ClawHub skill install — use `/download` ZIP endpoint ([#1060](https://github.com/NousResearch/hermes-agent/pull/1060))
+- Avoid mislabeling local skills as builtin — by @arceus77-7 ([#862](https://github.com/NousResearch/hermes-agent/pull/862))
+
+### New Skills
+- **Linear** project management ([#1230](https://github.com/NousResearch/hermes-agent/pull/1230))
+- **X/Twitter** via x-cli ([#1285](https://github.com/NousResearch/hermes-agent/pull/1285))
+- **Telephony** — Twilio, SMS, and AI calls ([#1289](https://github.com/NousResearch/hermes-agent/pull/1289))
+- **1Password** — by @arceus77-7 ([#883](https://github.com/NousResearch/hermes-agent/pull/883), [#1179](https://github.com/NousResearch/hermes-agent/pull/1179))
+- **NeuroSkill BCI** integration ([#1135](https://github.com/NousResearch/hermes-agent/pull/1135))
+- **Blender MCP** for 3D modeling ([#1531](https://github.com/NousResearch/hermes-agent/pull/1531))
+- **OSS Security Forensics** ([#1482](https://github.com/NousResearch/hermes-agent/pull/1482))
+- **Parallel CLI** research skill ([#1301](https://github.com/NousResearch/hermes-agent/pull/1301))
+- **OpenCode** CLI skill ([#1174](https://github.com/NousResearch/hermes-agent/pull/1174))
+- **ASCII Video** skill refactored — by @SHL0MS ([#1213](https://github.com/NousResearch/hermes-agent/pull/1213), [#1598](https://github.com/NousResearch/hermes-agent/pull/1598))
+
+---
+
+## 🎙️ Voice Mode
+
+- Voice mode foundation — push-to-talk CLI, Telegram/Discord voice notes ([#1299](https://github.com/NousResearch/hermes-agent/pull/1299))
+- Free local Whisper transcription via faster-whisper ([#1185](https://github.com/NousResearch/hermes-agent/pull/1185))
+- Discord voice channel reliability fixes ([#1429](https://github.com/NousResearch/hermes-agent/pull/1429))
+- Restore local STT fallback for gateway voice notes ([#1490](https://github.com/NousResearch/hermes-agent/pull/1490))
+- Honor `stt.enabled: false` across gateway transcription ([#1394](https://github.com/NousResearch/hermes-agent/pull/1394))
+- Fix bogus incapability message on Telegram voice notes (Issue [#1033](https://github.com/NousResearch/hermes-agent/issues/1033))
+
+---
+
+## 🔌 ACP (IDE Integration)
+
+- Restore ACP server implementation ([#1254](https://github.com/NousResearch/hermes-agent/pull/1254))
+- Support slash commands in ACP adapter ([#1532](https://github.com/NousResearch/hermes-agent/pull/1532))
+
+---
+
+## 🧪 RL Training
+
+- **Agentic On-Policy Distillation (OPD)** environment — new RL training environment for agent policy distillation ([#1149](https://github.com/NousResearch/hermes-agent/pull/1149))
+- Make tinker-atropos RL training fully optional ([#1062](https://github.com/NousResearch/hermes-agent/pull/1062))
+
+---
+
+## 🔒 Security & Reliability
+
+### Security Hardening
+- **Tirith pre-exec command scanning** — static analysis of terminal commands before execution ([#1256](https://github.com/NousResearch/hermes-agent/pull/1256))
+- **PII redaction** when `privacy.redact_pii` is enabled ([#1542](https://github.com/NousResearch/hermes-agent/pull/1542))
+- Strip Hermes provider/gateway/tool env vars from all subprocess environments ([#1157](https://github.com/NousResearch/hermes-agent/pull/1157), [#1172](https://github.com/NousResearch/hermes-agent/pull/1172), [#1399](https://github.com/NousResearch/hermes-agent/pull/1399), [#1419](https://github.com/NousResearch/hermes-agent/pull/1419))
+- Docker cwd workspace mount now explicit opt-in — never auto-mount host directories ([#1534](https://github.com/NousResearch/hermes-agent/pull/1534))
+- Escape parens and braces in fork bomb regex pattern ([#1397](https://github.com/NousResearch/hermes-agent/pull/1397))
+- Harden `.worktreeinclude` path containment ([#1388](https://github.com/NousResearch/hermes-agent/pull/1388))
+- Use description as `pattern_key` to prevent approval collisions ([#1395](https://github.com/NousResearch/hermes-agent/pull/1395))
+
+### Reliability
+- Guard init-time stdio writes ([#1271](https://github.com/NousResearch/hermes-agent/pull/1271))
+- Session log writes reuse shared atomic JSON helper ([#1280](https://github.com/NousResearch/hermes-agent/pull/1280))
+- Atomic temp cleanup protected on interrupts ([#1401](https://github.com/NousResearch/hermes-agent/pull/1401))
+
+---
+
+## 🐛 Notable Bug Fixes
+
+- **`/status` always showing 0 tokens** — now reports live state (Issue [#1465](https://github.com/NousResearch/hermes-agent/issues/1465), [#1476](https://github.com/NousResearch/hermes-agent/pull/1476))
+- **Custom model endpoints not working** — restored config-saved endpoint resolution (Issue [#1460](https://github.com/NousResearch/hermes-agent/issues/1460), [#1373](https://github.com/NousResearch/hermes-agent/pull/1373))
+- **MCP tools not visible until restart** — auto-reload on config change (Issue [#1036](https://github.com/NousResearch/hermes-agent/issues/1036), [#1474](https://github.com/NousResearch/hermes-agent/pull/1474))
+- **`hermes tools` removing MCP tools** — preserve MCP toolsets when saving (Issue [#1247](https://github.com/NousResearch/hermes-agent/issues/1247), [#1421](https://github.com/NousResearch/hermes-agent/pull/1421))
+- **Terminal subprocesses inheriting `OPENAI_BASE_URL`** breaking external tools (Issue [#1002](https://github.com/NousResearch/hermes-agent/issues/1002), [#1399](https://github.com/NousResearch/hermes-agent/pull/1399))
+- **Background process lost on gateway restart** — improved recovery (Issue [#1144](https://github.com/NousResearch/hermes-agent/issues/1144))
+- **Cron jobs not persisting state** — now stored in SQLite (Issue [#1416](https://github.com/NousResearch/hermes-agent/issues/1416), [#1255](https://github.com/NousResearch/hermes-agent/pull/1255))
+- **Cronjob `deliver: origin` not preserving thread context** (Issue [#1219](https://github.com/NousResearch/hermes-agent/issues/1219), [#1437](https://github.com/NousResearch/hermes-agent/pull/1437))
+- **Gateway systemd service failing to auto-restart** when browser processes orphaned (Issue [#1617](https://github.com/NousResearch/hermes-agent/issues/1617))
+- **`/background` completion report cut off in Telegram** (Issue [#1443](https://github.com/NousResearch/hermes-agent/issues/1443))
+- **Model switching not taking effect** (Issue [#1244](https://github.com/NousResearch/hermes-agent/issues/1244), [#1183](https://github.com/NousResearch/hermes-agent/pull/1183))
+- **`hermes doctor` reporting cronjob as unavailable** (Issue [#878](https://github.com/NousResearch/hermes-agent/issues/878), [#1180](https://github.com/NousResearch/hermes-agent/pull/1180))
+- **WhatsApp bridge messages not received** from mobile (Issue [#1142](https://github.com/NousResearch/hermes-agent/issues/1142))
+- **Setup wizard hanging on headless SSH** (Issue [#905](https://github.com/NousResearch/hermes-agent/issues/905), [#1274](https://github.com/NousResearch/hermes-agent/pull/1274))
+- **Log handler accumulation** degrading gateway performance (Issue [#990](https://github.com/NousResearch/hermes-agent/issues/990), [#1251](https://github.com/NousResearch/hermes-agent/pull/1251))
+- **Gateway NULL model in DB** (Issue [#987](https://github.com/NousResearch/hermes-agent/issues/987), [#1306](https://github.com/NousResearch/hermes-agent/pull/1306))
+- **Strict endpoints rejecting replayed tool_calls** (Issue [#893](https://github.com/NousResearch/hermes-agent/issues/893))
+- **Remaining hardcoded `~/.hermes` paths** — all now respect `HERMES_HOME` (Issue [#892](https://github.com/NousResearch/hermes-agent/issues/892), [#1233](https://github.com/NousResearch/hermes-agent/pull/1233))
+- **Delegate tool not working with custom inference providers** (Issue [#1011](https://github.com/NousResearch/hermes-agent/issues/1011), [#1328](https://github.com/NousResearch/hermes-agent/pull/1328))
+- **Skills Guard blocking official skills** (Issue [#1006](https://github.com/NousResearch/hermes-agent/issues/1006), [#1330](https://github.com/NousResearch/hermes-agent/pull/1330))
+- **Setup writing provider before model selection** (Issue [#1182](https://github.com/NousResearch/hermes-agent/issues/1182))
+- **`GatewayConfig.get()` AttributeError** crashing all message handling (Issue [#1158](https://github.com/NousResearch/hermes-agent/issues/1158), [#1287](https://github.com/NousResearch/hermes-agent/pull/1287))
+- **`/update` hard-failing with "command not found"** (Issue [#1049](https://github.com/NousResearch/hermes-agent/issues/1049))
+- **Image analysis failing silently** (Issue [#1034](https://github.com/NousResearch/hermes-agent/issues/1034), [#1338](https://github.com/NousResearch/hermes-agent/pull/1338))
+- **API `BadRequestError` from `'dict'` object has no attribute `'strip'`** (Issue [#1071](https://github.com/NousResearch/hermes-agent/issues/1071))
+- **Slash commands requiring exact full name** — now uses prefix matching (Issue [#928](https://github.com/NousResearch/hermes-agent/issues/928), [#1320](https://github.com/NousResearch/hermes-agent/pull/1320))
+- **Gateway stops responding when terminal is closed on headless** (Issue [#1005](https://github.com/NousResearch/hermes-agent/issues/1005))
+
+---
+
+## 🧪 Testing
+
+- Cover empty cached Anthropic tool-call turns ([#1222](https://github.com/NousResearch/hermes-agent/pull/1222))
+- Fix stale CI assumptions in parser and quick-command coverage ([#1236](https://github.com/NousResearch/hermes-agent/pull/1236))
+- Fix gateway async tests without implicit event loop ([#1278](https://github.com/NousResearch/hermes-agent/pull/1278))
+- Make gateway async tests xdist-safe ([#1281](https://github.com/NousResearch/hermes-agent/pull/1281))
+- Cross-timezone naive timestamp regression for cron ([#1319](https://github.com/NousResearch/hermes-agent/pull/1319))
+- Isolate codex provider tests from local env ([#1335](https://github.com/NousResearch/hermes-agent/pull/1335))
+- Lock retry replacement semantics ([#1379](https://github.com/NousResearch/hermes-agent/pull/1379))
+- Improve error logging in session search tool — by @aydnOktay ([#1533](https://github.com/NousResearch/hermes-agent/pull/1533))
+
+---
+
+## 📚 Documentation
+
+- Comprehensive SOUL.md guide ([#1315](https://github.com/NousResearch/hermes-agent/pull/1315))
+- Voice mode documentation ([#1316](https://github.com/NousResearch/hermes-agent/pull/1316), [#1362](https://github.com/NousResearch/hermes-agent/pull/1362))
+- Provider contribution guide ([#1361](https://github.com/NousResearch/hermes-agent/pull/1361))
+- ACP and internal systems implementation guides ([#1259](https://github.com/NousResearch/hermes-agent/pull/1259))
+- Expand Docusaurus coverage across CLI, tools, skills, and skins ([#1232](https://github.com/NousResearch/hermes-agent/pull/1232))
+- Terminal backend and Windows troubleshooting ([#1297](https://github.com/NousResearch/hermes-agent/pull/1297))
+- Skills hub reference section ([#1317](https://github.com/NousResearch/hermes-agent/pull/1317))
+- Checkpoint, /rollback, and git worktrees guide ([#1493](https://github.com/NousResearch/hermes-agent/pull/1493), [#1524](https://github.com/NousResearch/hermes-agent/pull/1524))
+- CLI status bar and /usage reference ([#1523](https://github.com/NousResearch/hermes-agent/pull/1523))
+- Fallback providers + /background command docs ([#1430](https://github.com/NousResearch/hermes-agent/pull/1430))
+- Gateway service scopes docs ([#1378](https://github.com/NousResearch/hermes-agent/pull/1378))
+- Slack thread reply behavior docs ([#1407](https://github.com/NousResearch/hermes-agent/pull/1407))
+- Redesigned landing page with Nous blue palette — by @austinpickett ([#974](https://github.com/NousResearch/hermes-agent/pull/974))
+- Fix several documentation typos — by @JackTheGit ([#953](https://github.com/NousResearch/hermes-agent/pull/953))
+- Stabilize website diagrams ([#1405](https://github.com/NousResearch/hermes-agent/pull/1405))
+- CLI vs messaging quick reference in README ([#1491](https://github.com/NousResearch/hermes-agent/pull/1491))
+- Add search to Docusaurus ([#1053](https://github.com/NousResearch/hermes-agent/pull/1053))
+- Home Assistant integration docs ([#1170](https://github.com/NousResearch/hermes-agent/pull/1170))
+
+---
+
+## 👥 Contributors
+
+### Core
+- **@teknium1** — 220+ PRs spanning every area of the codebase
+
+### Top Community Contributors
+
+- **@0xbyt4** (4 PRs) — Anthropic adapter fixes (max_tokens, fallback crash, 429/529 retry), Slack file upload thread context, setup NameError fix
+- **@erosika** (1 PR) — Honcho memory integration: async writes, memory modes, session title integration
+- **@SHL0MS** (2 PRs) — ASCII video skill design patterns and refactoring
+- **@alt-glitch** (2 PRs) — Persistent shell mode for local/SSH backends, setuptools packaging fix
+- **@arceus77-7** (2 PRs) — 1Password skill, fix skills list mislabeling
+- **@kshitijk4poor** (1 PR) — OpenClaw migration during setup wizard
+- **@ASRagab** (1 PR) — Fix adaptive thinking for Claude 4.6 models
+- **@eren-karakus0** (1 PR) — Strip Hermes provider env vars from subprocess environment
+- **@mr-emmett-one** (1 PR) — Fix DeepSeek V3 parser multi-tool call support
+- **@jplew** (1 PR) — Gateway restart on retryable startup failures
+- **@brandtcormorant** (1 PR) — Fix Anthropic cache control for empty text blocks
+- **@aydnOktay** (1 PR) — Improve error logging in session search tool
+- **@austinpickett** (1 PR) — Landing page redesign with Nous blue palette
+- **@JackTheGit** (1 PR) — Documentation typo fixes
+
+### All Contributors
+
+@0xbyt4, @alt-glitch, @arceus77-7, @ASRagab, @austinpickett, @aydnOktay, @brandtcormorant, @eren-karakus0, @erosika, @JackTheGit, @jplew, @kshitijk4poor, @mr-emmett-one, @SHL0MS, @teknium1
+
+---
+
+**Full Changelog**: [v2026.3.12...v2026.3.17](https://github.com/NousResearch/hermes-agent/compare/v2026.3.12...v2026.3.17)

RELEASE_v0.4.0.md

@@ -0,0 +1,400 @@
+# Hermes Agent v0.4.0 (v2026.3.23)
+
+**Release Date:** March 23, 2026
+
+> The platform expansion release — OpenAI-compatible API server, 6 new messaging adapters, 4 new inference providers, MCP server management with OAuth 2.1, @ context references, gateway prompt caching, streaming enabled by default, and a sweeping reliability pass with 200+ bug fixes.
+
+---
+
+## ✨ Highlights
+
+- **OpenAI-compatible API server** — Expose Hermes as an `/v1/chat/completions` endpoint with a new `/api/jobs` REST API for cron job management, hardened with input limits, field whitelists, SQLite-backed response persistence, and CORS origin protection ([#1756](https://github.com/NousResearch/hermes-agent/pull/1756), [#2450](https://github.com/NousResearch/hermes-agent/pull/2450), [#2456](https://github.com/NousResearch/hermes-agent/pull/2456), [#2451](https://github.com/NousResearch/hermes-agent/pull/2451), [#2472](https://github.com/NousResearch/hermes-agent/pull/2472))
+
+- **6 new messaging platform adapters** — Signal, DingTalk, SMS (Twilio), Mattermost, Matrix, and Webhook adapters join Telegram, Discord, and WhatsApp. Gateway auto-reconnects failed platforms with exponential backoff ([#2206](https://github.com/NousResearch/hermes-agent/pull/2206), [#1685](https://github.com/NousResearch/hermes-agent/pull/1685), [#1688](https://github.com/NousResearch/hermes-agent/pull/1688), [#1683](https://github.com/NousResearch/hermes-agent/pull/1683), [#2166](https://github.com/NousResearch/hermes-agent/pull/2166), [#2584](https://github.com/NousResearch/hermes-agent/pull/2584))
+
+- **@ context references** — Claude Code-style `@file` and `@url` context injection with tab completions in the CLI ([#2343](https://github.com/NousResearch/hermes-agent/pull/2343), [#2482](https://github.com/NousResearch/hermes-agent/pull/2482))
+
+- **4 new inference providers** — GitHub Copilot (OAuth + token validation), Alibaba Cloud / DashScope, Kilo Code, and OpenCode Zen/Go ([#1924](https://github.com/NousResearch/hermes-agent/pull/1924), [#1879](https://github.com/NousResearch/hermes-agent/pull/1879) by @mchzimm, [#1673](https://github.com/NousResearch/hermes-agent/pull/1673), [#1666](https://github.com/NousResearch/hermes-agent/pull/1666), [#1650](https://github.com/NousResearch/hermes-agent/pull/1650))
+
+- **MCP server management CLI** — `hermes mcp` commands for installing, configuring, and authenticating MCP servers with full OAuth 2.1 PKCE flow ([#2465](https://github.com/NousResearch/hermes-agent/pull/2465))
+
+- **Gateway prompt caching** — Cache AIAgent instances per session, preserving Anthropic prompt cache across turns for dramatic cost reduction on long conversations ([#2282](https://github.com/NousResearch/hermes-agent/pull/2282), [#2284](https://github.com/NousResearch/hermes-agent/pull/2284), [#2361](https://github.com/NousResearch/hermes-agent/pull/2361))
+
+- **Context compression overhaul** — Structured summaries with iterative updates, token-budget tail protection, configurable summary endpoint, and fallback model support ([#2323](https://github.com/NousResearch/hermes-agent/pull/2323), [#1727](https://github.com/NousResearch/hermes-agent/pull/1727), [#2224](https://github.com/NousResearch/hermes-agent/pull/2224))
+
+- **Streaming enabled by default** — CLI streaming on by default with proper spinner/tool progress display during streaming mode, plus extensive linebreak and concatenation fixes ([#2340](https://github.com/NousResearch/hermes-agent/pull/2340), [#2161](https://github.com/NousResearch/hermes-agent/pull/2161), [#2258](https://github.com/NousResearch/hermes-agent/pull/2258))
+
+---
+
+## 🖥️ CLI & User Experience
+
+### New Commands & Interactions
+- **@ context completions** — Tab-completable `@file`/`@url` references that inject file content or web pages into the conversation ([#2482](https://github.com/NousResearch/hermes-agent/pull/2482), [#2343](https://github.com/NousResearch/hermes-agent/pull/2343))
+- **`/statusbar`** — Toggle a persistent config bar showing model + provider info in the prompt ([#2240](https://github.com/NousResearch/hermes-agent/pull/2240), [#1917](https://github.com/NousResearch/hermes-agent/pull/1917))
+- **`/queue`** — Queue prompts for the agent without interrupting the current run ([#2191](https://github.com/NousResearch/hermes-agent/pull/2191), [#2469](https://github.com/NousResearch/hermes-agent/pull/2469))
+- **`/permission`** — Switch approval mode dynamically during a session ([#2207](https://github.com/NousResearch/hermes-agent/pull/2207))
+- **`/browser`** — Interactive browser sessions from the CLI ([#2273](https://github.com/NousResearch/hermes-agent/pull/2273), [#1814](https://github.com/NousResearch/hermes-agent/pull/1814))
+- **`/cost`** — Live pricing and usage tracking in gateway mode ([#2180](https://github.com/NousResearch/hermes-agent/pull/2180))
+- **`/approve` and `/deny`** — Replaced bare text approval in gateway with explicit commands ([#2002](https://github.com/NousResearch/hermes-agent/pull/2002))
+
+### Streaming & Display
+- Streaming enabled by default in CLI ([#2340](https://github.com/NousResearch/hermes-agent/pull/2340))
+- Show spinners and tool progress during streaming mode ([#2161](https://github.com/NousResearch/hermes-agent/pull/2161))
+- Show reasoning/thinking blocks when `show_reasoning` enabled ([#2118](https://github.com/NousResearch/hermes-agent/pull/2118))
+- Context pressure warnings for CLI and gateway ([#2159](https://github.com/NousResearch/hermes-agent/pull/2159))
+- Fix: streaming chunks concatenated without whitespace ([#2258](https://github.com/NousResearch/hermes-agent/pull/2258))
+- Fix: iteration boundary linebreak prevents stream concatenation ([#2413](https://github.com/NousResearch/hermes-agent/pull/2413))
+- Fix: defer streaming linebreak to prevent blank line stacking ([#2473](https://github.com/NousResearch/hermes-agent/pull/2473))
+- Fix: suppress spinner animation in non-TTY environments ([#2216](https://github.com/NousResearch/hermes-agent/pull/2216))
+- Fix: display provider and endpoint in API error messages ([#2266](https://github.com/NousResearch/hermes-agent/pull/2266))
+- Fix: resolve garbled ANSI escape codes in status printouts ([#2448](https://github.com/NousResearch/hermes-agent/pull/2448))
+- Fix: update gold ANSI color to true-color format ([#2246](https://github.com/NousResearch/hermes-agent/pull/2246))
+- Fix: normalize toolset labels and use skin colors in banner ([#1912](https://github.com/NousResearch/hermes-agent/pull/1912))
+
+### CLI Polish
+- Fix: prevent 'Press ENTER to continue...' on exit ([#2555](https://github.com/NousResearch/hermes-agent/pull/2555))
+- Fix: flush stdout during agent loop to prevent macOS display freeze ([#1654](https://github.com/NousResearch/hermes-agent/pull/1654))
+- Fix: show human-readable error when `hermes setup` hits permissions error ([#2196](https://github.com/NousResearch/hermes-agent/pull/2196))
+- Fix: `/stop` command crash + UnboundLocalError in streaming media delivery ([#2463](https://github.com/NousResearch/hermes-agent/pull/2463))
+- Fix: allow custom/local endpoints without API key ([#2556](https://github.com/NousResearch/hermes-agent/pull/2556))
+- Fix: Kitty keyboard protocol Shift+Enter for Ghostty/WezTerm (attempted + reverted due to prompt_toolkit crash) ([#2345](https://github.com/NousResearch/hermes-agent/pull/2345), [#2349](https://github.com/NousResearch/hermes-agent/pull/2349))
+
+### Configuration
+- **`${ENV_VAR}` substitution** in config.yaml ([#2684](https://github.com/NousResearch/hermes-agent/pull/2684))
+- **Real-time config reload** — config.yaml changes apply without restart ([#2210](https://github.com/NousResearch/hermes-agent/pull/2210))
+- **`custom_models.yaml`** for user-managed model additions ([#2214](https://github.com/NousResearch/hermes-agent/pull/2214))
+- **Priority-based context file selection** + CLAUDE.md support ([#2301](https://github.com/NousResearch/hermes-agent/pull/2301))
+- **Merge nested YAML sections** instead of replacing on config update ([#2213](https://github.com/NousResearch/hermes-agent/pull/2213))
+- Fix: config.yaml provider key overrides env var silently ([#2272](https://github.com/NousResearch/hermes-agent/pull/2272))
+- Fix: log warning instead of silently swallowing config.yaml errors ([#2683](https://github.com/NousResearch/hermes-agent/pull/2683))
+- Fix: disabled toolsets re-enable themselves after `hermes tools` ([#2268](https://github.com/NousResearch/hermes-agent/pull/2268))
+- Fix: platform default toolsets silently override tool deselection ([#2624](https://github.com/NousResearch/hermes-agent/pull/2624))
+- Fix: honor bare YAML `approvals.mode: off` ([#2620](https://github.com/NousResearch/hermes-agent/pull/2620))
+- Fix: `hermes update` use `.[all]` extras with fallback ([#1728](https://github.com/NousResearch/hermes-agent/pull/1728))
+- Fix: `hermes update` prompt before resetting working tree on stash conflicts ([#2390](https://github.com/NousResearch/hermes-agent/pull/2390))
+- Fix: use git pull --rebase in update/install to avoid divergent branch error ([#2274](https://github.com/NousResearch/hermes-agent/pull/2274))
+- Fix: add zprofile fallback and create zshrc on fresh macOS installs ([#2320](https://github.com/NousResearch/hermes-agent/pull/2320))
+- Fix: remove `ANTHROPIC_BASE_URL` env var to avoid collisions ([#1675](https://github.com/NousResearch/hermes-agent/pull/1675))
+- Fix: don't ask IMAP password if already in keyring or env ([#2212](https://github.com/NousResearch/hermes-agent/pull/2212))
+- Fix: OpenCode Zen/Go show OpenRouter models instead of their own ([#2277](https://github.com/NousResearch/hermes-agent/pull/2277))
+
+---
+
+## 🏗️ Core Agent & Architecture
+
+### New Providers
+- **GitHub Copilot** — Full OAuth auth, API routing, token validation, and 400k context. ([#1924](https://github.com/NousResearch/hermes-agent/pull/1924), [#1896](https://github.com/NousResearch/hermes-agent/pull/1896), [#1879](https://github.com/NousResearch/hermes-agent/pull/1879) by @mchzimm, [#2507](https://github.com/NousResearch/hermes-agent/pull/2507))
+- **Alibaba Cloud / DashScope** — Full integration with DashScope v1 runtime, model dot preservation, and 401 auth fixes ([#1673](https://github.com/NousResearch/hermes-agent/pull/1673), [#2332](https://github.com/NousResearch/hermes-agent/pull/2332), [#2459](https://github.com/NousResearch/hermes-agent/pull/2459))
+- **Kilo Code** — First-class inference provider ([#1666](https://github.com/NousResearch/hermes-agent/pull/1666))
+- **OpenCode Zen and OpenCode Go** — New provider backends ([#1650](https://github.com/NousResearch/hermes-agent/pull/1650), [#2393](https://github.com/NousResearch/hermes-agent/pull/2393) by @0xbyt4)
+- **NeuTTS** — Local TTS provider backend with built-in setup flow, replacing the old optional skill ([#1657](https://github.com/NousResearch/hermes-agent/pull/1657), [#1664](https://github.com/NousResearch/hermes-agent/pull/1664))
+
+### Provider Improvements
+- **Eager fallback** to backup model on rate-limit errors ([#1730](https://github.com/NousResearch/hermes-agent/pull/1730))
+- **Endpoint metadata** for custom model context and pricing; query local servers for actual context window size ([#1906](https://github.com/NousResearch/hermes-agent/pull/1906), [#2091](https://github.com/NousResearch/hermes-agent/pull/2091) by @dusterbloom)
+- **Context length detection overhaul** — models.dev integration, provider-aware resolution, fuzzy matching for custom endpoints, `/v1/props` for llama.cpp ([#2158](https://github.com/NousResearch/hermes-agent/pull/2158), [#2051](https://github.com/NousResearch/hermes-agent/pull/2051), [#2403](https://github.com/NousResearch/hermes-agent/pull/2403))
+- **Model catalog updates** — gpt-5.4-mini, gpt-5.4-nano, healer-alpha, haiku-4.5, minimax-m2.7, claude 4.6 at 1M context ([#1913](https://github.com/NousResearch/hermes-agent/pull/1913), [#1915](https://github.com/NousResearch/hermes-agent/pull/1915), [#1900](https://github.com/NousResearch/hermes-agent/pull/1900), [#2155](https://github.com/NousResearch/hermes-agent/pull/2155), [#2474](https://github.com/NousResearch/hermes-agent/pull/2474))
+- **Custom endpoint improvements** — `model.base_url` in config.yaml, `api_mode` override for responses API, allow endpoints without API key, fail fast on missing keys ([#2330](https://github.com/NousResearch/hermes-agent/pull/2330), [#1651](https://github.com/NousResearch/hermes-agent/pull/1651), [#2556](https://github.com/NousResearch/hermes-agent/pull/2556), [#2445](https://github.com/NousResearch/hermes-agent/pull/2445), [#1994](https://github.com/NousResearch/hermes-agent/pull/1994), [#1998](https://github.com/NousResearch/hermes-agent/pull/1998))
+- Inject model and provider into system prompt ([#1929](https://github.com/NousResearch/hermes-agent/pull/1929))
+- Tie `api_mode` to provider config instead of env var ([#1656](https://github.com/NousResearch/hermes-agent/pull/1656))
+- Fix: prevent Anthropic token leaking to third-party `anthropic_messages` providers ([#2389](https://github.com/NousResearch/hermes-agent/pull/2389))
+- Fix: prevent Anthropic fallback from inheriting non-Anthropic `base_url` ([#2388](https://github.com/NousResearch/hermes-agent/pull/2388))
+- Fix: `auxiliary_is_nous` flag never resets — leaked Nous tags to other providers ([#1713](https://github.com/NousResearch/hermes-agent/pull/1713))
+- Fix: Anthropic `tool_choice 'none'` still allowed tool calls ([#1714](https://github.com/NousResearch/hermes-agent/pull/1714))
+- Fix: Mistral parser nested JSON fallback extraction ([#2335](https://github.com/NousResearch/hermes-agent/pull/2335))
+- Fix: MiniMax 401 auth resolved by defaulting to `anthropic_messages` ([#2103](https://github.com/NousResearch/hermes-agent/pull/2103))
+- Fix: case-insensitive model family matching ([#2350](https://github.com/NousResearch/hermes-agent/pull/2350))
+- Fix: ignore placeholder provider keys in activation checks ([#2358](https://github.com/NousResearch/hermes-agent/pull/2358))
+- Fix: Preserve Ollama model:tag colons in context length detection ([#2149](https://github.com/NousResearch/hermes-agent/pull/2149))
+- Fix: recognize Claude Code OAuth credentials in startup gate ([#1663](https://github.com/NousResearch/hermes-agent/pull/1663))
+- Fix: detect Claude Code version dynamically for OAuth user-agent ([#1670](https://github.com/NousResearch/hermes-agent/pull/1670))
+- Fix: OAuth flag stale after refresh/fallback ([#1890](https://github.com/NousResearch/hermes-agent/pull/1890))
+- Fix: auxiliary client skips expired Codex JWT ([#2397](https://github.com/NousResearch/hermes-agent/pull/2397))
+
+### Agent Loop
+- **Gateway prompt caching** — Cache AIAgent per session, keep assistant turns, fix session restore ([#2282](https://github.com/NousResearch/hermes-agent/pull/2282), [#2284](https://github.com/NousResearch/hermes-agent/pull/2284), [#2361](https://github.com/NousResearch/hermes-agent/pull/2361))
+- **Context compression overhaul** — Structured summaries, iterative updates, token-budget tail protection, configurable `summary_base_url` ([#2323](https://github.com/NousResearch/hermes-agent/pull/2323), [#1727](https://github.com/NousResearch/hermes-agent/pull/1727), [#2224](https://github.com/NousResearch/hermes-agent/pull/2224))
+- **Pre-call sanitization and post-call tool guardrails** ([#1732](https://github.com/NousResearch/hermes-agent/pull/1732))
+- **Auto-recover** from provider-rejected `tool_choice` by retrying without ([#2174](https://github.com/NousResearch/hermes-agent/pull/2174))
+- **Background memory/skill review** replaces inline nudges ([#2235](https://github.com/NousResearch/hermes-agent/pull/2235))
+- **SOUL.md as primary agent identity** instead of hardcoded default ([#1922](https://github.com/NousResearch/hermes-agent/pull/1922))
+- Fix: prevent silent tool result loss during context compression ([#1993](https://github.com/NousResearch/hermes-agent/pull/1993))
+- Fix: handle empty/null function arguments in tool call recovery ([#2163](https://github.com/NousResearch/hermes-agent/pull/2163))
+- Fix: handle API refusal responses gracefully instead of crashing ([#2156](https://github.com/NousResearch/hermes-agent/pull/2156))
+- Fix: prevent stuck agent loop on malformed tool calls ([#2114](https://github.com/NousResearch/hermes-agent/pull/2114))
+- Fix: return JSON parse error to model instead of dispatching with empty args ([#2342](https://github.com/NousResearch/hermes-agent/pull/2342))
+- Fix: consecutive assistant message merge drops content on mixed types ([#1703](https://github.com/NousResearch/hermes-agent/pull/1703))
+- Fix: message role alternation violations in JSON recovery and error handler ([#1722](https://github.com/NousResearch/hermes-agent/pull/1722))
+- Fix: `compression_attempts` resets each iteration — allowed unlimited compressions ([#1723](https://github.com/NousResearch/hermes-agent/pull/1723))
+- Fix: `length_continue_retries` never resets — later truncations got fewer retries ([#1717](https://github.com/NousResearch/hermes-agent/pull/1717))
+- Fix: compressor summary role violated consecutive-role constraint ([#1720](https://github.com/NousResearch/hermes-agent/pull/1720), [#1743](https://github.com/NousResearch/hermes-agent/pull/1743))
+- Fix: remove hardcoded `gemini-3-flash-preview` as default summary model ([#2464](https://github.com/NousResearch/hermes-agent/pull/2464))
+- Fix: correctly handle empty tool results ([#2201](https://github.com/NousResearch/hermes-agent/pull/2201))
+- Fix: crash on None entry in `tool_calls` list ([#2209](https://github.com/NousResearch/hermes-agent/pull/2209) by @0xbyt4, [#2316](https://github.com/NousResearch/hermes-agent/pull/2316))
+- Fix: per-thread persistent event loops in worker threads ([#2214](https://github.com/NousResearch/hermes-agent/pull/2214) by @jquesnelle)
+- Fix: prevent 'event loop already running' when async tools run in parallel ([#2207](https://github.com/NousResearch/hermes-agent/pull/2207))
+- Fix: strip ANSI at the source — clean terminal output before it reaches the model ([#2115](https://github.com/NousResearch/hermes-agent/pull/2115))
+- Fix: skip top-level `cache_control` on role:tool for OpenRouter ([#2391](https://github.com/NousResearch/hermes-agent/pull/2391))
+- Fix: delegate tool — save parent tool names before child construction mutates global ([#2083](https://github.com/NousResearch/hermes-agent/pull/2083) by @ygd58, [#1894](https://github.com/NousResearch/hermes-agent/pull/1894))
+- Fix: only strip last assistant message if empty string ([#2326](https://github.com/NousResearch/hermes-agent/pull/2326))
+
+### Session & Memory
+- **Session search** and management slash commands ([#2198](https://github.com/NousResearch/hermes-agent/pull/2198))
+- **Auto session titles** and `.hermes.md` project config ([#1712](https://github.com/NousResearch/hermes-agent/pull/1712))
+- Fix: concurrent memory writes silently drop entries — added file locking ([#1726](https://github.com/NousResearch/hermes-agent/pull/1726))
+- Fix: search all sources by default in `session_search` ([#1892](https://github.com/NousResearch/hermes-agent/pull/1892))
+- Fix: handle hyphenated FTS5 queries and preserve quoted literals ([#1776](https://github.com/NousResearch/hermes-agent/pull/1776))
+- Fix: skip corrupt lines in `load_transcript` instead of crashing ([#1744](https://github.com/NousResearch/hermes-agent/pull/1744))
+- Fix: normalize session keys to prevent case-sensitive duplicates ([#2157](https://github.com/NousResearch/hermes-agent/pull/2157))
+- Fix: prevent `session_search` crash when no sessions exist ([#2194](https://github.com/NousResearch/hermes-agent/pull/2194))
+- Fix: reset token counters on new session for accurate usage display ([#2101](https://github.com/NousResearch/hermes-agent/pull/2101) by @InB4DevOps)
+- Fix: prevent stale memory overwrites by flush agent ([#2687](https://github.com/NousResearch/hermes-agent/pull/2687))
+- Fix: remove synthetic error message injection, fix session resume after repeated failures ([#2303](https://github.com/NousResearch/hermes-agent/pull/2303))
+- Fix: quiet mode with `--resume` now passes conversation_history ([#2357](https://github.com/NousResearch/hermes-agent/pull/2357))
+- Fix: unify resume logic in batch mode ([#2331](https://github.com/NousResearch/hermes-agent/pull/2331))
+
+### Honcho Memory
+- Honcho config fixes and @ context reference integration ([#2343](https://github.com/NousResearch/hermes-agent/pull/2343))
+- Self-hosted / Docker configuration documentation ([#2475](https://github.com/NousResearch/hermes-agent/pull/2475))
+
+---
+
+## 📱 Messaging Platforms (Gateway)
+
+### New Platform Adapters
+- **Signal Messenger** — Full adapter with attachment handling, group message filtering, and Note to Self echo-back protection ([#2206](https://github.com/NousResearch/hermes-agent/pull/2206), [#2400](https://github.com/NousResearch/hermes-agent/pull/2400), [#2297](https://github.com/NousResearch/hermes-agent/pull/2297), [#2156](https://github.com/NousResearch/hermes-agent/pull/2156))
+- **DingTalk** — Adapter with gateway wiring and setup docs ([#1685](https://github.com/NousResearch/hermes-agent/pull/1685), [#1690](https://github.com/NousResearch/hermes-agent/pull/1690), [#1692](https://github.com/NousResearch/hermes-agent/pull/1692))
+- **SMS (Twilio)** ([#1688](https://github.com/NousResearch/hermes-agent/pull/1688))
+- **Mattermost** — With @-mention-only channel filter ([#1683](https://github.com/NousResearch/hermes-agent/pull/1683), [#2443](https://github.com/NousResearch/hermes-agent/pull/2443))
+- **Matrix** — With vision support and image caching ([#1683](https://github.com/NousResearch/hermes-agent/pull/1683), [#2520](https://github.com/NousResearch/hermes-agent/pull/2520))
+- **Webhook** — Platform adapter for external event triggers ([#2166](https://github.com/NousResearch/hermes-agent/pull/2166))
+- **OpenAI-compatible API server** — `/v1/chat/completions` endpoint with `/api/jobs` cron management ([#1756](https://github.com/NousResearch/hermes-agent/pull/1756), [#2450](https://github.com/NousResearch/hermes-agent/pull/2450), [#2456](https://github.com/NousResearch/hermes-agent/pull/2456))
+
+### Telegram Improvements
+- MarkdownV2 support — strikethrough, spoiler, blockquotes, escape parentheses/braces/backslashes/backticks ([#2199](https://github.com/NousResearch/hermes-agent/pull/2199), [#2200](https://github.com/NousResearch/hermes-agent/pull/2200) by @llbn, [#2386](https://github.com/NousResearch/hermes-agent/pull/2386))
+- Auto-detect HTML tags and use `parse_mode=HTML` ([#1709](https://github.com/NousResearch/hermes-agent/pull/1709))
+- Telegram group vision support + thread-based sessions ([#2153](https://github.com/NousResearch/hermes-agent/pull/2153))
+- Auto-reconnect polling after network interruption ([#2517](https://github.com/NousResearch/hermes-agent/pull/2517))
+- Aggregate split text messages before dispatching ([#1674](https://github.com/NousResearch/hermes-agent/pull/1674))
+- Fix: streaming config bridge, not-modified, flood control ([#1782](https://github.com/NousResearch/hermes-agent/pull/1782), [#1783](https://github.com/NousResearch/hermes-agent/pull/1783))
+- Fix: edited_message event crashes ([#2074](https://github.com/NousResearch/hermes-agent/pull/2074))
+- Fix: retry 409 polling conflicts before giving up ([#2312](https://github.com/NousResearch/hermes-agent/pull/2312))
+- Fix: topic delivery via `platform:chat_id:thread_id` format ([#2455](https://github.com/NousResearch/hermes-agent/pull/2455))
+
+### Discord Improvements
+- Document caching and text-file injection ([#2503](https://github.com/NousResearch/hermes-agent/pull/2503))
+- Persistent typing indicator for DMs ([#2468](https://github.com/NousResearch/hermes-agent/pull/2468))
+- Discord DM vision — inline images + attachment analysis ([#2186](https://github.com/NousResearch/hermes-agent/pull/2186))
+- Persist thread participation across gateway restarts ([#1661](https://github.com/NousResearch/hermes-agent/pull/1661))
+- Fix: gateway crash on non-ASCII guild names ([#2302](https://github.com/NousResearch/hermes-agent/pull/2302))
+- Fix: thread permission errors ([#2073](https://github.com/NousResearch/hermes-agent/pull/2073))
+- Fix: slash event routing in threads ([#2460](https://github.com/NousResearch/hermes-agent/pull/2460))
+- Fix: remove bugged followup messages + `/ask` command ([#1836](https://github.com/NousResearch/hermes-agent/pull/1836))
+- Fix: graceful WebSocket reconnection ([#2127](https://github.com/NousResearch/hermes-agent/pull/2127))
+- Fix: voice channel TTS when streaming enabled ([#2322](https://github.com/NousResearch/hermes-agent/pull/2322))
+
+### WhatsApp & Other Adapters
+- WhatsApp: outbound `send_message` routing ([#1769](https://github.com/NousResearch/hermes-agent/pull/1769) by @sai-samarth), LID format self-chat ([#1667](https://github.com/NousResearch/hermes-agent/pull/1667)), `reply_prefix` config fix ([#1923](https://github.com/NousResearch/hermes-agent/pull/1923)), restart on bridge child exit ([#2334](https://github.com/NousResearch/hermes-agent/pull/2334)), image/bridge improvements ([#2181](https://github.com/NousResearch/hermes-agent/pull/2181))
+- Matrix: correct `reply_to_message_id` parameter ([#1895](https://github.com/NousResearch/hermes-agent/pull/1895)), bare media types fix ([#1736](https://github.com/NousResearch/hermes-agent/pull/1736))
+- Mattermost: MIME types for media attachments ([#2329](https://github.com/NousResearch/hermes-agent/pull/2329))
+
+### Gateway Core
+- **Auto-reconnect** failed platforms with exponential backoff ([#2584](https://github.com/NousResearch/hermes-agent/pull/2584))
+- **Notify users when session auto-resets** ([#2519](https://github.com/NousResearch/hermes-agent/pull/2519))
+- **Reply-to message context** for out-of-session replies ([#1662](https://github.com/NousResearch/hermes-agent/pull/1662))
+- **Ignore unauthorized DMs** config option ([#1919](https://github.com/NousResearch/hermes-agent/pull/1919))
+- Fix: `/reset` in thread-mode resets global session instead of thread ([#2254](https://github.com/NousResearch/hermes-agent/pull/2254))
+- Fix: deliver MEDIA: files after streaming responses ([#2382](https://github.com/NousResearch/hermes-agent/pull/2382))
+- Fix: cap interrupt recursion depth to prevent resource exhaustion ([#1659](https://github.com/NousResearch/hermes-agent/pull/1659))
+- Fix: detect stopped processes and release stale locks on `--replace` ([#2406](https://github.com/NousResearch/hermes-agent/pull/2406), [#1908](https://github.com/NousResearch/hermes-agent/pull/1908))
+- Fix: PID-based wait with force-kill for gateway restart ([#1902](https://github.com/NousResearch/hermes-agent/pull/1902))
+- Fix: prevent `--replace` mode from killing the caller process ([#2185](https://github.com/NousResearch/hermes-agent/pull/2185))
+- Fix: `/model` shows active fallback model instead of config default ([#1660](https://github.com/NousResearch/hermes-agent/pull/1660))
+- Fix: `/title` command fails when session doesn't exist in SQLite yet ([#2379](https://github.com/NousResearch/hermes-agent/pull/2379) by @ten-jampa)
+- Fix: process `/queue`'d messages after agent completion ([#2469](https://github.com/NousResearch/hermes-agent/pull/2469))
+- Fix: strip orphaned `tool_results` + let `/reset` bypass running agent ([#2180](https://github.com/NousResearch/hermes-agent/pull/2180))
+- Fix: prevent agents from starting gateway outside systemd management ([#2617](https://github.com/NousResearch/hermes-agent/pull/2617))
+- Fix: prevent systemd restart storm on gateway connection failure ([#2327](https://github.com/NousResearch/hermes-agent/pull/2327))
+- Fix: include resolved node path in systemd unit ([#1767](https://github.com/NousResearch/hermes-agent/pull/1767) by @sai-samarth)
+- Fix: send error details to user in gateway outer exception handler ([#1966](https://github.com/NousResearch/hermes-agent/pull/1966))
+- Fix: improve error handling for 429 usage limits and 500 context overflow ([#1839](https://github.com/NousResearch/hermes-agent/pull/1839))
+- Fix: add all missing platform allowlist env vars to startup warning check ([#2628](https://github.com/NousResearch/hermes-agent/pull/2628))
+- Fix: media delivery fails for file paths containing spaces ([#2621](https://github.com/NousResearch/hermes-agent/pull/2621))
+- Fix: duplicate session-key collision in multi-platform gateway ([#2171](https://github.com/NousResearch/hermes-agent/pull/2171))
+- Fix: Matrix and Mattermost never report as connected ([#1711](https://github.com/NousResearch/hermes-agent/pull/1711))
+- Fix: PII redaction config never read — missing yaml import ([#1701](https://github.com/NousResearch/hermes-agent/pull/1701))
+- Fix: NameError on skill slash commands ([#1697](https://github.com/NousResearch/hermes-agent/pull/1697))
+- Fix: persist watcher metadata in checkpoint for crash recovery ([#1706](https://github.com/NousResearch/hermes-agent/pull/1706))
+- Fix: pass `message_thread_id` in send_image_file, send_document, send_video ([#2339](https://github.com/NousResearch/hermes-agent/pull/2339))
+- Fix: media-group aggregation on rapid successive photo messages ([#2160](https://github.com/NousResearch/hermes-agent/pull/2160))
+
+---
+
+## 🔧 Tool System
+
+### MCP Enhancements
+- **MCP server management CLI** + OAuth 2.1 PKCE auth ([#2465](https://github.com/NousResearch/hermes-agent/pull/2465))
+- **Expose MCP servers as standalone toolsets** ([#1907](https://github.com/NousResearch/hermes-agent/pull/1907))
+- **Interactive MCP tool configuration** in `hermes tools` ([#1694](https://github.com/NousResearch/hermes-agent/pull/1694))
+- Fix: MCP-OAuth port mismatch, path traversal, and shared handler state ([#2552](https://github.com/NousResearch/hermes-agent/pull/2552))
+- Fix: preserve MCP tool registrations across session resets ([#2124](https://github.com/NousResearch/hermes-agent/pull/2124))
+- Fix: concurrent file access crash + duplicate MCP registration ([#2154](https://github.com/NousResearch/hermes-agent/pull/2154))
+- Fix: normalise MCP schemas + expand session list columns ([#2102](https://github.com/NousResearch/hermes-agent/pull/2102))
+- Fix: `tool_choice` `mcp_` prefix handling ([#1775](https://github.com/NousResearch/hermes-agent/pull/1775))
+
+### Web Tool Backends
+- **Tavily** as web search/extract/crawl backend ([#1731](https://github.com/NousResearch/hermes-agent/pull/1731))
+- **Parallel** as alternative web search/extract backend ([#1696](https://github.com/NousResearch/hermes-agent/pull/1696))
+- **Configurable web backend** — Firecrawl/BeautifulSoup/Playwright selection ([#2256](https://github.com/NousResearch/hermes-agent/pull/2256))
+- Fix: whitespace-only env vars bypass web backend detection ([#2341](https://github.com/NousResearch/hermes-agent/pull/2341))
+
+### New Tools
+- **IMAP email** reading and sending ([#2173](https://github.com/NousResearch/hermes-agent/pull/2173))
+- **STT (speech-to-text)** tool using Whisper API ([#2072](https://github.com/NousResearch/hermes-agent/pull/2072))
+- **Route-aware pricing estimates** ([#1695](https://github.com/NousResearch/hermes-agent/pull/1695))
+
+### Tool Improvements
+- TTS: `base_url` support for OpenAI TTS provider ([#2064](https://github.com/NousResearch/hermes-agent/pull/2064) by @hanai)
+- Vision: configurable timeout, tilde expansion in file paths, DM vision with multi-image and base64 fallback ([#2480](https://github.com/NousResearch/hermes-agent/pull/2480), [#2585](https://github.com/NousResearch/hermes-agent/pull/2585), [#2211](https://github.com/NousResearch/hermes-agent/pull/2211))
+- Browser: race condition fix in session creation ([#1721](https://github.com/NousResearch/hermes-agent/pull/1721)), TypeError on unexpected LLM params ([#1735](https://github.com/NousResearch/hermes-agent/pull/1735))
+- File tools: strip ANSI escape codes from write_file and patch content ([#2532](https://github.com/NousResearch/hermes-agent/pull/2532)), include pagination args in repeated search key ([#1824](https://github.com/NousResearch/hermes-agent/pull/1824) by @cutepawss), improve fuzzy matching accuracy + position calculation refactor ([#2096](https://github.com/NousResearch/hermes-agent/pull/2096), [#1681](https://github.com/NousResearch/hermes-agent/pull/1681))
+- Code execution: resource leak and double socket close fix ([#2381](https://github.com/NousResearch/hermes-agent/pull/2381))
+- Delegate: thread safety for concurrent subagent delegation ([#1672](https://github.com/NousResearch/hermes-agent/pull/1672)), preserve parent agent's tool list after delegation ([#1778](https://github.com/NousResearch/hermes-agent/pull/1778))
+- Fix: make concurrent tool batching path-aware for file mutations ([#1914](https://github.com/NousResearch/hermes-agent/pull/1914))
+- Fix: chunk long messages in `send_message_tool` before platform dispatch ([#1646](https://github.com/NousResearch/hermes-agent/pull/1646))
+- Fix: add missing 'messaging' toolset ([#1718](https://github.com/NousResearch/hermes-agent/pull/1718))
+- Fix: prevent unavailable tool names from leaking into model schemas ([#2072](https://github.com/NousResearch/hermes-agent/pull/2072))
+- Fix: pass visited set by reference to prevent diamond dependency duplication ([#2311](https://github.com/NousResearch/hermes-agent/pull/2311))
+- Fix: Daytona sandbox lookup migrated from `find_one` to `get/list` ([#2063](https://github.com/NousResearch/hermes-agent/pull/2063) by @rovle)
+
+---
+
+## 🧩 Skills Ecosystem
+
+### Skills System Improvements
+- **Agent-created skills** — Caution-level findings allowed, dangerous skills ask instead of block ([#1840](https://github.com/NousResearch/hermes-agent/pull/1840), [#2446](https://github.com/NousResearch/hermes-agent/pull/2446))
+- **`--yes` flag** to bypass confirmation in `/skills install` and uninstall ([#1647](https://github.com/NousResearch/hermes-agent/pull/1647))
+- **Disabled skills respected** across banner, system prompt, and slash commands ([#1897](https://github.com/NousResearch/hermes-agent/pull/1897))
+- Fix: skills custom_tools import crash + sandbox file_tools integration ([#2239](https://github.com/NousResearch/hermes-agent/pull/2239))
+- Fix: agent-created skills with pip requirements crash on install ([#2145](https://github.com/NousResearch/hermes-agent/pull/2145))
+- Fix: race condition in `Skills.__init__` when `hub.yaml` missing ([#2242](https://github.com/NousResearch/hermes-agent/pull/2242))
+- Fix: validate skill metadata before install and block duplicates ([#2241](https://github.com/NousResearch/hermes-agent/pull/2241))
+- Fix: skills hub inspect/resolve — 4 bugs in inspect, redirects, discovery, tap list ([#2447](https://github.com/NousResearch/hermes-agent/pull/2447))
+- Fix: agent-created skills keep working after session reset ([#2121](https://github.com/NousResearch/hermes-agent/pull/2121))
+
+### New Skills
+- **OCR-and-documents** — PDF/DOCX/XLS/PPTX/image OCR with optional GPU ([#2236](https://github.com/NousResearch/hermes-agent/pull/2236), [#2461](https://github.com/NousResearch/hermes-agent/pull/2461))
+- **Huggingface-hub** bundled skill ([#1921](https://github.com/NousResearch/hermes-agent/pull/1921))
+- **Sherlock OSINT** username search ([#1671](https://github.com/NousResearch/hermes-agent/pull/1671))
+- **Meme-generation** — Image generator with Pillow ([#2344](https://github.com/NousResearch/hermes-agent/pull/2344))
+- **Bioinformatics** gateway skill — index to 400+ bio skills ([#2387](https://github.com/NousResearch/hermes-agent/pull/2387))
+- **Inference.sh** skill (terminal-based) ([#1686](https://github.com/NousResearch/hermes-agent/pull/1686))
+- **Base blockchain** optional skill ([#1643](https://github.com/NousResearch/hermes-agent/pull/1643))
+- **3D-model-viewer** optional skill ([#2226](https://github.com/NousResearch/hermes-agent/pull/2226))
+- **FastMCP** optional skill ([#2113](https://github.com/NousResearch/hermes-agent/pull/2113))
+- **Hermes-agent-setup** skill ([#1905](https://github.com/NousResearch/hermes-agent/pull/1905))
+
+---
+
+## 🔌 Plugin System Enhancements
+
+- **TUI extension hooks** — Build custom CLIs on top of Hermes ([#2333](https://github.com/NousResearch/hermes-agent/pull/2333))
+- **`hermes plugins install/remove/list`** commands ([#2337](https://github.com/NousResearch/hermes-agent/pull/2337))
+- **Slash command registration** for plugins ([#2359](https://github.com/NousResearch/hermes-agent/pull/2359))
+- **`session:end` lifecycle event** hook ([#1725](https://github.com/NousResearch/hermes-agent/pull/1725))
+- Fix: require opt-in for project plugin discovery ([#2215](https://github.com/NousResearch/hermes-agent/pull/2215))
+
+---
+
+## 🔒 Security & Reliability
+
+### Security
+- **SSRF protection** for vision_tools and web_tools ([#2679](https://github.com/NousResearch/hermes-agent/pull/2679))
+- **Shell injection prevention** in `_expand_path` via `~user` path suffix ([#2685](https://github.com/NousResearch/hermes-agent/pull/2685))
+- **Block untrusted browser-origin** API server access ([#2451](https://github.com/NousResearch/hermes-agent/pull/2451))
+- **Block sandbox backend creds** from subprocess env ([#1658](https://github.com/NousResearch/hermes-agent/pull/1658))
+- **Block @ references** from reading secrets outside workspace ([#2601](https://github.com/NousResearch/hermes-agent/pull/2601) by @Gutslabs)
+- **Malicious code pattern pre-exec scanner** for terminal_tool ([#2245](https://github.com/NousResearch/hermes-agent/pull/2245))
+- **Harden terminal safety** and sandbox file writes ([#1653](https://github.com/NousResearch/hermes-agent/pull/1653))
+- **PKCE verifier leak** fix + OAuth refresh Content-Type ([#1775](https://github.com/NousResearch/hermes-agent/pull/1775))
+- **Eliminate SQL string formatting** in `execute()` calls ([#2061](https://github.com/NousResearch/hermes-agent/pull/2061) by @dusterbloom)
+- **Harden jobs API** — input limits, field whitelist, startup check ([#2456](https://github.com/NousResearch/hermes-agent/pull/2456))
+
+### Reliability
+- Thread locks on 4 SessionDB methods ([#1704](https://github.com/NousResearch/hermes-agent/pull/1704))
+- File locking for concurrent memory writes ([#1726](https://github.com/NousResearch/hermes-agent/pull/1726))
+- Handle OpenRouter errors gracefully ([#2112](https://github.com/NousResearch/hermes-agent/pull/2112))
+- Guard print() calls against OSError ([#1668](https://github.com/NousResearch/hermes-agent/pull/1668))
+- Safely handle non-string inputs in redacting formatter ([#2392](https://github.com/NousResearch/hermes-agent/pull/2392), [#1700](https://github.com/NousResearch/hermes-agent/pull/1700))
+- ACP: preserve session provider on model switch, persist sessions to disk ([#2380](https://github.com/NousResearch/hermes-agent/pull/2380), [#2071](https://github.com/NousResearch/hermes-agent/pull/2071))
+- API server: persist ResponseStore to SQLite across restarts ([#2472](https://github.com/NousResearch/hermes-agent/pull/2472))
+- Fix: `fetch_nous_models` always TypeError from positional args ([#1699](https://github.com/NousResearch/hermes-agent/pull/1699))
+- Fix: resolve merge conflict markers in cli.py breaking startup ([#2347](https://github.com/NousResearch/hermes-agent/pull/2347))
+- Fix: `minisweagent_path.py` missing from wheel ([#2098](https://github.com/NousResearch/hermes-agent/pull/2098) by @JiwaniZakir)
+
+### Cron System
+- **`[SILENT]` response** — cron agents can suppress delivery ([#1833](https://github.com/NousResearch/hermes-agent/pull/1833))
+- **Scale missed-job grace window** with schedule frequency ([#2449](https://github.com/NousResearch/hermes-agent/pull/2449))
+- **Recover recent one-shot jobs** ([#1918](https://github.com/NousResearch/hermes-agent/pull/1918))
+- Fix: normalize `repeat<=0` to None — jobs deleted after first run when LLM passes -1 ([#2612](https://github.com/NousResearch/hermes-agent/pull/2612) by @Mibayy)
+- Fix: Matrix added to scheduler delivery platform_map ([#2167](https://github.com/NousResearch/hermes-agent/pull/2167) by @buntingszn)
+- Fix: naive ISO timestamps without timezone — jobs fire at wrong time ([#1729](https://github.com/NousResearch/hermes-agent/pull/1729))
+- Fix: `get_due_jobs` reads `jobs.json` twice — race condition ([#1716](https://github.com/NousResearch/hermes-agent/pull/1716))
+- Fix: silent jobs return empty response for delivery skip ([#2442](https://github.com/NousResearch/hermes-agent/pull/2442))
+- Fix: stop injecting cron outputs into gateway session history ([#2313](https://github.com/NousResearch/hermes-agent/pull/2313))
+- Fix: close abandoned coroutine when `asyncio.run()` raises RuntimeError ([#2317](https://github.com/NousResearch/hermes-agent/pull/2317))
+
+---
+
+## 🧪 Testing
+
+- Resolve all consistently failing tests ([#2488](https://github.com/NousResearch/hermes-agent/pull/2488))
+- Replace `FakePath` with `monkeypatch` for Python 3.12 compat ([#2444](https://github.com/NousResearch/hermes-agent/pull/2444))
+- Align Hermes setup and full-suite expectations ([#1710](https://github.com/NousResearch/hermes-agent/pull/1710))
+
+---
+
+## 📚 Documentation
+
+- Comprehensive docs update for recent features ([#1693](https://github.com/NousResearch/hermes-agent/pull/1693), [#2183](https://github.com/NousResearch/hermes-agent/pull/2183))
+- Alibaba Cloud and DingTalk setup guides ([#1687](https://github.com/NousResearch/hermes-agent/pull/1687), [#1692](https://github.com/NousResearch/hermes-agent/pull/1692))
+- Detailed skills documentation ([#2244](https://github.com/NousResearch/hermes-agent/pull/2244))
+- Honcho self-hosted / Docker configuration ([#2475](https://github.com/NousResearch/hermes-agent/pull/2475))
+- Context length detection FAQ and quickstart references ([#2179](https://github.com/NousResearch/hermes-agent/pull/2179))
+- Fix docs inconsistencies across reference and user guides ([#1995](https://github.com/NousResearch/hermes-agent/pull/1995))
+- Fix MCP install commands — use uv, not bare pip ([#1909](https://github.com/NousResearch/hermes-agent/pull/1909))
+- Replace ASCII diagrams with Mermaid/lists ([#2402](https://github.com/NousResearch/hermes-agent/pull/2402))
+- Gemini OAuth provider implementation plan ([#2467](https://github.com/NousResearch/hermes-agent/pull/2467))
+- Discord Server Members Intent marked as required ([#2330](https://github.com/NousResearch/hermes-agent/pull/2330))
+- Fix MDX build error in api-server.md ([#1787](https://github.com/NousResearch/hermes-agent/pull/1787))
+- Align venv path to match installer ([#2114](https://github.com/NousResearch/hermes-agent/pull/2114))
+- New skills added to hub index ([#2281](https://github.com/NousResearch/hermes-agent/pull/2281))
+
+---
+
+## 👥 Contributors
+
+### Core
+- **@teknium1** (Teknium) — 280 PRs
+
+### Community Contributors
+- **@mchzimm** (to_the_max) — GitHub Copilot provider integration ([#1879](https://github.com/NousResearch/hermes-agent/pull/1879))
+- **@jquesnelle** (Jeffrey Quesnelle) — Per-thread persistent event loops fix ([#2214](https://github.com/NousResearch/hermes-agent/pull/2214))
+- **@llbn** (lbn) — Telegram MarkdownV2 strikethrough, spoiler, blockquotes, and escape fixes ([#2199](https://github.com/NousResearch/hermes-agent/pull/2199), [#2200](https://github.com/NousResearch/hermes-agent/pull/2200))
+- **@dusterbloom** — SQL injection prevention + local server context window querying ([#2061](https://github.com/NousResearch/hermes-agent/pull/2061), [#2091](https://github.com/NousResearch/hermes-agent/pull/2091))
+- **@0xbyt4** — Anthropic tool_calls None guard + OpenCode-Go provider config fix ([#2209](https://github.com/NousResearch/hermes-agent/pull/2209), [#2393](https://github.com/NousResearch/hermes-agent/pull/2393))
+- **@sai-samarth** (Saisamarth) — WhatsApp send_message routing + systemd node path ([#1769](https://github.com/NousResearch/hermes-agent/pull/1769), [#1767](https://github.com/NousResearch/hermes-agent/pull/1767))
+- **@Gutslabs** (Guts) — Block @ references from reading secrets ([#2601](https://github.com/NousResearch/hermes-agent/pull/2601))
+- **@Mibayy** (Mibay) — Cron job repeat normalization ([#2612](https://github.com/NousResearch/hermes-agent/pull/2612))
+- **@ten-jampa** (Tenzin Jampa) — Gateway /title command fix ([#2379](https://github.com/NousResearch/hermes-agent/pull/2379))
+- **@cutepawss** (lila) — File tools search pagination fix ([#1824](https://github.com/NousResearch/hermes-agent/pull/1824))
+- **@hanai** (Hanai) — OpenAI TTS base_url support ([#2064](https://github.com/NousResearch/hermes-agent/pull/2064))
+- **@rovle** (Lovre Pešut) — Daytona sandbox API migration ([#2063](https://github.com/NousResearch/hermes-agent/pull/2063))
+- **@buntingszn** (bunting szn) — Matrix cron delivery support ([#2167](https://github.com/NousResearch/hermes-agent/pull/2167))
+- **@InB4DevOps** — Token counter reset on new session ([#2101](https://github.com/NousResearch/hermes-agent/pull/2101))
+- **@JiwaniZakir** (Zakir Jiwani) — Missing file in wheel fix ([#2098](https://github.com/NousResearch/hermes-agent/pull/2098))
+- **@ygd58** (buray) — Delegate tool parent tool names fix ([#2083](https://github.com/NousResearch/hermes-agent/pull/2083))
+
+---
+
+**Full Changelog**: [v2026.3.17...v2026.3.23](https://github.com/NousResearch/hermes-agent/compare/v2026.3.17...v2026.3.23)

RELEASE_v0.5.0.md

@@ -0,0 +1,348 @@
+# Hermes Agent v0.5.0 (v2026.3.28)
+
+**Release Date:** March 28, 2026
+
+> The hardening release — Hugging Face provider, /model command overhaul, Telegram Private Chat Topics, native Modal SDK, plugin lifecycle hooks, tool-use enforcement for GPT models, Nix flake, 50+ security and reliability fixes, and a comprehensive supply chain audit.
+
+---
+
+## ✨ Highlights
+
+- **Nous Portal now supports 400+ models** — The Nous Research inference portal has expanded dramatically, giving Hermes Agent users access to over 400 models through a single provider endpoint
+
+- **Hugging Face as a first-class inference provider** — Full integration with HF Inference API including curated agentic model picker that maps to OpenRouter analogues, live `/models` endpoint probe, and setup wizard flow ([#3419](https://github.com/NousResearch/hermes-agent/pull/3419), [#3440](https://github.com/NousResearch/hermes-agent/pull/3440))
+
+- **Telegram Private Chat Topics** — Project-based conversations with functional skill binding per topic, enabling isolated workflows within a single Telegram chat ([#3163](https://github.com/NousResearch/hermes-agent/pull/3163))
+
+- **Native Modal SDK backend** — Replaced swe-rex dependency with native Modal SDK (`Sandbox.create.aio` + `exec.aio`), eliminating tunnels and simplifying the Modal terminal backend ([#3538](https://github.com/NousResearch/hermes-agent/pull/3538))
+
+- **Plugin lifecycle hooks activated** — `pre_llm_call`, `post_llm_call`, `on_session_start`, and `on_session_end` hooks now fire in the agent loop and CLI/gateway, completing the plugin hook system ([#3542](https://github.com/NousResearch/hermes-agent/pull/3542))
+
+- **Improved OpenAI Model Reliability** — Added `GPT_TOOL_USE_GUIDANCE` to prevent GPT models from describing intended actions instead of making tool calls, plus automatic stripping of stale budget warnings from conversation history that caused models to avoid tools across turns ([#3528](https://github.com/NousResearch/hermes-agent/pull/3528))
+
+- **Nix flake** — Full uv2nix build, NixOS module with persistent container mode, auto-generated config keys from Python source, and suffix PATHs for agent-friendliness ([#20](https://github.com/NousResearch/hermes-agent/pull/20), [#3274](https://github.com/NousResearch/hermes-agent/pull/3274), [#3061](https://github.com/NousResearch/hermes-agent/pull/3061)) by @alt-glitch
+
+- **Supply chain hardening** — Removed compromised `litellm` dependency, pinned all dependency version ranges, regenerated `uv.lock` with hashes, added CI workflow scanning PRs for supply chain attack patterns, and bumped deps to fix CVEs ([#2796](https://github.com/NousResearch/hermes-agent/pull/2796), [#2810](https://github.com/NousResearch/hermes-agent/pull/2810), [#2812](https://github.com/NousResearch/hermes-agent/pull/2812), [#2816](https://github.com/NousResearch/hermes-agent/pull/2816), [#3073](https://github.com/NousResearch/hermes-agent/pull/3073))
+
+- **Anthropic output limits fix** — Replaced hardcoded 16K `max_tokens` with per-model native output limits (128K for Opus 4.6, 64K for Sonnet 4.6), fixing "Response truncated" and thinking-budget exhaustion on direct Anthropic API ([#3426](https://github.com/NousResearch/hermes-agent/pull/3426), [#3444](https://github.com/NousResearch/hermes-agent/pull/3444))
+
+---
+
+## 🏗️ Core Agent & Architecture
+
+### New Provider: Hugging Face
+- First-class Hugging Face Inference API integration with auth, setup wizard, and model picker ([#3419](https://github.com/NousResearch/hermes-agent/pull/3419))
+- Curated model list mapping OpenRouter agentic defaults to HF equivalents — providers with 8+ curated models skip live `/models` probe for speed ([#3440](https://github.com/NousResearch/hermes-agent/pull/3440))
+- Added glm-5-turbo to Z.AI provider model list ([#3095](https://github.com/NousResearch/hermes-agent/pull/3095))
+
+### Provider & Model Improvements
+- `/model` command overhaul — extracted shared `switch_model()` pipeline for CLI and gateway, custom endpoint support, provider-aware routing ([#2795](https://github.com/NousResearch/hermes-agent/pull/2795), [#2799](https://github.com/NousResearch/hermes-agent/pull/2799))
+- Removed `/model` slash command from CLI and gateway in favor of `hermes model` subcommand ([#3080](https://github.com/NousResearch/hermes-agent/pull/3080))
+- Preserve `custom` provider instead of silently remapping to `openrouter` ([#2792](https://github.com/NousResearch/hermes-agent/pull/2792))
+- Read root-level `provider` and `base_url` from config.yaml into model config ([#3112](https://github.com/NousResearch/hermes-agent/pull/3112))
+- Align Nous Portal model slugs with OpenRouter naming ([#3253](https://github.com/NousResearch/hermes-agent/pull/3253))
+- Fix Alibaba provider default endpoint and model list ([#3484](https://github.com/NousResearch/hermes-agent/pull/3484))
+- Allow MiniMax users to override `/v1` → `/anthropic` auto-correction ([#3553](https://github.com/NousResearch/hermes-agent/pull/3553))
+- Migrate OAuth token refresh to `platform.claude.com` with fallback ([#3246](https://github.com/NousResearch/hermes-agent/pull/3246))
+
+### Agent Loop & Conversation
+- **Improved OpenAI model reliability** — `GPT_TOOL_USE_GUIDANCE` prevents GPT models from describing actions instead of calling tools + automatic budget warning stripping from history ([#3528](https://github.com/NousResearch/hermes-agent/pull/3528))
+- **Surface lifecycle events** — All retry, fallback, and compression events now surface to the user as formatted messages ([#3153](https://github.com/NousResearch/hermes-agent/pull/3153))
+- **Anthropic output limits** — Per-model native output limits instead of hardcoded 16K `max_tokens` ([#3426](https://github.com/NousResearch/hermes-agent/pull/3426))
+- **Thinking-budget exhaustion detection** — Skip useless continuation retries when model uses all output tokens on reasoning ([#3444](https://github.com/NousResearch/hermes-agent/pull/3444))
+- Always prefer streaming for API calls to prevent hung subagents ([#3120](https://github.com/NousResearch/hermes-agent/pull/3120))
+- Restore safe non-streaming fallback after stream failures ([#3020](https://github.com/NousResearch/hermes-agent/pull/3020))
+- Give subagents independent iteration budgets ([#3004](https://github.com/NousResearch/hermes-agent/pull/3004))
+- Update `api_key` in `_try_activate_fallback` for subagent auth ([#3103](https://github.com/NousResearch/hermes-agent/pull/3103))
+- Graceful return on max retries instead of crashing thread ([untagged commit](https://github.com/NousResearch/hermes-agent))
+- Count compression restarts toward retry limit ([#3070](https://github.com/NousResearch/hermes-agent/pull/3070))
+- Include tool tokens in preflight estimate, guard context probe persistence ([#3164](https://github.com/NousResearch/hermes-agent/pull/3164))
+- Update context compressor limits after fallback activation ([#3305](https://github.com/NousResearch/hermes-agent/pull/3305))
+- Validate empty user messages to prevent Anthropic API 400 errors ([#3322](https://github.com/NousResearch/hermes-agent/pull/3322))
+- GLM reasoning-only and max-length handling ([#3010](https://github.com/NousResearch/hermes-agent/pull/3010))
+- Increase API timeout default from 900s to 1800s for slow-thinking models ([#3431](https://github.com/NousResearch/hermes-agent/pull/3431))
+- Send `max_tokens` for Claude/OpenRouter + retry SSE connection errors ([#3497](https://github.com/NousResearch/hermes-agent/pull/3497))
+- Prevent AsyncOpenAI/httpx cross-loop deadlock in gateway mode ([#2701](https://github.com/NousResearch/hermes-agent/pull/2701)) by @ctlst
+
+### Streaming & Reasoning
+- **Persist reasoning across gateway session turns** with new schema v6 columns (`reasoning`, `reasoning_details`, `codex_reasoning_items`) ([#2974](https://github.com/NousResearch/hermes-agent/pull/2974))
+- Detect and kill stale SSE connections ([untagged commit](https://github.com/NousResearch/hermes-agent))
+- Fix stale stream detector race causing spurious `RemoteProtocolError` ([untagged commit](https://github.com/NousResearch/hermes-agent))
+- Skip duplicate callback for `<think>`-extracted reasoning during streaming ([#3116](https://github.com/NousResearch/hermes-agent/pull/3116))
+- Preserve reasoning fields in `rewrite_transcript` ([#3311](https://github.com/NousResearch/hermes-agent/pull/3311))
+- Preserve Gemini thought signatures in streamed tool calls ([#2997](https://github.com/NousResearch/hermes-agent/pull/2997))
+- Ensure first delta is fired during reasoning updates ([untagged commit](https://github.com/NousResearch/hermes-agent))
+
+### Session & Memory
+- **Session search recent sessions mode** — Omit query to browse recent sessions with titles, previews, and timestamps ([#2533](https://github.com/NousResearch/hermes-agent/pull/2533))
+- **Session config surfacing** on `/new`, `/reset`, and auto-reset ([#3321](https://github.com/NousResearch/hermes-agent/pull/3321))
+- **Third-party session isolation** — `--source` flag for isolating sessions by origin ([#3255](https://github.com/NousResearch/hermes-agent/pull/3255))
+- Add `/resume` CLI handler, session log truncation guard, `reopen_session` API ([#3315](https://github.com/NousResearch/hermes-agent/pull/3315))
+- Clear compressor summary and turn counter on `/clear` and `/new` ([#3102](https://github.com/NousResearch/hermes-agent/pull/3102))
+- Surface silent SessionDB failures that cause session data loss ([#2999](https://github.com/NousResearch/hermes-agent/pull/2999))
+- Session search fallback preview on summarization failure ([#3478](https://github.com/NousResearch/hermes-agent/pull/3478))
+- Prevent stale memory overwrites by flush agent ([#2687](https://github.com/NousResearch/hermes-agent/pull/2687))
+
+### Context Compression
+- Replace dead `summary_target_tokens` with ratio-based scaling ([#2554](https://github.com/NousResearch/hermes-agent/pull/2554))
+- Expose `compression.target_ratio`, `protect_last_n`, and `threshold` in `DEFAULT_CONFIG` ([untagged commit](https://github.com/NousResearch/hermes-agent))
+- Restore sane defaults and cap summary at 12K tokens ([untagged commit](https://github.com/NousResearch/hermes-agent))
+- Preserve transcript on `/compress` and hygiene compression ([#3556](https://github.com/NousResearch/hermes-agent/pull/3556))
+- Update context pressure warnings and token estimates after compaction ([untagged commit](https://github.com/NousResearch/hermes-agent))
+
+### Architecture & Dependencies
+- **Remove mini-swe-agent dependency** — Inline Docker and Modal backends directly ([#2804](https://github.com/NousResearch/hermes-agent/pull/2804))
+- **Replace swe-rex with native Modal SDK** for Modal backend ([#3538](https://github.com/NousResearch/hermes-agent/pull/3538))
+- **Plugin lifecycle hooks** — `pre_llm_call`, `post_llm_call`, `on_session_start`, `on_session_end` now fire in the agent loop ([#3542](https://github.com/NousResearch/hermes-agent/pull/3542))
+- Fix plugin toolsets invisible in `hermes tools` and standalone processes ([#3457](https://github.com/NousResearch/hermes-agent/pull/3457))
+- Consolidate `get_hermes_home()` and `parse_reasoning_effort()` ([#3062](https://github.com/NousResearch/hermes-agent/pull/3062))
+- Remove unused Hermes-native PKCE OAuth flow ([#3107](https://github.com/NousResearch/hermes-agent/pull/3107))
+- Remove ~100 unused imports across 55 files ([#3016](https://github.com/NousResearch/hermes-agent/pull/3016))
+- Fix 154 f-strings, simplify getattr/URL patterns, remove dead code ([#3119](https://github.com/NousResearch/hermes-agent/pull/3119))
+
+---
+
+## 📱 Messaging Platforms (Gateway)
+
+### Telegram
+- **Private Chat Topics** — Project-based conversations with functional skill binding per topic, enabling isolated workflows within a single Telegram chat ([#3163](https://github.com/NousResearch/hermes-agent/pull/3163))
+- **Auto-discover fallback IPs via DNS-over-HTTPS** when `api.telegram.org` is unreachable ([#3376](https://github.com/NousResearch/hermes-agent/pull/3376))
+- **Configurable reply threading mode** ([#2907](https://github.com/NousResearch/hermes-agent/pull/2907))
+- Fall back to no `thread_id` on "Message thread not found" BadRequest ([#3390](https://github.com/NousResearch/hermes-agent/pull/3390))
+- Self-reschedule reconnect when `start_polling` fails after 502 ([#3268](https://github.com/NousResearch/hermes-agent/pull/3268))
+
+### Discord
+- Stop phantom typing indicator after agent turn completes ([#3003](https://github.com/NousResearch/hermes-agent/pull/3003))
+
+### Slack
+- Send tool call progress messages to correct Slack thread ([#3063](https://github.com/NousResearch/hermes-agent/pull/3063))
+- Scope progress thread fallback to Slack only ([#3488](https://github.com/NousResearch/hermes-agent/pull/3488))
+
+### WhatsApp
+- Download documents, audio, and video media from messages ([#2978](https://github.com/NousResearch/hermes-agent/pull/2978))
+
+### Matrix
+- Add missing Matrix entry in `PLATFORMS` dict ([#3473](https://github.com/NousResearch/hermes-agent/pull/3473))
+- Harden e2ee access-token handling ([#3562](https://github.com/NousResearch/hermes-agent/pull/3562))
+- Add backoff for `SyncError` in sync loop ([#3280](https://github.com/NousResearch/hermes-agent/pull/3280))
+
+### Signal
+- Track SSE keepalive comments as connection activity ([#3316](https://github.com/NousResearch/hermes-agent/pull/3316))
+
+### Email
+- Prevent unbounded growth of `_seen_uids` in EmailAdapter ([#3490](https://github.com/NousResearch/hermes-agent/pull/3490))
+
+### Gateway Core
+- **Config-gated `/verbose` command** for messaging platforms — toggle tool output verbosity from chat ([#3262](https://github.com/NousResearch/hermes-agent/pull/3262))
+- **Background review notifications** delivered to user chat ([#3293](https://github.com/NousResearch/hermes-agent/pull/3293))
+- **Retry transient send failures** and notify user on exhaustion ([#3288](https://github.com/NousResearch/hermes-agent/pull/3288))
+- Recover from hung agents — `/stop` hard-kills session lock ([#3104](https://github.com/NousResearch/hermes-agent/pull/3104))
+- Thread-safe `SessionStore` — protect `_entries` with `threading.Lock` ([#3052](https://github.com/NousResearch/hermes-agent/pull/3052))
+- Fix gateway token double-counting with cached agents — use absolute set instead of increment ([#3306](https://github.com/NousResearch/hermes-agent/pull/3306), [#3317](https://github.com/NousResearch/hermes-agent/pull/3317))
+- Fingerprint full auth token in agent cache signature ([#3247](https://github.com/NousResearch/hermes-agent/pull/3247))
+- Silence background agent terminal output ([#3297](https://github.com/NousResearch/hermes-agent/pull/3297))
+- Include per-platform `ALLOW_ALL` and `SIGNAL_GROUP` in startup allowlist check ([#3313](https://github.com/NousResearch/hermes-agent/pull/3313))
+- Include user-local bin paths in systemd unit PATH ([#3527](https://github.com/NousResearch/hermes-agent/pull/3527))
+- Track background task references in `GatewayRunner` ([#3254](https://github.com/NousResearch/hermes-agent/pull/3254))
+- Add request timeouts to HA, Email, Mattermost, SMS adapters ([#3258](https://github.com/NousResearch/hermes-agent/pull/3258))
+- Add media download retry to Mattermost, Slack, and base cache ([#3323](https://github.com/NousResearch/hermes-agent/pull/3323))
+- Detect virtualenv path instead of hardcoding `venv/` ([#2797](https://github.com/NousResearch/hermes-agent/pull/2797))
+- Use `TERMINAL_CWD` for context file discovery, not process cwd ([untagged commit](https://github.com/NousResearch/hermes-agent))
+- Stop loading hermes repo AGENTS.md into gateway sessions (~10k wasted tokens) ([#2891](https://github.com/NousResearch/hermes-agent/pull/2891))
+
+---
+
+## 🖥️ CLI & User Experience
+
+### Interactive CLI
+- **Configurable busy input mode** + fix `/queue` always working ([#3298](https://github.com/NousResearch/hermes-agent/pull/3298))
+- **Preserve user input on multiline paste** ([#3065](https://github.com/NousResearch/hermes-agent/pull/3065))
+- **Tool generation callback** — streaming "preparing terminal…" updates during tool argument generation ([untagged commit](https://github.com/NousResearch/hermes-agent))
+- Show tool progress for substantive tools, not just "preparing" ([untagged commit](https://github.com/NousResearch/hermes-agent))
+- Buffer reasoning preview chunks and fix duplicate display ([#3013](https://github.com/NousResearch/hermes-agent/pull/3013))
+- Prevent reasoning box from rendering 3x during tool-calling loops ([#3405](https://github.com/NousResearch/hermes-agent/pull/3405))
+- Eliminate "Event loop is closed" / "Press ENTER to continue" during idle — three-layer fix with `neuter_async_httpx_del()`, custom exception handler, and stale client cleanup ([#3398](https://github.com/NousResearch/hermes-agent/pull/3398))
+- Fix status bar shows 26K instead of 260K for token counts with trailing zeros ([#3024](https://github.com/NousResearch/hermes-agent/pull/3024))
+- Fix status bar duplicates and degrades during long sessions ([#3291](https://github.com/NousResearch/hermes-agent/pull/3291))
+- Refresh TUI before background task output to prevent status bar overlap ([#3048](https://github.com/NousResearch/hermes-agent/pull/3048))
+- Suppress KawaiiSpinner animation under `patch_stdout` ([#2994](https://github.com/NousResearch/hermes-agent/pull/2994))
+- Skip KawaiiSpinner when TUI handles tool progress ([#2973](https://github.com/NousResearch/hermes-agent/pull/2973))
+- Guard `isatty()` against closed streams via `_is_tty` property ([#3056](https://github.com/NousResearch/hermes-agent/pull/3056))
+- Ensure single closure of streaming boxes during tool generation ([untagged commit](https://github.com/NousResearch/hermes-agent))
+- Cap context pressure percentage at 100% in display ([#3480](https://github.com/NousResearch/hermes-agent/pull/3480))
+- Clean up HTML error messages in CLI display ([#3069](https://github.com/NousResearch/hermes-agent/pull/3069))
+- Show HTTP status code and 400 body in API error output ([#3096](https://github.com/NousResearch/hermes-agent/pull/3096))
+- Extract useful info from HTML error pages, dump debug on max retries ([untagged commit](https://github.com/NousResearch/hermes-agent))
+- Prevent TypeError on startup when `base_url` is None ([#3068](https://github.com/NousResearch/hermes-agent/pull/3068))
+- Prevent update crash in non-TTY environments ([#3094](https://github.com/NousResearch/hermes-agent/pull/3094))
+- Handle EOFError in sessions delete/prune confirmation prompts ([#3101](https://github.com/NousResearch/hermes-agent/pull/3101))
+- Catch KeyboardInterrupt during `flush_memories` on exit and in exit cleanup handlers ([#3025](https://github.com/NousResearch/hermes-agent/pull/3025), [#3257](https://github.com/NousResearch/hermes-agent/pull/3257))
+- Guard `.strip()` against None values from YAML config ([#3552](https://github.com/NousResearch/hermes-agent/pull/3552))
+- Guard `config.get()` against YAML null values to prevent AttributeError ([#3377](https://github.com/NousResearch/hermes-agent/pull/3377))
+- Store asyncio task references to prevent GC mid-execution ([#3267](https://github.com/NousResearch/hermes-agent/pull/3267))
+
+### Setup & Configuration
+- Use explicit key mapping for returning-user menu dispatch instead of positional index ([#3083](https://github.com/NousResearch/hermes-agent/pull/3083))
+- Use `sys.executable` for pip in update commands to fix PEP 668 ([#3099](https://github.com/NousResearch/hermes-agent/pull/3099))
+- Harden `hermes update` against diverged history, non-main branches, and gateway edge cases ([#3492](https://github.com/NousResearch/hermes-agent/pull/3492))
+- OpenClaw migration overwrites defaults and setup wizard skips imported sections — fixed ([#3282](https://github.com/NousResearch/hermes-agent/pull/3282))
+- Stop recursive AGENTS.md walk, load top-level only ([#3110](https://github.com/NousResearch/hermes-agent/pull/3110))
+- Add macOS Homebrew paths to browser and terminal PATH resolution ([#2713](https://github.com/NousResearch/hermes-agent/pull/2713))
+- YAML boolean handling for `tool_progress` config ([#3300](https://github.com/NousResearch/hermes-agent/pull/3300))
+- Reset default SOUL.md to baseline identity text ([#3159](https://github.com/NousResearch/hermes-agent/pull/3159))
+- Reject relative cwd paths for container terminal backends ([untagged commit](https://github.com/NousResearch/hermes-agent))
+- Add explicit `hermes-api-server` toolset for API server platform ([#3304](https://github.com/NousResearch/hermes-agent/pull/3304))
+- Reorder setup wizard providers — OpenRouter first ([untagged commit](https://github.com/NousResearch/hermes-agent))
+
+---
+
+## 🔧 Tool System
+
+### API Server
+- **Idempotency-Key support**, body size limit, and OpenAI error envelope ([#2903](https://github.com/NousResearch/hermes-agent/pull/2903))
+- Allow Idempotency-Key in CORS headers ([#3530](https://github.com/NousResearch/hermes-agent/pull/3530))
+- Cancel orphaned agent + true interrupt on SSE disconnect ([#3427](https://github.com/NousResearch/hermes-agent/pull/3427))
+- Fix streaming breaks when agent makes tool calls ([#2985](https://github.com/NousResearch/hermes-agent/pull/2985))
+
+### Terminal & File Operations
+- Handle addition-only hunks in V4A patch parser ([#3325](https://github.com/NousResearch/hermes-agent/pull/3325))
+- Exponential backoff for persistent shell polling ([#2996](https://github.com/NousResearch/hermes-agent/pull/2996))
+- Add timeout to subprocess calls in `context_references` ([#3469](https://github.com/NousResearch/hermes-agent/pull/3469))
+
+### Browser & Vision
+- Handle 402 insufficient credits error in vision tool ([#2802](https://github.com/NousResearch/hermes-agent/pull/2802))
+- Fix `browser_vision` ignores `auxiliary.vision.timeout` config ([#2901](https://github.com/NousResearch/hermes-agent/pull/2901))
+- Make browser command timeout configurable via config.yaml ([#2801](https://github.com/NousResearch/hermes-agent/pull/2801))
+
+### MCP
+- MCP toolset resolution for runtime and config ([#3252](https://github.com/NousResearch/hermes-agent/pull/3252))
+- Add MCP tool name collision protection ([#3077](https://github.com/NousResearch/hermes-agent/pull/3077))
+
+### Auxiliary LLM
+- Guard aux LLM calls against None content + reasoning fallback + retry ([#3449](https://github.com/NousResearch/hermes-agent/pull/3449))
+- Catch ImportError from `build_anthropic_client` in vision auto-detection ([#3312](https://github.com/NousResearch/hermes-agent/pull/3312))
+
+### Other Tools
+- Add request timeouts to `send_message_tool` HTTP calls ([#3162](https://github.com/NousResearch/hermes-agent/pull/3162)) by @memosr
+- Auto-repair `jobs.json` with invalid control characters ([#3537](https://github.com/NousResearch/hermes-agent/pull/3537))
+- Enable fine-grained tool streaming for Claude/OpenRouter ([#3497](https://github.com/NousResearch/hermes-agent/pull/3497))
+
+---
+
+## 🧩 Skills Ecosystem
+
+### Skills System
+- **Env var passthrough** for skills and user config — skills can declare environment variables to pass through ([#2807](https://github.com/NousResearch/hermes-agent/pull/2807))
+- Cache skills prompt with shared `skill_utils` module for faster TTFT ([#3421](https://github.com/NousResearch/hermes-agent/pull/3421))
+- Avoid redundant file re-read for skill conditions ([#2992](https://github.com/NousResearch/hermes-agent/pull/2992))
+- Use Git Trees API to prevent silent subdirectory loss during install ([#2995](https://github.com/NousResearch/hermes-agent/pull/2995))
+- Fix skills-sh install for deeply nested repo structures ([#2980](https://github.com/NousResearch/hermes-agent/pull/2980))
+- Handle null metadata in skill frontmatter ([untagged commit](https://github.com/NousResearch/hermes-agent))
+- Preserve trust for skills-sh identifiers + reduce resolution churn ([#3251](https://github.com/NousResearch/hermes-agent/pull/3251))
+- Agent-created skills were incorrectly treated as untrusted community content — fixed ([untagged commit](https://github.com/NousResearch/hermes-agent))
+
+### New Skills
+- **G0DM0D3 godmode jailbreaking skill** + docs ([#3157](https://github.com/NousResearch/hermes-agent/pull/3157))
+- **Docker management skill** added to optional-skills ([#3060](https://github.com/NousResearch/hermes-agent/pull/3060))
+- **OpenClaw migration v2** — 17 new modules, terminal recap for migrating from OpenClaw to Hermes ([#2906](https://github.com/NousResearch/hermes-agent/pull/2906))
+
+---
+
+## 🔒 Security & Reliability
+
+### Security Hardening
+- **SSRF protection** added to `browser_navigate` ([#3058](https://github.com/NousResearch/hermes-agent/pull/3058))
+- **SSRF protection** added to `vision_tools` and `web_tools` (hardened) ([#2679](https://github.com/NousResearch/hermes-agent/pull/2679))
+- **Restrict subagent toolsets** to parent's enabled set ([#3269](https://github.com/NousResearch/hermes-agent/pull/3269))
+- **Prevent zip-slip path traversal** in self-update ([#3250](https://github.com/NousResearch/hermes-agent/pull/3250))
+- **Prevent shell injection** in `_expand_path` via `~user` path suffix ([#2685](https://github.com/NousResearch/hermes-agent/pull/2685))
+- **Normalize input** before dangerous command detection ([#3260](https://github.com/NousResearch/hermes-agent/pull/3260))
+- Make tirith block verdicts approvable instead of hard-blocking ([#3428](https://github.com/NousResearch/hermes-agent/pull/3428))
+- Remove compromised `litellm`/`typer`/`platformdirs` from deps ([#2796](https://github.com/NousResearch/hermes-agent/pull/2796))
+- Pin all dependency version ranges ([#2810](https://github.com/NousResearch/hermes-agent/pull/2810))
+- Regenerate `uv.lock` with hashes, use lockfile in setup ([#2812](https://github.com/NousResearch/hermes-agent/pull/2812))
+- Bump dependencies to fix CVEs + regenerate `uv.lock` ([#3073](https://github.com/NousResearch/hermes-agent/pull/3073))
+- Supply chain audit CI workflow for PR scanning ([#2816](https://github.com/NousResearch/hermes-agent/pull/2816))
+
+### Reliability
+- **SQLite WAL write-lock contention** causing 15-20s TUI freeze — fixed ([#3385](https://github.com/NousResearch/hermes-agent/pull/3385))
+- **SQLite concurrency hardening** + session transcript integrity ([#3249](https://github.com/NousResearch/hermes-agent/pull/3249))
+- Prevent recurring cron job re-fire on gateway crash/restart loop ([#3396](https://github.com/NousResearch/hermes-agent/pull/3396))
+- Mark cron session as ended after job completes ([#2998](https://github.com/NousResearch/hermes-agent/pull/2998))
+
+---
+
+## ⚡ Performance
+
+- **TTFT startup optimizations** — salvaged easy-win startup improvements ([#3395](https://github.com/NousResearch/hermes-agent/pull/3395))
+- Cache skills prompt with shared `skill_utils` module ([#3421](https://github.com/NousResearch/hermes-agent/pull/3421))
+- Avoid redundant file re-read for skill conditions in prompt builder ([#2992](https://github.com/NousResearch/hermes-agent/pull/2992))
+
+---
+
+## 🐛 Notable Bug Fixes
+
+- Fix gateway token double-counting with cached agents ([#3306](https://github.com/NousResearch/hermes-agent/pull/3306), [#3317](https://github.com/NousResearch/hermes-agent/pull/3317))
+- Fix "Event loop is closed" / "Press ENTER to continue" during idle sessions ([#3398](https://github.com/NousResearch/hermes-agent/pull/3398))
+- Fix reasoning box rendering 3x during tool-calling loops ([#3405](https://github.com/NousResearch/hermes-agent/pull/3405))
+- Fix status bar shows 26K instead of 260K for token counts ([#3024](https://github.com/NousResearch/hermes-agent/pull/3024))
+- Fix `/queue` always working regardless of config ([#3298](https://github.com/NousResearch/hermes-agent/pull/3298))
+- Fix phantom Discord typing indicator after agent turn ([#3003](https://github.com/NousResearch/hermes-agent/pull/3003))
+- Fix Slack progress messages appearing in wrong thread ([#3063](https://github.com/NousResearch/hermes-agent/pull/3063))
+- Fix WhatsApp media downloads (documents, audio, video) ([#2978](https://github.com/NousResearch/hermes-agent/pull/2978))
+- Fix Telegram "Message thread not found" killing progress messages ([#3390](https://github.com/NousResearch/hermes-agent/pull/3390))
+- Fix OpenClaw migration overwriting defaults ([#3282](https://github.com/NousResearch/hermes-agent/pull/3282))
+- Fix returning-user setup menu dispatching wrong section ([#3083](https://github.com/NousResearch/hermes-agent/pull/3083))
+- Fix `hermes update` PEP 668 "externally-managed-environment" error ([#3099](https://github.com/NousResearch/hermes-agent/pull/3099))
+- Fix subagents hitting `max_iterations` prematurely via shared budget ([#3004](https://github.com/NousResearch/hermes-agent/pull/3004))
+- Fix YAML boolean handling for `tool_progress` config ([#3300](https://github.com/NousResearch/hermes-agent/pull/3300))
+- Fix `config.get()` crashes on YAML null values ([#3377](https://github.com/NousResearch/hermes-agent/pull/3377))
+- Fix `.strip()` crash on None values from YAML config ([#3552](https://github.com/NousResearch/hermes-agent/pull/3552))
+- Fix hung agents on gateway — `/stop` now hard-kills session lock ([#3104](https://github.com/NousResearch/hermes-agent/pull/3104))
+- Fix `_custom` provider silently remapped to `openrouter` ([#2792](https://github.com/NousResearch/hermes-agent/pull/2792))
+- Fix Matrix missing from `PLATFORMS` dict ([#3473](https://github.com/NousResearch/hermes-agent/pull/3473))
+- Fix Email adapter unbounded `_seen_uids` growth ([#3490](https://github.com/NousResearch/hermes-agent/pull/3490))
+
+---
+
+## 🧪 Testing
+
+- Pin `agent-client-protocol` < 0.9 to handle breaking upstream release ([#3320](https://github.com/NousResearch/hermes-agent/pull/3320))
+- Catch anthropic ImportError in vision auto-detection tests ([#3312](https://github.com/NousResearch/hermes-agent/pull/3312))
+- Update retry-exhaust test for new graceful return behavior ([#3320](https://github.com/NousResearch/hermes-agent/pull/3320))
+- Add regression tests for null metadata frontmatter ([untagged commit](https://github.com/NousResearch/hermes-agent))
+
+---
+
+## 📚 Documentation
+
+- Update all docs for `/model` command overhaul and custom provider support ([#2800](https://github.com/NousResearch/hermes-agent/pull/2800))
+- Fix stale and incorrect documentation across 18 files ([#2805](https://github.com/NousResearch/hermes-agent/pull/2805))
+- Document 9 previously undocumented features ([#2814](https://github.com/NousResearch/hermes-agent/pull/2814))
+- Add missing skills, CLI commands, and messaging env vars to docs ([#2809](https://github.com/NousResearch/hermes-agent/pull/2809))
+- Fix api-server response storage documentation — SQLite, not in-memory ([#2819](https://github.com/NousResearch/hermes-agent/pull/2819))
+- Quote pip install extras to fix zsh glob errors ([#2815](https://github.com/NousResearch/hermes-agent/pull/2815))
+- Unify hooks documentation — add plugin hooks to hooks page, add `session:end` event ([untagged commit](https://github.com/NousResearch/hermes-agent))
+- Clarify two-mode behavior in `session_search` schema description ([untagged commit](https://github.com/NousResearch/hermes-agent))
+- Fix Discord Public Bot setting for Discord-provided invite link ([#3519](https://github.com/NousResearch/hermes-agent/pull/3519)) by @mehmoodosman
+- Revise v0.4.0 changelog — fix feature attribution, reorder sections ([untagged commit](https://github.com/NousResearch/hermes-agent))
+
+---
+
+## 👥 Contributors
+
+### Core
+- **@teknium1** — 157 PRs covering the full scope of this release
+
+### Community Contributors
+- **@alt-glitch** (Siddharth Balyan) — 2 PRs: Nix flake with uv2nix build, NixOS module, and persistent container mode ([#20](https://github.com/NousResearch/hermes-agent/pull/20)); auto-generated config keys and suffix PATHs for Nix builds ([#3061](https://github.com/NousResearch/hermes-agent/pull/3061), [#3274](https://github.com/NousResearch/hermes-agent/pull/3274))
+- **@ctlst** — 1 PR: Prevent AsyncOpenAI/httpx cross-loop deadlock in gateway mode ([#2701](https://github.com/NousResearch/hermes-agent/pull/2701))
+- **@memosr** (memosr.eth) — 1 PR: Add request timeouts to `send_message_tool` HTTP calls ([#3162](https://github.com/NousResearch/hermes-agent/pull/3162))
+- **@mehmoodosman** (Osman Mehmood) — 1 PR: Fix Discord docs for Public Bot setting ([#3519](https://github.com/NousResearch/hermes-agent/pull/3519))
+
+### All Contributors
+@alt-glitch, @ctlst, @mehmoodosman, @memosr, @teknium1
+
+---
+
+**Full Changelog**: [v2026.3.23...v2026.3.28](https://github.com/NousResearch/hermes-agent/compare/v2026.3.23...v2026.3.28)

RELEASE_v0.6.0.md

@@ -0,0 +1,249 @@
+# Hermes Agent v0.6.0 (v2026.3.30)
+
+**Release Date:** March 30, 2026
+
+> The multi-instance release — Profiles for running isolated agent instances, MCP server mode, Docker container, fallback provider chains, two new messaging platforms (Feishu/Lark and WeCom), Telegram webhook mode, Slack multi-workspace OAuth, 95 PRs and 16 resolved issues in 2 days.
+
+---
+
+## ✨ Highlights
+
+- **Profiles — Multi-Instance Hermes** — Run multiple isolated Hermes instances from the same installation. Each profile gets its own config, memory, sessions, skills, and gateway service. Create with `hermes profile create`, switch with `hermes -p <name>`, export/import for sharing. Full token-lock isolation prevents two profiles from using the same bot credential. ([#3681](https://github.com/NousResearch/hermes-agent/pull/3681))
+
+- **MCP Server Mode** — Expose Hermes conversations and sessions to any MCP-compatible client (Claude Desktop, Cursor, VS Code, etc.) via `hermes mcp serve`. Browse conversations, read messages, search across sessions, and manage attachments — all through the Model Context Protocol. Supports both stdio and Streamable HTTP transports. ([#3795](https://github.com/NousResearch/hermes-agent/pull/3795))
+
+- **Docker Container** — Official Dockerfile for running Hermes Agent in a container. Supports both CLI and gateway modes with volume-mounted config. ([#3668](https://github.com/NousResearch/hermes-agent/pull/3668), closes [#850](https://github.com/NousResearch/hermes-agent/issues/850))
+
+- **Ordered Fallback Provider Chain** — Configure multiple inference providers with automatic failover. When your primary provider returns errors or is unreachable, Hermes automatically tries the next provider in the chain. Configure via `fallback_providers` in config.yaml. ([#3813](https://github.com/NousResearch/hermes-agent/pull/3813), closes [#1734](https://github.com/NousResearch/hermes-agent/issues/1734))
+
+- **Feishu/Lark Platform Support** — Full gateway adapter for Feishu (飞书) and Lark with event subscriptions, message cards, group chat, image/file attachments, and interactive card callbacks. ([#3799](https://github.com/NousResearch/hermes-agent/pull/3799), [#3817](https://github.com/NousResearch/hermes-agent/pull/3817), closes [#1788](https://github.com/NousResearch/hermes-agent/issues/1788))
+
+- **WeCom (Enterprise WeChat) Platform Support** — New gateway adapter for WeCom (企业微信) with text/image/voice messages, group chats, and callback verification. ([#3847](https://github.com/NousResearch/hermes-agent/pull/3847))
+
+- **Slack Multi-Workspace OAuth** — Connect a single Hermes gateway to multiple Slack workspaces via OAuth token file. Each workspace gets its own bot token, resolved dynamically per incoming event. ([#3903](https://github.com/NousResearch/hermes-agent/pull/3903))
+
+- **Telegram Webhook Mode & Group Controls** — Run the Telegram adapter in webhook mode as an alternative to polling — faster response times and better for production deployments behind a reverse proxy. New group mention gating controls when the bot responds: always, only when @mentioned, or via regex triggers. ([#3880](https://github.com/NousResearch/hermes-agent/pull/3880), [#3870](https://github.com/NousResearch/hermes-agent/pull/3870))
+
+- **Exa Search Backend** — Add Exa as an alternative web search and content extraction backend alongside Firecrawl and DuckDuckGo. Set `EXA_API_KEY` and configure as preferred backend. ([#3648](https://github.com/NousResearch/hermes-agent/pull/3648))
+
+- **Skills & Credentials on Remote Backends** — Mount skill directories and credential files into Modal and Docker containers, so remote terminal sessions have access to the same skills and secrets as local execution. ([#3890](https://github.com/NousResearch/hermes-agent/pull/3890), [#3671](https://github.com/NousResearch/hermes-agent/pull/3671), closes [#3665](https://github.com/NousResearch/hermes-agent/issues/3665), [#3433](https://github.com/NousResearch/hermes-agent/issues/3433))
+
+---
+
+## 🏗️ Core Agent & Architecture
+
+### Provider & Model Support
+- **Ordered fallback provider chain** — automatic failover across multiple configured providers ([#3813](https://github.com/NousResearch/hermes-agent/pull/3813))
+- **Fix api_mode on provider switch** — switching providers via `hermes model` now correctly clears stale `api_mode` instead of hardcoding `chat_completions`, fixing 404s for providers with Anthropic-compatible endpoints ([#3726](https://github.com/NousResearch/hermes-agent/pull/3726), [#3857](https://github.com/NousResearch/hermes-agent/pull/3857), closes [#3685](https://github.com/NousResearch/hermes-agent/issues/3685))
+- **Stop silent OpenRouter fallback** — when no provider is configured, Hermes now raises a clear error instead of silently routing to OpenRouter ([#3807](https://github.com/NousResearch/hermes-agent/pull/3807), [#3862](https://github.com/NousResearch/hermes-agent/pull/3862))
+- **Gemini 3.1 preview models** — added to OpenRouter and Nous Portal catalogs ([#3803](https://github.com/NousResearch/hermes-agent/pull/3803), closes [#3753](https://github.com/NousResearch/hermes-agent/issues/3753))
+- **Gemini direct API context length** — full context length resolution for direct Google AI endpoints ([#3876](https://github.com/NousResearch/hermes-agent/pull/3876))
+- **gpt-5.4-mini** added to Codex fallback catalog ([#3855](https://github.com/NousResearch/hermes-agent/pull/3855))
+- **Curated model lists preferred** over live API probe when the probe returns fewer models ([#3856](https://github.com/NousResearch/hermes-agent/pull/3856), [#3867](https://github.com/NousResearch/hermes-agent/pull/3867))
+- **User-friendly 429 rate limit messages** with Retry-After countdown ([#3809](https://github.com/NousResearch/hermes-agent/pull/3809))
+- **Auxiliary client placeholder key** for local servers without auth requirements ([#3842](https://github.com/NousResearch/hermes-agent/pull/3842))
+- **INFO-level logging** for auxiliary provider resolution ([#3866](https://github.com/NousResearch/hermes-agent/pull/3866))
+
+### Agent Loop & Conversation
+- **Subagent status reporting** — reports `completed` status when summary exists instead of generic failure ([#3829](https://github.com/NousResearch/hermes-agent/pull/3829))
+- **Session log file updated during compression** — prevents stale file references after context compression ([#3835](https://github.com/NousResearch/hermes-agent/pull/3835))
+- **Omit empty tools param** — sends no `tools` parameter when empty instead of `None`, fixing compatibility with strict providers ([#3820](https://github.com/NousResearch/hermes-agent/pull/3820))
+
+### Profiles & Multi-Instance
+- **Profiles system** — `hermes profile create/list/switch/delete/export/import/rename`. Each profile gets isolated HERMES_HOME, gateway service, CLI wrapper. Token locks prevent credential collisions. Tab completion for profile names. ([#3681](https://github.com/NousResearch/hermes-agent/pull/3681))
+- **Profile-aware display paths** — all user-facing `~/.hermes` paths replaced with `display_hermes_home()` to show the correct profile directory ([#3623](https://github.com/NousResearch/hermes-agent/pull/3623))
+- **Lazy display_hermes_home imports** — prevents `ImportError` during `hermes update` when modules cache stale bytecode ([#3776](https://github.com/NousResearch/hermes-agent/pull/3776))
+- **HERMES_HOME for protected paths** — `.env` write-deny path now respects HERMES_HOME instead of hardcoded `~/.hermes` ([#3840](https://github.com/NousResearch/hermes-agent/pull/3840))
+
+---
+
+## 📱 Messaging Platforms (Gateway)
+
+### New Platforms
+- **Feishu/Lark** — Full adapter with event subscriptions, message cards, group chat, image/file attachments, interactive card callbacks ([#3799](https://github.com/NousResearch/hermes-agent/pull/3799), [#3817](https://github.com/NousResearch/hermes-agent/pull/3817))
+- **WeCom (Enterprise WeChat)** — Text/image/voice messages, group chats, callback verification ([#3847](https://github.com/NousResearch/hermes-agent/pull/3847))
+
+### Telegram
+- **Webhook mode** — run as webhook endpoint instead of polling for production deployments ([#3880](https://github.com/NousResearch/hermes-agent/pull/3880))
+- **Group mention gating & regex triggers** — configurable bot response behavior in groups: always, @mention-only, or regex-matched ([#3870](https://github.com/NousResearch/hermes-agent/pull/3870))
+- **Gracefully handle deleted reply targets** — no more crashes when the message being replied to was deleted ([#3858](https://github.com/NousResearch/hermes-agent/pull/3858), closes [#3229](https://github.com/NousResearch/hermes-agent/issues/3229))
+
+### Discord
+- **Message processing reactions** — adds a reaction emoji while processing and removes it when done, giving visual feedback in channels ([#3871](https://github.com/NousResearch/hermes-agent/pull/3871))
+- **DISCORD_IGNORE_NO_MENTION** — skip messages that @mention other users/bots but not Hermes ([#3640](https://github.com/NousResearch/hermes-agent/pull/3640))
+- **Clean up deferred "thinking..."** — properly removes the "thinking..." indicator after slash commands complete ([#3674](https://github.com/NousResearch/hermes-agent/pull/3674), closes [#3595](https://github.com/NousResearch/hermes-agent/issues/3595))
+
+### Slack
+- **Multi-workspace OAuth** — connect to multiple Slack workspaces from a single gateway via OAuth token file ([#3903](https://github.com/NousResearch/hermes-agent/pull/3903))
+
+### WhatsApp
+- **Persistent aiohttp session** — reuse HTTP sessions across requests instead of creating new ones per message ([#3818](https://github.com/NousResearch/hermes-agent/pull/3818))
+- **LID↔phone alias resolution** — correctly match Linked ID and phone number formats in allowlists ([#3830](https://github.com/NousResearch/hermes-agent/pull/3830))
+- **Skip reply prefix in bot mode** — cleaner message formatting when running as a WhatsApp bot ([#3931](https://github.com/NousResearch/hermes-agent/pull/3931))
+
+### Matrix
+- **Native voice messages via MSC3245** — send voice messages as proper Matrix voice events instead of file attachments ([#3877](https://github.com/NousResearch/hermes-agent/pull/3877))
+
+### Mattermost
+- **Configurable mention behavior** — respond to messages without requiring @mention ([#3664](https://github.com/NousResearch/hermes-agent/pull/3664))
+
+### Signal
+- **URL-encode phone numbers** and correct attachment RPC parameter — fixes delivery failures with certain phone number formats ([#3670](https://github.com/NousResearch/hermes-agent/pull/3670)) — @kshitijk4poor
+
+### Email
+- **Close SMTP/IMAP connections on failure** — prevents connection leaks during error scenarios ([#3804](https://github.com/NousResearch/hermes-agent/pull/3804))
+
+### Gateway Core
+- **Atomic config writes** — use atomic file writes for config.yaml to prevent data loss during crashes ([#3800](https://github.com/NousResearch/hermes-agent/pull/3800))
+- **Home channel env overrides** — apply environment variable overrides for home channels consistently ([#3796](https://github.com/NousResearch/hermes-agent/pull/3796), [#3808](https://github.com/NousResearch/hermes-agent/pull/3808))
+- **Replace print() with logger** — BasePlatformAdapter now uses proper logging instead of print statements ([#3669](https://github.com/NousResearch/hermes-agent/pull/3669))
+- **Cron delivery labels** — resolve human-friendly delivery labels via channel directory ([#3860](https://github.com/NousResearch/hermes-agent/pull/3860), closes [#1945](https://github.com/NousResearch/hermes-agent/issues/1945))
+- **Cron [SILENT] tightening** — prevent agents from prefixing reports with [SILENT] to suppress delivery ([#3901](https://github.com/NousResearch/hermes-agent/pull/3901))
+- **Background task media delivery** and vision download timeout fixes ([#3919](https://github.com/NousResearch/hermes-agent/pull/3919))
+- **Boot-md hook** — example built-in hook to run a BOOT.md file on gateway startup ([#3733](https://github.com/NousResearch/hermes-agent/pull/3733))
+
+---
+
+## 🖥️ CLI & User Experience
+
+### Interactive CLI
+- **Configurable tool preview length** — show full file paths by default instead of truncating at 40 chars ([#3841](https://github.com/NousResearch/hermes-agent/pull/3841))
+- **Tool token context display** — `hermes tools` checklist now shows estimated token cost per toolset ([#3805](https://github.com/NousResearch/hermes-agent/pull/3805))
+- **/bg spinner TUI fix** — route background task spinner through the TUI widget to prevent status bar collision ([#3643](https://github.com/NousResearch/hermes-agent/pull/3643))
+- **Prevent status bar wrapping** into duplicate rows ([#3883](https://github.com/NousResearch/hermes-agent/pull/3883)) — @kshitijk4poor
+- **Handle closed stdout ValueError** in safe print paths — fixes crashes when stdout is closed during gateway thread shutdown ([#3843](https://github.com/NousResearch/hermes-agent/pull/3843), closes [#3534](https://github.com/NousResearch/hermes-agent/issues/3534))
+- **Remove input() from /tools disable** — eliminates freeze in terminal when disabling tools ([#3918](https://github.com/NousResearch/hermes-agent/pull/3918))
+- **TTY guard for interactive CLI commands** — prevent CPU spin when launched without a terminal ([#3933](https://github.com/NousResearch/hermes-agent/pull/3933))
+- **Argparse entrypoint** — use argparse in the top-level launcher for cleaner error handling ([#3874](https://github.com/NousResearch/hermes-agent/pull/3874))
+- **Lazy-initialized tools show yellow** in banner instead of red, reducing false alarm about "missing" tools ([#3822](https://github.com/NousResearch/hermes-agent/pull/3822))
+- **Honcho tools shown in banner** when configured ([#3810](https://github.com/NousResearch/hermes-agent/pull/3810))
+
+### Setup & Configuration
+- **Auto-install matrix-nio** during `hermes setup` when Matrix is selected ([#3802](https://github.com/NousResearch/hermes-agent/pull/3802), [#3873](https://github.com/NousResearch/hermes-agent/pull/3873))
+- **Session export stdout support** — export sessions to stdout with `-` for piping ([#3641](https://github.com/NousResearch/hermes-agent/pull/3641), closes [#3609](https://github.com/NousResearch/hermes-agent/issues/3609))
+- **Configurable approval timeouts** — set how long dangerous command approval prompts wait before auto-denying ([#3886](https://github.com/NousResearch/hermes-agent/pull/3886), closes [#3765](https://github.com/NousResearch/hermes-agent/issues/3765))
+- **Clear __pycache__ during update** — prevents stale bytecode ImportError after `hermes update` ([#3819](https://github.com/NousResearch/hermes-agent/pull/3819))
+
+---
+
+## 🔧 Tool System
+
+### MCP
+- **MCP Server Mode** — `hermes mcp serve` exposes conversations, sessions, and attachments to MCP clients via stdio or Streamable HTTP ([#3795](https://github.com/NousResearch/hermes-agent/pull/3795))
+- **Dynamic tool discovery** — respond to `notifications/tools/list_changed` events to pick up new tools from MCP servers without reconnecting ([#3812](https://github.com/NousResearch/hermes-agent/pull/3812))
+- **Non-deprecated HTTP transport** — switched from `sse_client` to `streamable_http_client` ([#3646](https://github.com/NousResearch/hermes-agent/pull/3646))
+
+### Web Tools
+- **Exa search backend** — alternative to Firecrawl and DuckDuckGo for web search and extraction ([#3648](https://github.com/NousResearch/hermes-agent/pull/3648))
+
+### Browser
+- **Guard against None LLM responses** in browser snapshot and vision tools ([#3642](https://github.com/NousResearch/hermes-agent/pull/3642))
+
+### Terminal & Remote Backends
+- **Mount skill directories** into Modal and Docker containers ([#3890](https://github.com/NousResearch/hermes-agent/pull/3890))
+- **Mount credential files** into remote backends with mtime+size caching ([#3671](https://github.com/NousResearch/hermes-agent/pull/3671))
+- **Preserve partial output** when commands time out instead of losing everything ([#3868](https://github.com/NousResearch/hermes-agent/pull/3868))
+- **Stop marking persisted env vars as missing** on remote backends ([#3650](https://github.com/NousResearch/hermes-agent/pull/3650))
+
+### Audio
+- **.aac format support** in transcription tool ([#3865](https://github.com/NousResearch/hermes-agent/pull/3865), closes [#1963](https://github.com/NousResearch/hermes-agent/issues/1963))
+- **Audio download retry** — retry logic for `cache_audio_from_url` matching the existing image download pattern ([#3401](https://github.com/NousResearch/hermes-agent/pull/3401)) — @binhnt92
+
+### Vision
+- **Reject non-image files** and enforce website-only policy for vision analysis ([#3845](https://github.com/NousResearch/hermes-agent/pull/3845))
+
+### Tool Schema
+- **Ensure name field** always present in tool definitions, fixing `KeyError: 'name'` crashes ([#3811](https://github.com/NousResearch/hermes-agent/pull/3811), closes [#3729](https://github.com/NousResearch/hermes-agent/issues/3729))
+
+### ACP (Editor Integration)
+- **Complete session management surface** for VS Code/Zed/JetBrains clients — proper task lifecycle, cancel support, session persistence ([#3675](https://github.com/NousResearch/hermes-agent/pull/3675))
+
+---
+
+## 🧩 Skills & Plugins
+
+### Skills System
+- **External skill directories** — configure additional skill directories via `skills.external_dirs` in config.yaml ([#3678](https://github.com/NousResearch/hermes-agent/pull/3678))
+- **Category path traversal blocked** — prevents `../` attacks in skill category names ([#3844](https://github.com/NousResearch/hermes-agent/pull/3844))
+- **parallel-cli moved to optional-skills** — reduces default skill footprint ([#3673](https://github.com/NousResearch/hermes-agent/pull/3673)) — @kshitijk4poor
+
+### New Skills
+- **memento-flashcards** — spaced repetition flashcard system ([#3827](https://github.com/NousResearch/hermes-agent/pull/3827))
+- **songwriting-and-ai-music** — songwriting craft and AI music generation prompts ([#3834](https://github.com/NousResearch/hermes-agent/pull/3834))
+- **SiYuan Note** — integration with SiYuan note-taking app ([#3742](https://github.com/NousResearch/hermes-agent/pull/3742))
+- **Scrapling** — web scraping skill using Scrapling library ([#3742](https://github.com/NousResearch/hermes-agent/pull/3742))
+- **one-three-one-rule** — communication framework skill ([#3797](https://github.com/NousResearch/hermes-agent/pull/3797))
+
+### Plugin System
+- **Plugin enable/disable commands** — `hermes plugins enable/disable <name>` for managing plugin state without removing them ([#3747](https://github.com/NousResearch/hermes-agent/pull/3747))
+- **Plugin message injection** — plugins can now inject messages into the conversation stream on behalf of the user via `ctx.inject_message()` ([#3778](https://github.com/NousResearch/hermes-agent/pull/3778)) — @winglian
+- **Honcho self-hosted support** — allow local Honcho instances without requiring an API key ([#3644](https://github.com/NousResearch/hermes-agent/pull/3644))
+
+---
+
+## 🔒 Security & Reliability
+
+### Security Hardening
+- **Hardened dangerous command detection** — expanded pattern matching for risky shell commands and added file tool path guards for sensitive locations (`/etc/`, `/boot/`, docker.sock) ([#3872](https://github.com/NousResearch/hermes-agent/pull/3872))
+- **Sensitive path write checks** in approval system — catch writes to system config files through file tools, not just terminal ([#3859](https://github.com/NousResearch/hermes-agent/pull/3859))
+- **Secret redaction expansion** — now covers ElevenLabs, Tavily, and Exa API keys ([#3920](https://github.com/NousResearch/hermes-agent/pull/3920))
+- **Vision file rejection** — reject non-image files passed to vision analysis to prevent information disclosure ([#3845](https://github.com/NousResearch/hermes-agent/pull/3845))
+- **Category path traversal blocking** — prevent directory traversal in skill category names ([#3844](https://github.com/NousResearch/hermes-agent/pull/3844))
+
+### Reliability
+- **Atomic config.yaml writes** — prevent data loss during gateway crashes ([#3800](https://github.com/NousResearch/hermes-agent/pull/3800))
+- **Clear __pycache__ on update** — prevent stale bytecode from causing ImportError after updates ([#3819](https://github.com/NousResearch/hermes-agent/pull/3819))
+- **Lazy imports for update safety** — prevent ImportError chains during `hermes update` when modules reference new functions ([#3776](https://github.com/NousResearch/hermes-agent/pull/3776))
+- **Restore terminalbench2 from patch corruption** — recovered file damaged by patch tool's secret redaction ([#3801](https://github.com/NousResearch/hermes-agent/pull/3801))
+- **Terminal timeout preserves partial output** — no more lost command output on timeout ([#3868](https://github.com/NousResearch/hermes-agent/pull/3868))
+
+---
+
+## 🐛 Notable Bug Fixes
+
+- **OpenClaw migration model config overwrite** — migration no longer overwrites model config dict with a string ([#3924](https://github.com/NousResearch/hermes-agent/pull/3924)) — @0xbyt4
+- **OpenClaw migration expanded** — covers full data footprint including sessions, cron, memory ([#3869](https://github.com/NousResearch/hermes-agent/pull/3869))
+- **Telegram deleted reply targets** — gracefully handle replies to deleted messages instead of crashing ([#3858](https://github.com/NousResearch/hermes-agent/pull/3858))
+- **Discord "thinking..." persistence** — properly cleans up deferred response indicators ([#3674](https://github.com/NousResearch/hermes-agent/pull/3674))
+- **WhatsApp LID↔phone aliases** — fixes allowlist matching failures with Linked ID format ([#3830](https://github.com/NousResearch/hermes-agent/pull/3830))
+- **Signal URL-encoded phone numbers** — fixes delivery failures with certain formats ([#3670](https://github.com/NousResearch/hermes-agent/pull/3670))
+- **Email connection leaks** — properly close SMTP/IMAP connections on error ([#3804](https://github.com/NousResearch/hermes-agent/pull/3804))
+- **_safe_print ValueError** — no more gateway thread crashes on closed stdout ([#3843](https://github.com/NousResearch/hermes-agent/pull/3843))
+- **Tool schema KeyError 'name'** — ensure name field always present in tool definitions ([#3811](https://github.com/NousResearch/hermes-agent/pull/3811))
+- **api_mode stale on provider switch** — correctly clear when switching providers via `hermes model` ([#3857](https://github.com/NousResearch/hermes-agent/pull/3857))
+
+---
+
+## 🧪 Testing
+
+- Resolved 10+ CI failures across hooks, tiktoken, plugins, and skill tests ([#3848](https://github.com/NousResearch/hermes-agent/pull/3848), [#3721](https://github.com/NousResearch/hermes-agent/pull/3721), [#3936](https://github.com/NousResearch/hermes-agent/pull/3936))
+
+---
+
+## 📚 Documentation
+
+- **Comprehensive OpenClaw migration guide** — step-by-step guide for migrating from OpenClaw/Claw3D to Hermes Agent ([#3864](https://github.com/NousResearch/hermes-agent/pull/3864), [#3900](https://github.com/NousResearch/hermes-agent/pull/3900))
+- **Credential file passthrough docs** — document how to forward credential files and env vars to remote backends ([#3677](https://github.com/NousResearch/hermes-agent/pull/3677))
+- **DuckDuckGo requirements clarified** — note runtime dependency on duckduckgo-search package ([#3680](https://github.com/NousResearch/hermes-agent/pull/3680))
+- **Skills catalog updated** — added red-teaming category and optional skills listing ([#3745](https://github.com/NousResearch/hermes-agent/pull/3745))
+- **Feishu docs MDX fix** — escape angle-bracket URLs that break Docusaurus build ([#3902](https://github.com/NousResearch/hermes-agent/pull/3902))
+
+---
+
+## 👥 Contributors
+
+### Core
+- **@teknium1** — 90 PRs across all subsystems
+
+### Community Contributors
+- **@kshitijk4poor** — 3 PRs: Signal phone number fix ([#3670](https://github.com/NousResearch/hermes-agent/pull/3670)), parallel-cli to optional-skills ([#3673](https://github.com/NousResearch/hermes-agent/pull/3673)), status bar wrapping fix ([#3883](https://github.com/NousResearch/hermes-agent/pull/3883))
+- **@winglian** — 1 PR: Plugin message injection interface ([#3778](https://github.com/NousResearch/hermes-agent/pull/3778))
+- **@binhnt92** — 1 PR: Audio download retry logic ([#3401](https://github.com/NousResearch/hermes-agent/pull/3401))
+- **@0xbyt4** — 1 PR: OpenClaw migration model config fix ([#3924](https://github.com/NousResearch/hermes-agent/pull/3924))
+
+### Issues Resolved from Community
+@Material-Scientist ([#850](https://github.com/NousResearch/hermes-agent/issues/850)), @hanxu98121 ([#1734](https://github.com/NousResearch/hermes-agent/issues/1734)), @penwyp ([#1788](https://github.com/NousResearch/hermes-agent/issues/1788)), @dan-and ([#1945](https://github.com/NousResearch/hermes-agent/issues/1945)), @AdrianScott ([#1963](https://github.com/NousResearch/hermes-agent/issues/1963)), @clawdbot47 ([#3229](https://github.com/NousResearch/hermes-agent/issues/3229)), @alanfwilliams ([#3404](https://github.com/NousResearch/hermes-agent/issues/3404)), @kentimsit ([#3433](https://github.com/NousResearch/hermes-agent/issues/3433)), @hayka-pacha ([#3534](https://github.com/NousResearch/hermes-agent/issues/3534)), @primmer ([#3595](https://github.com/NousResearch/hermes-agent/issues/3595)), @dagelf ([#3609](https://github.com/NousResearch/hermes-agent/issues/3609)), @HenkDz ([#3685](https://github.com/NousResearch/hermes-agent/issues/3685)), @tmdgusya ([#3729](https://github.com/NousResearch/hermes-agent/issues/3729)), @TypQxQ ([#3753](https://github.com/NousResearch/hermes-agent/issues/3753)), @acsezen ([#3765](https://github.com/NousResearch/hermes-agent/issues/3765))
+
+---
+
+**Full Changelog**: [v2026.3.28...v2026.3.30](https://github.com/NousResearch/hermes-agent/compare/v2026.3.28...v2026.3.30)

acp_adapter/__init__.py

@@ -0,0 +1 @@
+"""ACP (Agent Communication Protocol) adapter for hermes-agent."""

acp_adapter/__main__.py

@@ -0,0 +1,5 @@
+"""Allow running the ACP adapter as ``python -m acp_adapter``."""
+
+from .entry import main
+
+main()

acp_adapter/auth.py

@@ -0,0 +1,24 @@
+"""ACP auth helpers — detect the currently configured Hermes provider."""
+
+from __future__ import annotations
+
+from typing import Optional
+
+
+def detect_provider() -> Optional[str]:
+    """Resolve the active Hermes runtime provider, or None if unavailable."""
+    try:
+        from hermes_cli.runtime_provider import resolve_runtime_provider
+        runtime = resolve_runtime_provider()
+        api_key = runtime.get("api_key")
+        provider = runtime.get("provider")
+        if isinstance(api_key, str) and api_key.strip() and isinstance(provider, str) and provider.strip():
+            return provider.strip().lower()
+    except Exception:
+        return None
+    return None
+
+
+def has_provider() -> bool:
+    """Return True if Hermes can resolve any runtime provider credentials."""
+    return detect_provider() is not None

acp_adapter/entry.py

@@ -0,0 +1,86 @@
+"""CLI entry point for the hermes-agent ACP adapter.
+
+Loads environment variables from ``~/.hermes/.env``, configures logging
+to write to stderr (so stdout is reserved for ACP JSON-RPC transport),
+and starts the ACP agent server.
+
+Usage::
+
+    python -m acp_adapter.entry
+    # or
+    hermes acp
+    # or
+    hermes-acp
+"""
+
+import asyncio
+import logging
+import os
+import sys
+from pathlib import Path
+from hermes_constants import get_hermes_home
+
+
+def _setup_logging() -> None:
+    """Route all logging to stderr so stdout stays clean for ACP stdio."""
+    handler = logging.StreamHandler(sys.stderr)
+    handler.setFormatter(
+        logging.Formatter(
+            "%(asctime)s [%(levelname)s] %(name)s: %(message)s",
+            datefmt="%Y-%m-%d %H:%M:%S",
+        )
+    )
+    root = logging.getLogger()
+    root.handlers.clear()
+    root.addHandler(handler)
+    root.setLevel(logging.INFO)
+
+    # Quiet down noisy libraries
+    logging.getLogger("httpx").setLevel(logging.WARNING)
+    logging.getLogger("httpcore").setLevel(logging.WARNING)
+    logging.getLogger("openai").setLevel(logging.WARNING)
+
+
+def _load_env() -> None:
+    """Load .env from HERMES_HOME (default ``~/.hermes``)."""
+    from hermes_cli.env_loader import load_hermes_dotenv
+
+    hermes_home = get_hermes_home()
+    loaded = load_hermes_dotenv(hermes_home=hermes_home)
+    if loaded:
+        for env_file in loaded:
+            logging.getLogger(__name__).info("Loaded env from %s", env_file)
+    else:
+        logging.getLogger(__name__).info(
+            "No .env found at %s, using system env", hermes_home / ".env"
+        )
+
+
+def main() -> None:
+    """Entry point: load env, configure logging, run the ACP agent."""
+    _setup_logging()
+    _load_env()
+
+    logger = logging.getLogger(__name__)
+    logger.info("Starting hermes-agent ACP adapter")
+
+    # Ensure the project root is on sys.path so ``from run_agent import AIAgent`` works
+    project_root = str(Path(__file__).resolve().parent.parent)
+    if project_root not in sys.path:
+        sys.path.insert(0, project_root)
+
+    import acp
+    from .server import HermesACPAgent
+
+    agent = HermesACPAgent()
+    try:
+        asyncio.run(acp.run_agent(agent, use_unstable_protocol=True))
+    except KeyboardInterrupt:
+        logger.info("Shutting down (KeyboardInterrupt)")
+    except Exception:
+        logger.exception("ACP agent crashed")
+        sys.exit(1)
+
+
+if __name__ == "__main__":
+    main()

acp_adapter/events.py

@@ -0,0 +1,171 @@
+"""Callback factories for bridging AIAgent events to ACP notifications.
+
+Each factory returns a callable with the signature that AIAgent expects
+for its callbacks. Internally, the callbacks push ACP session updates
+to the client via ``conn.session_update()`` using
+``asyncio.run_coroutine_threadsafe()`` (since AIAgent runs in a worker
+thread while the event loop lives on the main thread).
+"""
+
+import asyncio
+import json
+import logging
+from collections import deque
+from typing import Any, Callable, Deque, Dict
+
+import acp
+
+from .tools import (
+    build_tool_complete,
+    build_tool_start,
+    make_tool_call_id,
+)
+
+logger = logging.getLogger(__name__)
+
+
+def _send_update(
+    conn: acp.Client,
+    session_id: str,
+    loop: asyncio.AbstractEventLoop,
+    update: Any,
+) -> None:
+    """Fire-and-forget an ACP session update from a worker thread."""
+    try:
+        future = asyncio.run_coroutine_threadsafe(
+            conn.session_update(session_id, update), loop
+        )
+        future.result(timeout=5)
+    except Exception:
+        logger.debug("Failed to send ACP update", exc_info=True)
+
+
+# ------------------------------------------------------------------
+# Tool progress callback
+# ------------------------------------------------------------------
+
+def make_tool_progress_cb(
+    conn: acp.Client,
+    session_id: str,
+    loop: asyncio.AbstractEventLoop,
+    tool_call_ids: Dict[str, Deque[str]],
+) -> Callable:
+    """Create a ``tool_progress_callback`` for AIAgent.
+
+    Signature expected by AIAgent::
+
+        tool_progress_callback(name: str, preview: str, args: dict)
+
+    Emits ``ToolCallStart`` for each tool invocation and tracks IDs in a FIFO
+    queue per tool name so duplicate/parallel same-name calls still complete
+    against the correct ACP tool call.
+    """
+
+    def _tool_progress(name: str, preview: str, args: Any = None) -> None:
+        if isinstance(args, str):
+            try:
+                args = json.loads(args)
+            except (json.JSONDecodeError, TypeError):
+                args = {"raw": args}
+        if not isinstance(args, dict):
+            args = {}
+
+        tc_id = make_tool_call_id()
+        queue = tool_call_ids.get(name)
+        if queue is None:
+            queue = deque()
+            tool_call_ids[name] = queue
+        elif isinstance(queue, str):
+            queue = deque([queue])
+            tool_call_ids[name] = queue
+        queue.append(tc_id)
+
+        update = build_tool_start(tc_id, name, args)
+        _send_update(conn, session_id, loop, update)
+
+    return _tool_progress
+
+
+# ------------------------------------------------------------------
+# Thinking callback
+# ------------------------------------------------------------------
+
+def make_thinking_cb(
+    conn: acp.Client,
+    session_id: str,
+    loop: asyncio.AbstractEventLoop,
+) -> Callable:
+    """Create a ``thinking_callback`` for AIAgent."""
+
+    def _thinking(text: str) -> None:
+        if not text:
+            return
+        update = acp.update_agent_thought_text(text)
+        _send_update(conn, session_id, loop, update)
+
+    return _thinking
+
+
+# ------------------------------------------------------------------
+# Step callback
+# ------------------------------------------------------------------
+
+def make_step_cb(
+    conn: acp.Client,
+    session_id: str,
+    loop: asyncio.AbstractEventLoop,
+    tool_call_ids: Dict[str, Deque[str]],
+) -> Callable:
+    """Create a ``step_callback`` for AIAgent.
+
+    Signature expected by AIAgent::
+
+        step_callback(api_call_count: int, prev_tools: list)
+    """
+
+    def _step(api_call_count: int, prev_tools: Any = None) -> None:
+        if prev_tools and isinstance(prev_tools, list):
+            for tool_info in prev_tools:
+                tool_name = None
+                result = None
+
+                if isinstance(tool_info, dict):
+                    tool_name = tool_info.get("name") or tool_info.get("function_name")
+                    result = tool_info.get("result") or tool_info.get("output")
+                elif isinstance(tool_info, str):
+                    tool_name = tool_info
+
+                queue = tool_call_ids.get(tool_name or "")
+                if isinstance(queue, str):
+                    queue = deque([queue])
+                    tool_call_ids[tool_name] = queue
+                if tool_name and queue:
+                    tc_id = queue.popleft()
+                    update = build_tool_complete(
+                        tc_id, tool_name, result=str(result) if result is not None else None
+                    )
+                    _send_update(conn, session_id, loop, update)
+                    if not queue:
+                        tool_call_ids.pop(tool_name, None)
+
+    return _step
+
+
+# ------------------------------------------------------------------
+# Agent message callback
+# ------------------------------------------------------------------
+
+def make_message_cb(
+    conn: acp.Client,
+    session_id: str,
+    loop: asyncio.AbstractEventLoop,
+) -> Callable:
+    """Create a callback that streams agent response text to the editor."""
+
+    def _message(text: str) -> None:
+        if not text:
+            return
+        update = acp.update_agent_message_text(text)
+        _send_update(conn, session_id, loop, update)
+
+    return _message

acp_adapter/permissions.py

@@ -0,0 +1,77 @@
+"""ACP permission bridging — maps ACP approval requests to hermes approval callbacks."""
+
+from __future__ import annotations
+
+import asyncio
+import logging
+from concurrent.futures import TimeoutError as FutureTimeout
+from typing import Callable
+
+from acp.schema import (
+    AllowedOutcome,
+    PermissionOption,
+)
+
+logger = logging.getLogger(__name__)
+
+# Maps ACP PermissionOptionKind -> hermes approval result strings
+_KIND_TO_HERMES = {
+    "allow_once": "once",
+    "allow_always": "always",
+    "reject_once": "deny",
+    "reject_always": "deny",
+}
+
+
+def make_approval_callback(
+    request_permission_fn: Callable,
+    loop: asyncio.AbstractEventLoop,
+    session_id: str,
+    timeout: float = 60.0,
+) -> Callable[[str, str], str]:
+    """
+    Return a hermes-compatible ``approval_callback(command, description) -> str``
+    that bridges to the ACP client's ``request_permission`` call.
+
+    Args:
+        request_permission_fn: The ACP connection's ``request_permission`` coroutine.
+        loop: The event loop on which the ACP connection lives.
+        session_id: Current ACP session id.
+        timeout: Seconds to wait for a response before auto-denying.
+    """
+
+    def _callback(command: str, description: str) -> str:
+        options = [
+            PermissionOption(option_id="allow_once", kind="allow_once", name="Allow once"),
+            PermissionOption(option_id="allow_always", kind="allow_always", name="Allow always"),
+            PermissionOption(option_id="deny", kind="reject_once", name="Deny"),
+        ]
+        import acp as _acp
+
+        tool_call = _acp.start_tool_call("perm-check", command, kind="execute")
+
+        coro = request_permission_fn(
+            session_id=session_id,
+            tool_call=tool_call,
+            options=options,
+        )
+
+        try:
+            future = asyncio.run_coroutine_threadsafe(coro, loop)
+            response = future.result(timeout=timeout)
+        except (FutureTimeout, Exception) as exc:
+            logger.warning("Permission request timed out or failed: %s", exc)
+            return "deny"
+
+        outcome = response.outcome
+        if isinstance(outcome, AllowedOutcome):
+            option_id = outcome.option_id
+            # Look up the kind from our options list
+            for opt in options:
+                if opt.option_id == option_id:
+                    return _KIND_TO_HERMES.get(opt.kind, "deny")
+            return "once"  # fallback for unknown option_id
+        else:
+            return "deny"
+
+    return _callback

acp_adapter/server.py

@@ -0,0 +1,535 @@
+"""ACP agent server — exposes Hermes Agent via the Agent Client Protocol."""
+
+from __future__ import annotations
+
+import asyncio
+import logging
+from collections import defaultdict, deque
+from concurrent.futures import ThreadPoolExecutor
+from typing import Any, Deque, Optional
+
+import acp
+from acp.schema import (
+    AgentCapabilities,
+    AuthenticateResponse,
+    AuthMethod,
+    ClientCapabilities,
+    EmbeddedResourceContentBlock,
+    ForkSessionResponse,
+    ImageContentBlock,
+    AudioContentBlock,
+    Implementation,
+    InitializeResponse,
+    ListSessionsResponse,
+    LoadSessionResponse,
+    NewSessionResponse,
+    PromptResponse,
+    ResumeSessionResponse,
+    SetSessionConfigOptionResponse,
+    SetSessionModelResponse,
+    SetSessionModeResponse,
+    ResourceContentBlock,
+    SessionCapabilities,
+    SessionForkCapabilities,
+    SessionListCapabilities,
+    SessionInfo,
+    TextContentBlock,
+    Usage,
+)
+
+from acp_adapter.auth import detect_provider, has_provider
+from acp_adapter.events import (
+    make_message_cb,
+    make_step_cb,
+    make_thinking_cb,
+    make_tool_progress_cb,
+)
+from acp_adapter.permissions import make_approval_callback
+from acp_adapter.session import SessionManager, SessionState
+
+logger = logging.getLogger(__name__)
+
+try:
+    from hermes_cli import __version__ as HERMES_VERSION
+except Exception:
+    HERMES_VERSION = "0.0.0"
+
+# Thread pool for running AIAgent (synchronous) in parallel.
+_executor = ThreadPoolExecutor(max_workers=4, thread_name_prefix="acp-agent")
+
+
+def _extract_text(
+    prompt: list[
+        TextContentBlock
+        | ImageContentBlock
+        | AudioContentBlock
+        | ResourceContentBlock
+        | EmbeddedResourceContentBlock
+    ],
+) -> str:
+    """Extract plain text from ACP content blocks."""
+    parts: list[str] = []
+    for block in prompt:
+        if isinstance(block, TextContentBlock):
+            parts.append(block.text)
+        elif hasattr(block, "text"):
+            parts.append(str(block.text))
+        # Non-text blocks are ignored for now.
+    return "\n".join(parts)
+
+
+class HermesACPAgent(acp.Agent):
+    """ACP Agent implementation wrapping Hermes AIAgent."""
+
+    def __init__(self, session_manager: SessionManager | None = None):
+        super().__init__()
+        self.session_manager = session_manager or SessionManager()
+        self._conn: Optional[acp.Client] = None
+
+    # ---- Connection lifecycle -----------------------------------------------
+
+    def on_connect(self, conn: acp.Client) -> None:
+        """Store the client connection for sending session updates."""
+        self._conn = conn
+        logger.info("ACP client connected")
+
+    # ---- ACP lifecycle ------------------------------------------------------
+
+    async def initialize(
+        self,
+        protocol_version: int | None = None,
+        client_capabilities: ClientCapabilities | None = None,
+        client_info: Implementation | None = None,
+        **kwargs: Any,
+    ) -> InitializeResponse:
+        resolved_protocol_version = (
+            protocol_version if isinstance(protocol_version, int) else acp.PROTOCOL_VERSION
+        )
+        provider = detect_provider()
+        auth_methods = None
+        if provider:
+            auth_methods = [
+                AuthMethod(
+                    id=provider,
+                    name=f"{provider} runtime credentials",
+                    description=f"Authenticate Hermes using the currently configured {provider} runtime credentials.",
+                )
+            ]
+
+        client_name = client_info.name if client_info else "unknown"
+        logger.info(
+            "Initialize from %s (protocol v%s)",
+            client_name,
+            resolved_protocol_version,
+        )
+
+        return InitializeResponse(
+            protocol_version=acp.PROTOCOL_VERSION,
+            agent_info=Implementation(name="hermes-agent", version=HERMES_VERSION),
+            agent_capabilities=AgentCapabilities(
+                session_capabilities=SessionCapabilities(
+                    fork=SessionForkCapabilities(),
+                    list=SessionListCapabilities(),
+                ),
+            ),
+            auth_methods=auth_methods,
+        )
+
+    async def authenticate(self, method_id: str, **kwargs: Any) -> AuthenticateResponse | None:
+        if has_provider():
+            return AuthenticateResponse()
+        return None
+
+    # ---- Session management -------------------------------------------------
+
+    async def new_session(
+        self,
+        cwd: str,
+        mcp_servers: list | None = None,
+        **kwargs: Any,
+    ) -> NewSessionResponse:
+        state = self.session_manager.create_session(cwd=cwd)
+        logger.info("New session %s (cwd=%s)", state.session_id, cwd)
+        return NewSessionResponse(session_id=state.session_id)
+
+    async def load_session(
+        self,
+        cwd: str,
+        session_id: str,
+        mcp_servers: list | None = None,
+        **kwargs: Any,
+    ) -> LoadSessionResponse | None:
+        state = self.session_manager.update_cwd(session_id, cwd)
+        if state is None:
+            logger.warning("load_session: session %s not found", session_id)
+            return None
+        logger.info("Loaded session %s", session_id)
+        return LoadSessionResponse()
+
+    async def resume_session(
+        self,
+        cwd: str,
+        session_id: str,
+        mcp_servers: list | None = None,
+        **kwargs: Any,
+    ) -> ResumeSessionResponse:
+        state = self.session_manager.update_cwd(session_id, cwd)
+        if state is None:
+            logger.warning("resume_session: session %s not found, creating new", session_id)
+            state = self.session_manager.create_session(cwd=cwd)
+        logger.info("Resumed session %s", state.session_id)
+        return ResumeSessionResponse()
+
+    async def cancel(self, session_id: str, **kwargs: Any) -> None:
+        state = self.session_manager.get_session(session_id)
+        if state and state.cancel_event:
+            state.cancel_event.set()
+            try:
+                if getattr(state, "agent", None) and hasattr(state.agent, "interrupt"):
+                    state.agent.interrupt()
+            except Exception:
+                logger.debug("Failed to interrupt ACP session %s", session_id, exc_info=True)
+            logger.info("Cancelled session %s", session_id)
+
+    async def fork_session(
+        self,
+        cwd: str,
+        session_id: str,
+        mcp_servers: list | None = None,
+        **kwargs: Any,
+    ) -> ForkSessionResponse:
+        state = self.session_manager.fork_session(session_id, cwd=cwd)
+        new_id = state.session_id if state else ""
+        logger.info("Forked session %s -> %s", session_id, new_id)
+        return ForkSessionResponse(session_id=new_id)
+
+    async def list_sessions(
+        self,
+        cursor: str | None = None,
+        cwd: str | None = None,
+        **kwargs: Any,
+    ) -> ListSessionsResponse:
+        infos = self.session_manager.list_sessions()
+        sessions = [
+            SessionInfo(session_id=s["session_id"], cwd=s["cwd"])
+            for s in infos
+        ]
+        return ListSessionsResponse(sessions=sessions)
+
+    # ---- Prompt (core) ------------------------------------------------------
+
+    async def prompt(
+        self,
+        prompt: list[
+            TextContentBlock
+            | ImageContentBlock
+            | AudioContentBlock
+            | ResourceContentBlock
+            | EmbeddedResourceContentBlock
+        ],
+        session_id: str,
+        **kwargs: Any,
+    ) -> PromptResponse:
+        """Run Hermes on the user's prompt and stream events back to the editor."""
+        state = self.session_manager.get_session(session_id)
+        if state is None:
+            logger.error("prompt: session %s not found", session_id)
+            return PromptResponse(stop_reason="refusal")
+
+        user_text = _extract_text(prompt).strip()
+        if not user_text:
+            return PromptResponse(stop_reason="end_turn")
+
+        # Intercept slash commands — handle locally without calling the LLM
+        if user_text.startswith("/"):
+            response_text = self._handle_slash_command(user_text, state)
+            if response_text is not None:
+                if self._conn:
+                    update = acp.update_agent_message_text(response_text)
+                    await self._conn.session_update(session_id, update)
+                return PromptResponse(stop_reason="end_turn")
+
+        logger.info("Prompt on session %s: %s", session_id, user_text[:100])
+
+        conn = self._conn
+        loop = asyncio.get_running_loop()
+
+        if state.cancel_event:
+            state.cancel_event.clear()
+
+        tool_call_ids: dict[str, Deque[str]] = defaultdict(deque)
+        previous_approval_cb = None
+
+        if conn:
+            tool_progress_cb = make_tool_progress_cb(conn, session_id, loop, tool_call_ids)
+            thinking_cb = make_thinking_cb(conn, session_id, loop)
+            step_cb = make_step_cb(conn, session_id, loop, tool_call_ids)
+            message_cb = make_message_cb(conn, session_id, loop)
+            approval_cb = make_approval_callback(conn.request_permission, loop, session_id)
+        else:
+            tool_progress_cb = None
+            thinking_cb = None
+            step_cb = None
+            message_cb = None
+            approval_cb = None
+
+        agent = state.agent
+        agent.tool_progress_callback = tool_progress_cb
+        agent.thinking_callback = thinking_cb
+        agent.step_callback = step_cb
+        agent.message_callback = message_cb
+
+        if approval_cb:
+            try:
+                from tools import terminal_tool as _terminal_tool
+                previous_approval_cb = getattr(_terminal_tool, "_approval_callback", None)
+                _terminal_tool.set_approval_callback(approval_cb)
+            except Exception:
+                logger.debug("Could not set ACP approval callback", exc_info=True)
+
+        def _run_agent() -> dict:
+            try:
+                result = agent.run_conversation(
+                    user_message=user_text,
+                    conversation_history=state.history,
+                    task_id=session_id,
+                )
+                return result
+            except Exception as e:
+                logger.exception("Agent error in session %s", session_id)
+                return {"final_response": f"Error: {e}", "messages": state.history}
+            finally:
+                if approval_cb:
+                    try:
+                        from tools import terminal_tool as _terminal_tool
+                        _terminal_tool.set_approval_callback(previous_approval_cb)
+                    except Exception:
+                        logger.debug("Could not restore approval callback", exc_info=True)
+
+        try:
+            result = await loop.run_in_executor(_executor, _run_agent)
+        except Exception:
+            logger.exception("Executor error for session %s", session_id)
+            return PromptResponse(stop_reason="end_turn")
+
+        if result.get("messages"):
+            state.history = result["messages"]
+            # Persist updated history so sessions survive process restarts.
+            self.session_manager.save_session(session_id)
+
+        final_response = result.get("final_response", "")
+        if final_response and conn:
+            update = acp.update_agent_message_text(final_response)
+            await conn.session_update(session_id, update)
+
+        usage = None
+        usage_data = result.get("usage")
+        if usage_data and isinstance(usage_data, dict):
+            usage = Usage(
+                input_tokens=usage_data.get("prompt_tokens", 0),
+                output_tokens=usage_data.get("completion_tokens", 0),
+                total_tokens=usage_data.get("total_tokens", 0),
+                thought_tokens=usage_data.get("reasoning_tokens"),
+                cached_read_tokens=usage_data.get("cached_tokens"),
+            )
+
+        stop_reason = "cancelled" if state.cancel_event and state.cancel_event.is_set() else "end_turn"
+        return PromptResponse(stop_reason=stop_reason, usage=usage)
+
+    # ---- Slash commands (headless) -------------------------------------------
+
+    _SLASH_COMMANDS = {
+        "help": "Show available commands",
+        "model": "Show or change current model",
+        "tools": "List available tools",
+        "context": "Show conversation context info",
+        "reset": "Clear conversation history",
+        "compact": "Compress conversation context",
+        "version": "Show Hermes version",
+    }
+
+    def _handle_slash_command(self, text: str, state: SessionState) -> str | None:
+        """Dispatch a slash command and return the response text.
+
+        Returns ``None`` for unrecognized commands so they fall through
+        to the LLM (the user may have typed ``/something`` as prose).
+        """
+        parts = text.split(maxsplit=1)
+        cmd = parts[0].lstrip("/").lower()
+        args = parts[1].strip() if len(parts) > 1 else ""
+
+        handler = {
+            "help": self._cmd_help,
+            "model": self._cmd_model,
+            "tools": self._cmd_tools,
+            "context": self._cmd_context,
+            "reset": self._cmd_reset,
+            "compact": self._cmd_compact,
+            "version": self._cmd_version,
+        }.get(cmd)
+
+        if handler is None:
+            return None  # not a known command — let the LLM handle it
+
+        try:
+            return handler(args, state)
+        except Exception as e:
+            logger.error("Slash command /%s error: %s", cmd, e, exc_info=True)
+            return f"Error executing /{cmd}: {e}"
+
+    def _cmd_help(self, args: str, state: SessionState) -> str:
+        lines = ["Available commands:", ""]
+        for cmd, desc in self._SLASH_COMMANDS.items():
+            lines.append(f"  /{cmd:10s}  {desc}")
+        lines.append("")
+        lines.append("Unrecognized /commands are sent to the model as normal messages.")
+        return "\n".join(lines)
+
+    def _cmd_model(self, args: str, state: SessionState) -> str:
+        if not args:
+            model = state.model or getattr(state.agent, "model", "unknown")
+            provider = getattr(state.agent, "provider", None) or "auto"
+            return f"Current model: {model}\nProvider: {provider}"
+
+        new_model = args.strip()
+        target_provider = None
+        current_provider = getattr(state.agent, "provider", None) or "openrouter"
+
+        # Auto-detect provider for the requested model
+        try:
+            from hermes_cli.models import parse_model_input, detect_provider_for_model
+            target_provider, new_model = parse_model_input(new_model, current_provider)
+            if target_provider == current_provider:
+                detected = detect_provider_for_model(new_model, current_provider)
+                if detected:
+                    target_provider, new_model = detected
+        except Exception:
+            logger.debug("Provider detection failed, using model as-is", exc_info=True)
+
+        state.model = new_model
+        state.agent = self.session_manager._make_agent(
+            session_id=state.session_id,
+            cwd=state.cwd,
+            model=new_model,
+            requested_provider=target_provider or current_provider,
+        )
+        self.session_manager.save_session(state.session_id)
+        provider_label = getattr(state.agent, "provider", None) or target_provider or current_provider
+        logger.info("Session %s: model switched to %s", state.session_id, new_model)
+        return f"Model switched to: {new_model}\nProvider: {provider_label}"
+
+    def _cmd_tools(self, args: str, state: SessionState) -> str:
+        try:
+            from model_tools import get_tool_definitions
+            toolsets = getattr(state.agent, "enabled_toolsets", None) or ["hermes-acp"]
+            tools = get_tool_definitions(enabled_toolsets=toolsets, quiet_mode=True)
+            if not tools:
+                return "No tools available."
+            lines = [f"Available tools ({len(tools)}):"]
+            for t in tools:
+                name = t.get("function", {}).get("name", "?")
+                desc = t.get("function", {}).get("description", "")
+                # Truncate long descriptions
+                if len(desc) > 80:
+                    desc = desc[:77] + "..."
+                lines.append(f"  {name}: {desc}")
+            return "\n".join(lines)
+        except Exception as e:
+            return f"Could not list tools: {e}"
+
+    def _cmd_context(self, args: str, state: SessionState) -> str:
+        n_messages = len(state.history)
+        if n_messages == 0:
+            return "Conversation is empty (no messages yet)."
+        # Count by role
+        roles: dict[str, int] = {}
+        for msg in state.history:
+            role = msg.get("role", "unknown")
+            roles[role] = roles.get(role, 0) + 1
+        lines = [
+            f"Conversation: {n_messages} messages",
+            f"  user: {roles.get('user', 0)}, assistant: {roles.get('assistant', 0)}, "
+            f"tool: {roles.get('tool', 0)}, system: {roles.get('system', 0)}",
+        ]
+        model = state.model or getattr(state.agent, "model", "")
+        if model:
+            lines.append(f"Model: {model}")
+        return "\n".join(lines)
+
+    def _cmd_reset(self, args: str, state: SessionState) -> str:
+        state.history.clear()
+        self.session_manager.save_session(state.session_id)
+        return "Conversation history cleared."
+
+    def _cmd_compact(self, args: str, state: SessionState) -> str:
+        if not state.history:
+            return "Nothing to compress — conversation is empty."
+        try:
+            agent = state.agent
+            if hasattr(agent, "compress_context"):
+                agent.compress_context(state.history)
+                self.session_manager.save_session(state.session_id)
+                return f"Context compressed. Messages: {len(state.history)}"
+            return "Context compression not available for this agent."
+        except Exception as e:
+            return f"Compression failed: {e}"
+
+    def _cmd_version(self, args: str, state: SessionState) -> str:
+        return f"Hermes Agent v{HERMES_VERSION}"
+
+    # ---- Model switching (ACP protocol method) -------------------------------
+
+    async def set_session_model(
+        self, model_id: str, session_id: str, **kwargs: Any
+    ) -> SetSessionModelResponse | None:
+        """Switch the model for a session (called by ACP protocol)."""
+        state = self.session_manager.get_session(session_id)
+        if state:
+            state.model = model_id
+            current_provider = getattr(state.agent, "provider", None)
+            current_base_url = getattr(state.agent, "base_url", None)
+            current_api_mode = getattr(state.agent, "api_mode", None)
+            state.agent = self.session_manager._make_agent(
+                session_id=session_id,
+                cwd=state.cwd,
+                model=model_id,
+                requested_provider=current_provider,
+                base_url=current_base_url,
+                api_mode=current_api_mode,
+            )
+            self.session_manager.save_session(session_id)
+            logger.info("Session %s: model switched to %s", session_id, model_id)
+            return SetSessionModelResponse()
+        logger.warning("Session %s: model switch requested for missing session", session_id)
+        return None
+
+    async def set_session_mode(
+        self, mode_id: str, session_id: str, **kwargs: Any
+    ) -> SetSessionModeResponse | None:
+        """Persist the editor-requested mode so ACP clients do not fail on mode switches."""
+        state = self.session_manager.get_session(session_id)
+        if state is None:
+            logger.warning("Session %s: mode switch requested for missing session", session_id)
+            return None
+        setattr(state, "mode", mode_id)
+        self.session_manager.save_session(session_id)
+        logger.info("Session %s: mode switched to %s", session_id, mode_id)
+        return SetSessionModeResponse()
+
+    async def set_config_option(
+        self, config_id: str, session_id: str, value: str, **kwargs: Any
+    ) -> SetSessionConfigOptionResponse | None:
+        """Accept ACP config option updates even when Hermes has no typed ACP config surface yet."""
+        state = self.session_manager.get_session(session_id)
+        if state is None:
+            logger.warning("Session %s: config update requested for missing session", session_id)
+            return None
+
+        options = getattr(state, "config_options", None)
+        if not isinstance(options, dict):
+            options = {}
+        options[str(config_id)] = value
+        setattr(state, "config_options", options)
+        self.session_manager.save_session(session_id)
+        logger.info("Session %s: config option %s updated", session_id, config_id)
+        return SetSessionConfigOptionResponse(config_options=[])

acp_adapter/session.py

@@ -0,0 +1,461 @@
+"""ACP session manager — maps ACP sessions to Hermes AIAgent instances.
+
+Sessions are persisted to the shared SessionDB (``~/.hermes/state.db``) so they
+survive process restarts and appear in ``session_search``.  When the editor
+reconnects after idle/restart, the ``load_session`` / ``resume_session`` calls
+find the persisted session in the database and restore the full conversation
+history.
+"""
+from __future__ import annotations
+
+from hermes_constants import get_hermes_home
+
+import copy
+import json
+import logging
+import uuid
+from dataclasses import dataclass, field
+from threading import Lock
+from typing import Any, Dict, List, Optional
+
+logger = logging.getLogger(__name__)
+
+
+def _register_task_cwd(task_id: str, cwd: str) -> None:
+    """Bind a task/session id to the editor's working directory for tools."""
+    if not task_id:
+        return
+    try:
+        from tools.terminal_tool import register_task_env_overrides
+        register_task_env_overrides(task_id, {"cwd": cwd})
+    except Exception:
+        logger.debug("Failed to register ACP task cwd override", exc_info=True)
+
+
+def _clear_task_cwd(task_id: str) -> None:
+    """Remove task-specific cwd overrides for an ACP session."""
+    if not task_id:
+        return
+    try:
+        from tools.terminal_tool import clear_task_env_overrides
+        clear_task_env_overrides(task_id)
+    except Exception:
+        logger.debug("Failed to clear ACP task cwd override", exc_info=True)
+
+
+@dataclass
+class SessionState:
+    """Tracks per-session state for an ACP-managed Hermes agent."""
+
+    session_id: str
+    agent: Any  # AIAgent instance
+    cwd: str = "."
+    model: str = ""
+    history: List[Dict[str, Any]] = field(default_factory=list)
+    cancel_event: Any = None  # threading.Event
+
+
+class SessionManager:
+    """Thread-safe manager for ACP sessions backed by Hermes AIAgent instances.
+
+    Sessions are held in-memory for fast access **and** persisted to the
+    shared SessionDB so they survive process restarts and are searchable
+    via ``session_search``.
+    """
+
+    def __init__(self, agent_factory=None, db=None):
+        """
+        Args:
+            agent_factory: Optional callable that creates an AIAgent-like object.
+                           Used by tests. When omitted, a real AIAgent is created
+                           using the current Hermes runtime provider configuration.
+            db:            Optional SessionDB instance. When omitted, the default
+                           SessionDB (``~/.hermes/state.db``) is lazily created.
+        """
+        self._sessions: Dict[str, SessionState] = {}
+        self._lock = Lock()
+        self._agent_factory = agent_factory
+        self._db_instance = db  # None → lazy-init on first use
+
+    # ---- public API ---------------------------------------------------------
+
+    def create_session(self, cwd: str = ".") -> SessionState:
+        """Create a new session with a unique ID and a fresh AIAgent."""
+        import threading
+
+        session_id = str(uuid.uuid4())
+        agent = self._make_agent(session_id=session_id, cwd=cwd)
+        state = SessionState(
+            session_id=session_id,
+            agent=agent,
+            cwd=cwd,
+            model=getattr(agent, "model", "") or "",
+            cancel_event=threading.Event(),
+        )
+        with self._lock:
+            self._sessions[session_id] = state
+        _register_task_cwd(session_id, cwd)
+        self._persist(state)
+        logger.info("Created ACP session %s (cwd=%s)", session_id, cwd)
+        return state
+
+    def get_session(self, session_id: str) -> Optional[SessionState]:
+        """Return the session for *session_id*, or ``None``.
+
+        If the session is not in memory but exists in the database (e.g. after
+        a process restart), it is transparently restored.
+        """
+        with self._lock:
+            state = self._sessions.get(session_id)
+        if state is not None:
+            return state
+        # Attempt to restore from database.
+        return self._restore(session_id)
+
+    def remove_session(self, session_id: str) -> bool:
+        """Remove a session from memory and database. Returns True if it existed."""
+        with self._lock:
+            existed = self._sessions.pop(session_id, None) is not None
+        db_existed = self._delete_persisted(session_id)
+        if existed or db_existed:
+            _clear_task_cwd(session_id)
+        return existed or db_existed
+
+    def fork_session(self, session_id: str, cwd: str = ".") -> Optional[SessionState]:
+        """Deep-copy a session's history into a new session."""
+        import threading
+
+        original = self.get_session(session_id)  # checks DB too
+        if original is None:
+            return None
+
+        new_id = str(uuid.uuid4())
+        agent = self._make_agent(
+            session_id=new_id,
+            cwd=cwd,
+            model=original.model or None,
+        )
+        state = SessionState(
+            session_id=new_id,
+            agent=agent,
+            cwd=cwd,
+            model=getattr(agent, "model", original.model) or original.model,
+            history=copy.deepcopy(original.history),
+            cancel_event=threading.Event(),
+        )
+        with self._lock:
+            self._sessions[new_id] = state
+        _register_task_cwd(new_id, cwd)
+        self._persist(state)
+        logger.info("Forked ACP session %s -> %s", session_id, new_id)
+        return state
+
+    def list_sessions(self) -> List[Dict[str, Any]]:
+        """Return lightweight info dicts for all sessions (memory + database)."""
+        # Collect in-memory sessions first.
+        with self._lock:
+            seen_ids = set(self._sessions.keys())
+            results = [
+                {
+                    "session_id": s.session_id,
+                    "cwd": s.cwd,
+                    "model": s.model,
+                    "history_len": len(s.history),
+                }
+                for s in self._sessions.values()
+            ]
+
+        # Merge any persisted sessions not currently in memory.
+        db = self._get_db()
+        if db is not None:
+            try:
+                rows = db.search_sessions(source="acp", limit=1000)
+                for row in rows:
+                    sid = row["id"]
+                    if sid in seen_ids:
+                        continue
+                    # Extract cwd from model_config JSON.
+                    cwd = "."
+                    mc = row.get("model_config")
+                    if mc:
+                        try:
+                            cwd = json.loads(mc).get("cwd", ".")
+                        except (json.JSONDecodeError, TypeError):
+                            pass
+                    results.append({
+                        "session_id": sid,
+                        "cwd": cwd,
+                        "model": row.get("model") or "",
+                        "history_len": row.get("message_count") or 0,
+                    })
+            except Exception:
+                logger.debug("Failed to list ACP sessions from DB", exc_info=True)
+
+        return results
+
+    def update_cwd(self, session_id: str, cwd: str) -> Optional[SessionState]:
+        """Update the working directory for a session and its tool overrides."""
+        state = self.get_session(session_id)  # checks DB too
+        if state is None:
+            return None
+        state.cwd = cwd
+        _register_task_cwd(session_id, cwd)
+        self._persist(state)
+        return state
+
+    def cleanup(self) -> None:
+        """Remove all sessions (memory and database) and clear task-specific cwd overrides."""
+        with self._lock:
+            session_ids = list(self._sessions.keys())
+            self._sessions.clear()
+        for session_id in session_ids:
+            _clear_task_cwd(session_id)
+            self._delete_persisted(session_id)
+        # Also remove any DB-only ACP sessions not currently in memory.
+        db = self._get_db()
+        if db is not None:
+            try:
+                rows = db.search_sessions(source="acp", limit=10000)
+                for row in rows:
+                    sid = row["id"]
+                    _clear_task_cwd(sid)
+                    db.delete_session(sid)
+            except Exception:
+                logger.debug("Failed to cleanup ACP sessions from DB", exc_info=True)
+
+    def save_session(self, session_id: str) -> None:
+        """Persist the current state of a session to the database.
+
+        Called by the server after prompt completion, slash commands that
+        mutate history, and model switches.
+        """
+        with self._lock:
+            state = self._sessions.get(session_id)
+        if state is not None:
+            self._persist(state)
+
+    # ---- persistence via SessionDB ------------------------------------------
+
+    def _get_db(self):
+        """Lazily initialise and return the SessionDB instance.
+
+        Returns ``None`` if the DB is unavailable (e.g. import error in a
+        minimal test environment).
+
+        Note: we resolve ``HERMES_HOME`` dynamically rather than relying on
+        the module-level ``DEFAULT_DB_PATH`` constant, because that constant
+        is evaluated at import time and won't reflect env-var changes made
+        later (e.g. by the test fixture ``_isolate_hermes_home``).
+        """
+        if self._db_instance is not None:
+            return self._db_instance
+        try:
+            import os
+            from pathlib import Path
+            from hermes_state import SessionDB
+            hermes_home = get_hermes_home()
+            self._db_instance = SessionDB(db_path=hermes_home / "state.db")
+            return self._db_instance
+        except Exception:
+            logger.debug("SessionDB unavailable for ACP persistence", exc_info=True)
+            return None
+
+    def _persist(self, state: SessionState) -> None:
+        """Write session state to the database.
+
+        Creates the session record if it doesn't exist, then replaces all
+        stored messages with the current in-memory history.
+        """
+        db = self._get_db()
+        if db is None:
+            return
+
+        # Ensure model is a plain string (not a MagicMock or other proxy).
+        model_str = str(state.model) if state.model else None
+        session_meta = {"cwd": state.cwd}
+        provider = getattr(state.agent, "provider", None)
+        base_url = getattr(state.agent, "base_url", None)
+        api_mode = getattr(state.agent, "api_mode", None)
+        if isinstance(provider, str) and provider.strip():
+            session_meta["provider"] = provider.strip()
+        if isinstance(base_url, str) and base_url.strip():
+            session_meta["base_url"] = base_url.strip()
+        if isinstance(api_mode, str) and api_mode.strip():
+            session_meta["api_mode"] = api_mode.strip()
+        cwd_json = json.dumps(session_meta)
+
+        try:
+            # Ensure the session record exists.
+            existing = db.get_session(state.session_id)
+            if existing is None:
+                db.create_session(
+                    session_id=state.session_id,
+                    source="acp",
+                    model=model_str,
+                    model_config={"cwd": state.cwd},
+                )
+            else:
+                # Update model_config (contains cwd) if changed.
+                try:
+                    with db._lock:
+                        db._conn.execute(
+                            "UPDATE sessions SET model_config = ?, model = COALESCE(?, model) WHERE id = ?",
+                            (cwd_json, model_str, state.session_id),
+                        )
+                        db._conn.commit()
+                except Exception:
+                    logger.debug("Failed to update ACP session metadata", exc_info=True)
+
+            # Replace stored messages with current history.
+            db.clear_messages(state.session_id)
+            for msg in state.history:
+                db.append_message(
+                    session_id=state.session_id,
+                    role=msg.get("role", "user"),
+                    content=msg.get("content"),
+                    tool_name=msg.get("tool_name") or msg.get("name"),
+                    tool_calls=msg.get("tool_calls"),
+                    tool_call_id=msg.get("tool_call_id"),
+                )
+        except Exception:
+            logger.warning("Failed to persist ACP session %s", state.session_id, exc_info=True)
+
+    def _restore(self, session_id: str) -> Optional[SessionState]:
+        """Load a session from the database into memory, recreating the AIAgent."""
+        import threading
+
+        db = self._get_db()
+        if db is None:
+            return None
+
+        try:
+            row = db.get_session(session_id)
+        except Exception:
+            logger.debug("Failed to query DB for ACP session %s", session_id, exc_info=True)
+            return None
+
+        if row is None:
+            return None
+
+        # Only restore ACP sessions.
+        if row.get("source") != "acp":
+            return None
+
+        # Extract cwd from model_config.
+        cwd = "."
+        requested_provider = row.get("billing_provider")
+        restored_base_url = row.get("billing_base_url")
+        restored_api_mode = None
+        mc = row.get("model_config")
+        if mc:
+            try:
+                meta = json.loads(mc)
+                if isinstance(meta, dict):
+                    cwd = meta.get("cwd", ".")
+                    requested_provider = meta.get("provider") or requested_provider
+                    restored_base_url = meta.get("base_url") or restored_base_url
+                    restored_api_mode = meta.get("api_mode") or restored_api_mode
+            except (json.JSONDecodeError, TypeError):
+                pass
+
+        model = row.get("model") or None
+
+        # Load conversation history.
+        try:
+            history = db.get_messages_as_conversation(session_id)
+        except Exception:
+            logger.warning("Failed to load messages for ACP session %s", session_id, exc_info=True)
+            history = []
+
+        try:
+            agent = self._make_agent(
+                session_id=session_id,
+                cwd=cwd,
+                model=model,
+                requested_provider=requested_provider,
+                base_url=restored_base_url,
+                api_mode=restored_api_mode,
+            )
+        except Exception:
+            logger.warning("Failed to recreate agent for ACP session %s", session_id, exc_info=True)
+            return None
+
+        state = SessionState(
+            session_id=session_id,
+            agent=agent,
+            cwd=cwd,
+            model=model or getattr(agent, "model", "") or "",
+            history=history,
+            cancel_event=threading.Event(),
+        )
+        with self._lock:
+            self._sessions[session_id] = state
+        _register_task_cwd(session_id, cwd)
+        logger.info("Restored ACP session %s from DB (%d messages)", session_id, len(history))
+        return state
+
+    def _delete_persisted(self, session_id: str) -> bool:
+        """Delete a session from the database. Returns True if it existed."""
+        db = self._get_db()
+        if db is None:
+            return False
+        try:
+            return db.delete_session(session_id)
+        except Exception:
+            logger.debug("Failed to delete ACP session %s from DB", session_id, exc_info=True)
+            return False
+
+    # ---- internal -----------------------------------------------------------
+
+    def _make_agent(
+        self,
+        *,
+        session_id: str,
+        cwd: str,
+        model: str | None = None,
+        requested_provider: str | None = None,
+        base_url: str | None = None,
+        api_mode: str | None = None,
+    ):
+        if self._agent_factory is not None:
+            return self._agent_factory()
+
+        from run_agent import AIAgent
+        from hermes_cli.config import load_config
+        from hermes_cli.runtime_provider import resolve_runtime_provider
+
+        config = load_config()
+        model_cfg = config.get("model")
+        default_model = "anthropic/claude-opus-4.6"
+        config_provider = None
+        if isinstance(model_cfg, dict):
+            default_model = str(model_cfg.get("default") or default_model)
+            config_provider = model_cfg.get("provider")
+        elif isinstance(model_cfg, str) and model_cfg.strip():
+            default_model = model_cfg.strip()
+
+        kwargs = {
+            "platform": "acp",
+            "enabled_toolsets": ["hermes-acp"],
+            "quiet_mode": True,
+            "session_id": session_id,
+            "model": model or default_model,
+        }
+
+        try:
+            runtime = resolve_runtime_provider(requested=requested_provider or config_provider)
+            kwargs.update(
+                {
+                    "provider": runtime.get("provider"),
+                    "api_mode": api_mode or runtime.get("api_mode"),
+                    "base_url": base_url or runtime.get("base_url"),
+                    "api_key": runtime.get("api_key"),
+                    "command": runtime.get("command"),
+                    "args": list(runtime.get("args") or []),
+                }
+            )
+        except Exception:
+            logger.debug("ACP session falling back to default provider resolution", exc_info=True)
+
+        _register_task_cwd(session_id, cwd)
+        return AIAgent(**kwargs)

acp_adapter/tools.py

@@ -0,0 +1,215 @@
+"""ACP tool-call helpers for mapping hermes tools to ACP ToolKind and building content."""
+
+from __future__ import annotations
+
+import uuid
+from typing import Any, Dict, List, Optional
+
+import acp
+from acp.schema import (
+    ToolCallLocation,
+    ToolCallStart,
+    ToolCallProgress,
+    ToolKind,
+)
+
+# ---------------------------------------------------------------------------
+# Map hermes tool names -> ACP ToolKind
+# ---------------------------------------------------------------------------
+
+TOOL_KIND_MAP: Dict[str, ToolKind] = {
+    # File operations
+    "read_file": "read",
+    "write_file": "edit",
+    "patch": "edit",
+    "search_files": "search",
+    # Terminal / execution
+    "terminal": "execute",
+    "process": "execute",
+    "execute_code": "execute",
+    # Web / fetch
+    "web_search": "fetch",
+    "web_extract": "fetch",
+    # Browser
+    "browser_navigate": "fetch",
+    "browser_click": "execute",
+    "browser_type": "execute",
+    "browser_snapshot": "read",
+    "browser_vision": "read",
+    "browser_scroll": "execute",
+    "browser_press": "execute",
+    "browser_back": "execute",
+    "browser_close": "execute",
+    "browser_get_images": "read",
+    # Agent internals
+    "delegate_task": "execute",
+    "vision_analyze": "read",
+    "image_generate": "execute",
+    "text_to_speech": "execute",
+    # Thinking / meta
+    "_thinking": "think",
+}
+
+
+def get_tool_kind(tool_name: str) -> ToolKind:
+    """Return the ACP ToolKind for a hermes tool, defaulting to 'other'."""
+    return TOOL_KIND_MAP.get(tool_name, "other")
+
+
+def make_tool_call_id() -> str:
+    """Generate a unique tool call ID."""
+    return f"tc-{uuid.uuid4().hex[:12]}"
+
+
+def build_tool_title(tool_name: str, args: Dict[str, Any]) -> str:
+    """Build a human-readable title for a tool call."""
+    if tool_name == "terminal":
+        cmd = args.get("command", "")
+        if len(cmd) > 80:
+            cmd = cmd[:77] + "..."
+        return f"terminal: {cmd}"
+    if tool_name == "read_file":
+        return f"read: {args.get('path', '?')}"
+    if tool_name == "write_file":
+        return f"write: {args.get('path', '?')}"
+    if tool_name == "patch":
+        mode = args.get("mode", "replace")
+        path = args.get("path", "?")
+        return f"patch ({mode}): {path}"
+    if tool_name == "search_files":
+        return f"search: {args.get('pattern', '?')}"
+    if tool_name == "web_search":
+        return f"web search: {args.get('query', '?')}"
+    if tool_name == "web_extract":
+        urls = args.get("urls", [])
+        if urls:
+            return f"extract: {urls[0]}" + (f" (+{len(urls)-1})" if len(urls) > 1 else "")
+        return "web extract"
+    if tool_name == "delegate_task":
+        goal = args.get("goal", "")
+        if goal and len(goal) > 60:
+            goal = goal[:57] + "..."
+        return f"delegate: {goal}" if goal else "delegate task"
+    if tool_name == "execute_code":
+        return "execute code"
+    if tool_name == "vision_analyze":
+        return f"analyze image: {args.get('question', '?')[:50]}"
+    return tool_name
+
+
+# ---------------------------------------------------------------------------
+# Build ACP content objects for tool-call events
+# ---------------------------------------------------------------------------
+
+
+def build_tool_start(
+    tool_call_id: str,
+    tool_name: str,
+    arguments: Dict[str, Any],
+) -> ToolCallStart:
+    """Create a ToolCallStart event for the given hermes tool invocation."""
+    kind = get_tool_kind(tool_name)
+    title = build_tool_title(tool_name, arguments)
+    locations = extract_locations(arguments)
+
+    if tool_name == "patch":
+        mode = arguments.get("mode", "replace")
+        if mode == "replace":
+            path = arguments.get("path", "")
+            old = arguments.get("old_string", "")
+            new = arguments.get("new_string", "")
+            content = [acp.tool_diff_content(path=path, new_text=new, old_text=old)]
+        else:
+            # Patch mode — show the patch content as text
+            patch_text = arguments.get("patch", "")
+            content = [acp.tool_content(acp.text_block(patch_text))]
+        return acp.start_tool_call(
+            tool_call_id, title, kind=kind, content=content, locations=locations,
+            raw_input=arguments,
+        )
+
+    if tool_name == "write_file":
+        path = arguments.get("path", "")
+        file_content = arguments.get("content", "")
+        content = [acp.tool_diff_content(path=path, new_text=file_content)]
+        return acp.start_tool_call(
+            tool_call_id, title, kind=kind, content=content, locations=locations,
+            raw_input=arguments,
+        )
+
+    if tool_name == "terminal":
+        command = arguments.get("command", "")
+        content = [acp.tool_content(acp.text_block(f"$ {command}"))]
+        return acp.start_tool_call(
+            tool_call_id, title, kind=kind, content=content, locations=locations,
+            raw_input=arguments,
+        )
+
+    if tool_name == "read_file":
+        path = arguments.get("path", "")
+        content = [acp.tool_content(acp.text_block(f"Reading {path}"))]
+        return acp.start_tool_call(
+            tool_call_id, title, kind=kind, content=content, locations=locations,
+            raw_input=arguments,
+        )
+
+    if tool_name == "search_files":
+        pattern = arguments.get("pattern", "")
+        target = arguments.get("target", "content")
+        content = [acp.tool_content(acp.text_block(f"Searching for '{pattern}' ({target})"))]
+        return acp.start_tool_call(
+            tool_call_id, title, kind=kind, content=content, locations=locations,
+            raw_input=arguments,
+        )
+
+    # Generic fallback
+    import json
+    try:
+        args_text = json.dumps(arguments, indent=2, default=str)
+    except (TypeError, ValueError):
+        args_text = str(arguments)
+    content = [acp.tool_content(acp.text_block(args_text))]
+    return acp.start_tool_call(
+        tool_call_id, title, kind=kind, content=content, locations=locations,
+        raw_input=arguments,
+    )
+
+
+def build_tool_complete(
+    tool_call_id: str,
+    tool_name: str,
+    result: Optional[str] = None,
+) -> ToolCallProgress:
+    """Create a ToolCallUpdate (progress) event for a completed tool call."""
+    kind = get_tool_kind(tool_name)
+
+    # Truncate very large results for the UI
+    display_result = result or ""
+    if len(display_result) > 5000:
+        display_result = display_result[:4900] + f"\n... ({len(result)} chars total, truncated)"
+
+    content = [acp.tool_content(acp.text_block(display_result))]
+    return acp.update_tool_call(
+        tool_call_id,
+        kind=kind,
+        status="completed",
+        content=content,
+        raw_output=result,
+    )
+
+
+# ---------------------------------------------------------------------------
+# Location extraction
+# ---------------------------------------------------------------------------
+
+
+def extract_locations(
+    arguments: Dict[str, Any],
+) -> List[ToolCallLocation]:
+    """Extract file-system locations from tool arguments."""
+    locations: List[ToolCallLocation] = []
+    path = arguments.get("path")
+    if path:
+        line = arguments.get("offset") or arguments.get("line")
+        locations.append(ToolCallLocation(path=path, line=line))
+    return locations

acp_registry/agent.json

@@ -0,0 +1,12 @@
+{
+  "schema_version": 1,
+  "name": "hermes-agent",
+  "display_name": "Hermes Agent",
+  "description": "AI agent by Nous Research with 90+ tools, persistent memory, and multi-platform support",
+  "icon": "icon.svg",
+  "distribution": {
+    "type": "command",
+    "command": "hermes",
+    "args": ["acp"]
+  }
+}

acp_registry/icon.svg

@@ -0,0 +1,25 @@
+<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 64 64" width="64" height="64">
+  <defs>
+    <linearGradient id="gold" x1="0%" y1="0%" x2="0%" y2="100%">
+      <stop offset="0%" style="stop-color:#F5C542;stop-opacity:1" />
+      <stop offset="100%" style="stop-color:#D4961C;stop-opacity:1" />
+    </linearGradient>
+  </defs>
+  <!-- Staff -->
+  <rect x="30" y="10" width="4" height="46" rx="2" fill="url(#gold)" />
+  <!-- Wings (left) -->
+  <path d="M30 18 C24 14, 14 14, 10 18 C14 16, 22 16, 28 20" fill="#F5C542" opacity="0.9" />
+  <path d="M30 22 C26 19, 18 19, 14 22 C18 20, 24 20, 28 24" fill="#D4961C" opacity="0.8" />
+  <!-- Wings (right) -->
+  <path d="M34 18 C40 14, 50 14, 54 18 C50 16, 42 16, 36 20" fill="#F5C542" opacity="0.9" />
+  <path d="M34 22 C38 19, 46 19, 50 22 C46 20, 40 20, 36 24" fill="#D4961C" opacity="0.8" />
+  <!-- Left serpent -->
+  <path d="M32 48 C22 44, 20 38, 26 34 C20 36, 18 42, 24 46 C18 40, 22 30, 30 28 C24 32, 22 38, 28 42"
+        fill="none" stroke="#F5C542" stroke-width="2.5" stroke-linecap="round" />
+  <!-- Right serpent -->
+  <path d="M32 48 C42 44, 44 38, 38 34 C44 36, 46 42, 40 46 C46 40, 42 30, 34 28 C40 32, 42 38, 36 42"
+        fill="none" stroke="#D4961C" stroke-width="2.5" stroke-linecap="round" />
+  <!-- Orb at top -->
+  <circle cx="32" cy="10" r="4" fill="#F5C542" />
+  <circle cx="32" cy="10" r="2" fill="#FFF8E1" opacity="0.7" />
+</svg>

agent/__init__.py

@@ -0,0 +1,6 @@
+"""Agent internals -- extracted modules from run_agent.py.
+
+These modules contain pure utility functions and self-contained classes
+that were previously embedded in the 3,600-line run_agent.py. Extracting
+them makes run_agent.py focused on the AIAgent orchestrator class.
+"""

agent/context_compressor.py

@@ -0,0 +1,676 @@
+"""Automatic context window compression for long conversations.
+
+Self-contained class with its own OpenAI client for summarization.
+Uses auxiliary model (cheap/fast) to summarize middle turns while
+protecting head and tail context.
+
+Improvements over v1:
+  - Structured summary template (Goal, Progress, Decisions, Files, Next Steps)
+  - Iterative summary updates (preserves info across multiple compactions)
+  - Token-budget tail protection instead of fixed message count
+  - Tool output pruning before LLM summarization (cheap pre-pass)
+  - Scaled summary budget (proportional to compressed content)
+  - Richer tool call/result detail in summarizer input
+"""
+
+import logging
+from typing import Any, Dict, List, Optional
+
+from agent.auxiliary_client import call_llm
+from agent.model_metadata import (
+    get_model_context_length,
+    estimate_messages_tokens_rough,
+)
+
+logger = logging.getLogger(__name__)
+
+SUMMARY_PREFIX = (
+    "[CONTEXT COMPACTION] Earlier turns in this conversation were compacted "
+    "to save context space. The summary below describes work that was "
+    "already completed, and the current session state may still reflect "
+    "that work (for example, files may already be changed). Use the summary "
+    "and the current state to continue from where things left off, and "
+    "avoid repeating work:"
+)
+LEGACY_SUMMARY_PREFIX = "[CONTEXT SUMMARY]:"
+
+# Minimum tokens for the summary output
+_MIN_SUMMARY_TOKENS = 2000
+# Proportion of compressed content to allocate for summary
+_SUMMARY_RATIO = 0.20
+# Absolute ceiling for summary tokens (even on very large context windows)
+_SUMMARY_TOKENS_CEILING = 12_000
+
+# Placeholder used when pruning old tool results
+_PRUNED_TOOL_PLACEHOLDER = "[Old tool output cleared to save context space]"
+
+# Chars per token rough estimate
+_CHARS_PER_TOKEN = 4
+
+
+class ContextCompressor:
+    """Compresses conversation context when approaching the model's context limit.
+
+    Algorithm:
+      1. Prune old tool results (cheap, no LLM call)
+      2. Protect head messages (system prompt + first exchange)
+      3. Protect tail messages by token budget (most recent ~20K tokens)
+      4. Summarize middle turns with structured LLM prompt
+      5. On subsequent compactions, iteratively update the previous summary
+    """
+
+    def __init__(
+        self,
+        model: str,
+        threshold_percent: float = 0.50,
+        protect_first_n: int = 3,
+        protect_last_n: int = 20,
+        summary_target_ratio: float = 0.20,
+        quiet_mode: bool = False,
+        summary_model_override: str = None,
+        base_url: str = "",
+        api_key: str = "",
+        config_context_length: int | None = None,
+        provider: str = "",
+    ):
+        self.model = model
+        self.base_url = base_url
+        self.api_key = api_key
+        self.provider = provider
+        self.threshold_percent = threshold_percent
+        self.protect_first_n = protect_first_n
+        self.protect_last_n = protect_last_n
+        self.summary_target_ratio = max(0.10, min(summary_target_ratio, 0.80))
+        self.quiet_mode = quiet_mode
+
+        self.context_length = get_model_context_length(
+            model, base_url=base_url, api_key=api_key,
+            config_context_length=config_context_length,
+            provider=provider,
+        )
+        self.threshold_tokens = int(self.context_length * threshold_percent)
+        self.compression_count = 0
+
+        # Derive token budgets: ratio is relative to the threshold, not total context
+        target_tokens = int(self.threshold_tokens * self.summary_target_ratio)
+        self.tail_token_budget = target_tokens
+        self.max_summary_tokens = min(
+            int(self.context_length * 0.05), _SUMMARY_TOKENS_CEILING,
+        )
+
+        if not quiet_mode:
+            logger.info(
+                "Context compressor initialized: model=%s context_length=%d "
+                "threshold=%d (%.0f%%) target_ratio=%.0f%% tail_budget=%d "
+                "provider=%s base_url=%s",
+                model, self.context_length, self.threshold_tokens,
+                threshold_percent * 100, self.summary_target_ratio * 100,
+                self.tail_token_budget,
+                provider or "none", base_url or "none",
+            )
+        self._context_probed = False  # True after a step-down from context error
+
+        self.last_prompt_tokens = 0
+        self.last_completion_tokens = 0
+        self.last_total_tokens = 0
+
+        self.summary_model = summary_model_override or ""
+
+        # Stores the previous compaction summary for iterative updates
+        self._previous_summary: Optional[str] = None
+
+    def update_from_response(self, usage: Dict[str, Any]):
+        """Update tracked token usage from API response."""
+        self.last_prompt_tokens = usage.get("prompt_tokens", 0)
+        self.last_completion_tokens = usage.get("completion_tokens", 0)
+        self.last_total_tokens = usage.get("total_tokens", 0)
+
+    def should_compress(self, prompt_tokens: int = None) -> bool:
+        """Check if context exceeds the compression threshold."""
+        tokens = prompt_tokens if prompt_tokens is not None else self.last_prompt_tokens
+        return tokens >= self.threshold_tokens
+
+    def should_compress_preflight(self, messages: List[Dict[str, Any]]) -> bool:
+        """Quick pre-flight check using rough estimate (before API call)."""
+        rough_estimate = estimate_messages_tokens_rough(messages)
+        return rough_estimate >= self.threshold_tokens
+
+    def get_status(self) -> Dict[str, Any]:
+        """Get current compression status for display/logging."""
+        return {
+            "last_prompt_tokens": self.last_prompt_tokens,
+            "threshold_tokens": self.threshold_tokens,
+            "context_length": self.context_length,
+            "usage_percent": min(100, (self.last_prompt_tokens / self.context_length * 100)) if self.context_length else 0,
+            "compression_count": self.compression_count,
+        }
+
+    # ------------------------------------------------------------------
+    # Tool output pruning (cheap pre-pass, no LLM call)
+    # ------------------------------------------------------------------
+
+    def _prune_old_tool_results(
+        self, messages: List[Dict[str, Any]], protect_tail_count: int,
+    ) -> tuple[List[Dict[str, Any]], int]:
+        """Replace old tool result contents with a short placeholder.
+
+        Walks backward from the end, protecting the most recent
+        ``protect_tail_count`` messages. Older tool results get their
+        content replaced with a placeholder string.
+
+        Returns (pruned_messages, pruned_count).
+        """
+        if not messages:
+            return messages, 0
+
+        result = [m.copy() for m in messages]
+        pruned = 0
+        prune_boundary = len(result) - protect_tail_count
+
+        for i in range(prune_boundary):
+            msg = result[i]
+            if msg.get("role") != "tool":
+                continue
+            content = msg.get("content", "")
+            if not content or content == _PRUNED_TOOL_PLACEHOLDER:
+                continue
+            # Only prune if the content is substantial (>200 chars)
+            if len(content) > 200:
+                result[i] = {**msg, "content": _PRUNED_TOOL_PLACEHOLDER}
+                pruned += 1
+
+        return result, pruned
+
+    # ------------------------------------------------------------------
+    # Summarization
+    # ------------------------------------------------------------------
+
+    def _compute_summary_budget(self, turns_to_summarize: List[Dict[str, Any]]) -> int:
+        """Scale summary token budget with the amount of content being compressed.
+
+        The maximum scales with the model's context window (5% of context,
+        capped at ``_SUMMARY_TOKENS_CEILING``) so large-context models get
+        richer summaries instead of being hard-capped at 8K tokens.
+        """
+        content_tokens = estimate_messages_tokens_rough(turns_to_summarize)
+        budget = int(content_tokens * _SUMMARY_RATIO)
+        return max(_MIN_SUMMARY_TOKENS, min(budget, self.max_summary_tokens))
+
+    def _serialize_for_summary(self, turns: List[Dict[str, Any]]) -> str:
+        """Serialize conversation turns into labeled text for the summarizer.
+
+        Includes tool call arguments and result content (up to 3000 chars
+        per message) so the summarizer can preserve specific details like
+        file paths, commands, and outputs.
+        """
+        parts = []
+        for msg in turns:
+            role = msg.get("role", "unknown")
+            content = msg.get("content") or ""
+
+            # Tool results: keep more content than before (3000 chars)
+            if role == "tool":
+                tool_id = msg.get("tool_call_id", "")
+                if len(content) > 3000:
+                    content = content[:2000] + "\n...[truncated]...\n" + content[-800:]
+                parts.append(f"[TOOL RESULT {tool_id}]: {content}")
+                continue
+
+            # Assistant messages: include tool call names AND arguments
+            if role == "assistant":
+                if len(content) > 3000:
+                    content = content[:2000] + "\n...[truncated]...\n" + content[-800:]
+                tool_calls = msg.get("tool_calls", [])
+                if tool_calls:
+                    tc_parts = []
+                    for tc in tool_calls:
+                        if isinstance(tc, dict):
+                            fn = tc.get("function", {})
+                            name = fn.get("name", "?")
+                            args = fn.get("arguments", "")
+                            # Truncate long arguments but keep enough for context
+                            if len(args) > 500:
+                                args = args[:400] + "..."
+                            tc_parts.append(f"  {name}({args})")
+                        else:
+                            fn = getattr(tc, "function", None)
+                            name = getattr(fn, "name", "?") if fn else "?"
+                            tc_parts.append(f"  {name}(...)")
+                    content += "\n[Tool calls:\n" + "\n".join(tc_parts) + "\n]"
+                parts.append(f"[ASSISTANT]: {content}")
+                continue
+
+            # User and other roles
+            if len(content) > 3000:
+                content = content[:2000] + "\n...[truncated]...\n" + content[-800:]
+            parts.append(f"[{role.upper()}]: {content}")
+
+        return "\n\n".join(parts)
+
+    def _generate_summary(self, turns_to_summarize: List[Dict[str, Any]]) -> Optional[str]:
+        """Generate a structured summary of conversation turns.
+
+        Uses a structured template (Goal, Progress, Decisions, Files, Next Steps)
+        inspired by Pi-mono and OpenCode. When a previous summary exists,
+        generates an iterative update instead of summarizing from scratch.
+
+        Returns None if all attempts fail — the caller should drop
+        the middle turns without a summary rather than inject a useless
+        placeholder.
+        """
+        summary_budget = self._compute_summary_budget(turns_to_summarize)
+        content_to_summarize = self._serialize_for_summary(turns_to_summarize)
+
+        if self._previous_summary:
+            # Iterative update: preserve existing info, add new progress
+            prompt = f"""You are updating a context compaction summary. A previous compaction produced the summary below. New conversation turns have occurred since then and need to be incorporated.
+
+PREVIOUS SUMMARY:
+{self._previous_summary}
+
+NEW TURNS TO INCORPORATE:
+{content_to_summarize}
+
+Update the summary using this exact structure. PRESERVE all existing information that is still relevant. ADD new progress. Move items from "In Progress" to "Done" when completed. Remove information only if it is clearly obsolete.
+
+## Goal
+[What the user is trying to accomplish — preserve from previous summary, update if goal evolved]
+
+## Constraints & Preferences
+[User preferences, coding style, constraints, important decisions — accumulate across compactions]
+
+## Progress
+### Done
+[Completed work — include specific file paths, commands run, results obtained]
+### In Progress
+[Work currently underway]
+### Blocked
+[Any blockers or issues encountered]
+
+## Key Decisions
+[Important technical decisions and why they were made]
+
+## Relevant Files
+[Files read, modified, or created — with brief note on each. Accumulate across compactions.]
+
+## Next Steps
+[What needs to happen next to continue the work]
+
+## Critical Context
+[Any specific values, error messages, configuration details, or data that would be lost without explicit preservation]
+
+Target ~{summary_budget} tokens. Be specific — include file paths, command outputs, error messages, and concrete values rather than vague descriptions.
+
+Write only the summary body. Do not include any preamble or prefix."""
+        else:
+            # First compaction: summarize from scratch
+            prompt = f"""Create a structured handoff summary for a later assistant that will continue this conversation after earlier turns are compacted.
+
+TURNS TO SUMMARIZE:
+{content_to_summarize}
+
+Use this exact structure:
+
+## Goal
+[What the user is trying to accomplish]
+
+## Constraints & Preferences
+[User preferences, coding style, constraints, important decisions]
+
+## Progress
+### Done
+[Completed work — include specific file paths, commands run, results obtained]
+### In Progress
+[Work currently underway]
+### Blocked
+[Any blockers or issues encountered]
+
+## Key Decisions
+[Important technical decisions and why they were made]
+
+## Relevant Files
+[Files read, modified, or created — with brief note on each]
+
+## Next Steps
+[What needs to happen next to continue the work]
+
+## Critical Context
+[Any specific values, error messages, configuration details, or data that would be lost without explicit preservation]
+
+Target ~{summary_budget} tokens. Be specific — include file paths, command outputs, error messages, and concrete values rather than vague descriptions. The goal is to prevent the next assistant from repeating work or losing important details.
+
+Write only the summary body. Do not include any preamble or prefix."""
+
+        try:
+            call_kwargs = {
+                "task": "compression",
+                "messages": [{"role": "user", "content": prompt}],
+                "temperature": 0.3,
+                "max_tokens": summary_budget * 2,
+                # timeout resolved from auxiliary.compression.timeout config by call_llm
+            }
+            if self.summary_model:
+                call_kwargs["model"] = self.summary_model
+            response = call_llm(**call_kwargs)
+            content = response.choices[0].message.content
+            # Handle cases where content is not a string (e.g., dict from llama.cpp)
+            if not isinstance(content, str):
+                content = str(content) if content else ""
+            summary = content.strip()
+            # Store for iterative updates on next compaction
+            self._previous_summary = summary
+            return self._with_summary_prefix(summary)
+        except RuntimeError:
+            logging.warning("Context compression: no provider available for "
+                            "summary. Middle turns will be dropped without summary.")
+            return None
+        except Exception as e:
+            logging.warning("Failed to generate context summary: %s", e)
+            return None
+
+    @staticmethod
+    def _with_summary_prefix(summary: str) -> str:
+        """Normalize summary text to the current compaction handoff format."""
+        text = (summary or "").strip()
+        for prefix in (LEGACY_SUMMARY_PREFIX, SUMMARY_PREFIX):
+            if text.startswith(prefix):
+                text = text[len(prefix):].lstrip()
+                break
+        return f"{SUMMARY_PREFIX}\n{text}" if text else SUMMARY_PREFIX
+
+    # ------------------------------------------------------------------
+    # Tool-call / tool-result pair integrity helpers
+    # ------------------------------------------------------------------
+
+    @staticmethod
+    def _get_tool_call_id(tc) -> str:
+        """Extract the call ID from a tool_call entry (dict or SimpleNamespace)."""
+        if isinstance(tc, dict):
+            return tc.get("id", "")
+        return getattr(tc, "id", "") or ""
+
+    def _sanitize_tool_pairs(self, messages: List[Dict[str, Any]]) -> List[Dict[str, Any]]:
+        """Fix orphaned tool_call / tool_result pairs after compression.
+
+        Two failure modes:
+        1. A tool *result* references a call_id whose assistant tool_call was
+           removed (summarized/truncated).  The API rejects this with
+           "No tool call found for function call output with call_id ...".
+        2. An assistant message has tool_calls whose results were dropped.
+           The API rejects this because every tool_call must be followed by
+           a tool result with the matching call_id.
+
+        This method removes orphaned results and inserts stub results for
+        orphaned calls so the message list is always well-formed.
+        """
+        surviving_call_ids: set = set()
+        for msg in messages:
+            if msg.get("role") == "assistant":
+                for tc in msg.get("tool_calls") or []:
+                    cid = self._get_tool_call_id(tc)
+                    if cid:
+                        surviving_call_ids.add(cid)
+
+        result_call_ids: set = set()
+        for msg in messages:
+            if msg.get("role") == "tool":
+                cid = msg.get("tool_call_id")
+                if cid:
+                    result_call_ids.add(cid)
+
+        # 1. Remove tool results whose call_id has no matching assistant tool_call
+        orphaned_results = result_call_ids - surviving_call_ids
+        if orphaned_results:
+            messages = [
+                m for m in messages
+                if not (m.get("role") == "tool" and m.get("tool_call_id") in orphaned_results)
+            ]
+            if not self.quiet_mode:
+                logger.info("Compression sanitizer: removed %d orphaned tool result(s)", len(orphaned_results))
+
+        # 2. Add stub results for assistant tool_calls whose results were dropped
+        missing_results = surviving_call_ids - result_call_ids
+        if missing_results:
+            patched: List[Dict[str, Any]] = []
+            for msg in messages:
+                patched.append(msg)
+                if msg.get("role") == "assistant":
+                    for tc in msg.get("tool_calls") or []:
+                        cid = self._get_tool_call_id(tc)
+                        if cid in missing_results:
+                            patched.append({
+                                "role": "tool",
+                                "content": "[Result from earlier conversation — see context summary above]",
+                                "tool_call_id": cid,
+                            })
+            messages = patched
+            if not self.quiet_mode:
+                logger.info("Compression sanitizer: added %d stub tool result(s)", len(missing_results))
+
+        return messages
+
+    def _align_boundary_forward(self, messages: List[Dict[str, Any]], idx: int) -> int:
+        """Push a compress-start boundary forward past any orphan tool results.
+
+        If ``messages[idx]`` is a tool result, slide forward until we hit a
+        non-tool message so we don't start the summarised region mid-group.
+        """
+        while idx < len(messages) and messages[idx].get("role") == "tool":
+            idx += 1
+        return idx
+
+    def _align_boundary_backward(self, messages: List[Dict[str, Any]], idx: int) -> int:
+        """Pull a compress-end boundary backward to avoid splitting a
+        tool_call / result group.
+
+        If the boundary falls in the middle of a tool-result group (i.e.
+        there are consecutive tool messages before ``idx``), walk backward
+        past all of them to find the parent assistant message.  If found,
+        move the boundary before the assistant so the entire
+        assistant + tool_results group is included in the summarised region
+        rather than being split (which causes silent data loss when
+        ``_sanitize_tool_pairs`` removes the orphaned tail results).
+        """
+        if idx <= 0 or idx >= len(messages):
+            return idx
+        # Walk backward past consecutive tool results
+        check = idx - 1
+        while check >= 0 and messages[check].get("role") == "tool":
+            check -= 1
+        # If we landed on the parent assistant with tool_calls, pull the
+        # boundary before it so the whole group gets summarised together.
+        if check >= 0 and messages[check].get("role") == "assistant" and messages[check].get("tool_calls"):
+            idx = check
+        return idx
+
+    # ------------------------------------------------------------------
+    # Tail protection by token budget
+    # ------------------------------------------------------------------
+
+    def _find_tail_cut_by_tokens(
+        self, messages: List[Dict[str, Any]], head_end: int,
+        token_budget: int | None = None,
+    ) -> int:
+        """Walk backward from the end of messages, accumulating tokens until
+        the budget is reached. Returns the index where the tail starts.
+
+        ``token_budget`` defaults to ``self.tail_token_budget`` which is
+        derived from ``summary_target_ratio * context_length``, so it
+        scales automatically with the model's context window.
+
+        Never cuts inside a tool_call/result group. Falls back to the old
+        ``protect_last_n`` if the budget would protect fewer messages.
+        """
+        if token_budget is None:
+            token_budget = self.tail_token_budget
+        n = len(messages)
+        min_tail = self.protect_last_n
+        accumulated = 0
+        cut_idx = n  # start from beyond the end
+
+        for i in range(n - 1, head_end - 1, -1):
+            msg = messages[i]
+            content = msg.get("content") or ""
+            msg_tokens = len(content) // _CHARS_PER_TOKEN + 10  # +10 for role/metadata
+            # Include tool call arguments in estimate
+            for tc in msg.get("tool_calls") or []:
+                if isinstance(tc, dict):
+                    args = tc.get("function", {}).get("arguments", "")
+                    msg_tokens += len(args) // _CHARS_PER_TOKEN
+            if accumulated + msg_tokens > token_budget and (n - i) >= min_tail:
+                break
+            accumulated += msg_tokens
+            cut_idx = i
+
+        # Ensure we protect at least protect_last_n messages
+        fallback_cut = n - min_tail
+        if cut_idx > fallback_cut:
+            cut_idx = fallback_cut
+
+        # If the token budget would protect everything (small conversations),
+        # fall back to the fixed protect_last_n approach so compression can
+        # still remove middle turns.
+        if cut_idx <= head_end:
+            cut_idx = fallback_cut
+
+        # Align to avoid splitting tool groups
+        cut_idx = self._align_boundary_backward(messages, cut_idx)
+
+        return max(cut_idx, head_end + 1)
+
+    # ------------------------------------------------------------------
+    # Main compression entry point
+    # ------------------------------------------------------------------
+
+    def compress(self, messages: List[Dict[str, Any]], current_tokens: int = None) -> List[Dict[str, Any]]:
+        """Compress conversation messages by summarizing middle turns.
+
+        Algorithm:
+          1. Prune old tool results (cheap pre-pass, no LLM call)
+          2. Protect head messages (system prompt + first exchange)
+          3. Find tail boundary by token budget (~20K tokens of recent context)
+          4. Summarize middle turns with structured LLM prompt
+          5. On re-compression, iteratively update the previous summary
+
+        After compression, orphaned tool_call / tool_result pairs are cleaned
+        up so the API never receives mismatched IDs.
+        """
+        n_messages = len(messages)
+        if n_messages <= self.protect_first_n + self.protect_last_n + 1:
+            if not self.quiet_mode:
+                logger.warning(
+                    "Cannot compress: only %d messages (need > %d)",
+                    n_messages,
+                    self.protect_first_n + self.protect_last_n + 1,
+                )
+            return messages
+
+        display_tokens = current_tokens if current_tokens else self.last_prompt_tokens or estimate_messages_tokens_rough(messages)
+
+        # Phase 1: Prune old tool results (cheap, no LLM call)
+        messages, pruned_count = self._prune_old_tool_results(
+            messages, protect_tail_count=self.protect_last_n * 3,
+        )
+        if pruned_count and not self.quiet_mode:
+            logger.info("Pre-compression: pruned %d old tool result(s)", pruned_count)
+
+        # Phase 2: Determine boundaries
+        compress_start = self.protect_first_n
+        compress_start = self._align_boundary_forward(messages, compress_start)
+
+        # Use token-budget tail protection instead of fixed message count
+        compress_end = self._find_tail_cut_by_tokens(messages, compress_start)
+
+        if compress_start >= compress_end:
+            return messages
+
+        turns_to_summarize = messages[compress_start:compress_end]
+
+        if not self.quiet_mode:
+            logger.info(
+                "Context compression triggered (%d tokens >= %d threshold)",
+                display_tokens,
+                self.threshold_tokens,
+            )
+            logger.info(
+                "Model context limit: %d tokens (%.0f%% = %d)",
+                self.context_length,
+                self.threshold_percent * 100,
+                self.threshold_tokens,
+            )
+            tail_msgs = n_messages - compress_end
+            logger.info(
+                "Summarizing turns %d-%d (%d turns), protecting %d head + %d tail messages",
+                compress_start + 1,
+                compress_end,
+                len(turns_to_summarize),
+                compress_start,
+                tail_msgs,
+            )
+
+        # Phase 3: Generate structured summary
+        summary = self._generate_summary(turns_to_summarize)
+
+        # Phase 4: Assemble compressed message list
+        compressed = []
+        for i in range(compress_start):
+            msg = messages[i].copy()
+            if i == 0 and msg.get("role") == "system" and self.compression_count == 0:
+                msg["content"] = (
+                    (msg.get("content") or "")
+                    + "\n\n[Note: Some earlier conversation turns have been compacted into a handoff summary to preserve context space. The current session state may still reflect earlier work, so build on that summary and state rather than re-doing work.]"
+                )
+            compressed.append(msg)
+
+        _merge_summary_into_tail = False
+        if summary:
+            last_head_role = messages[compress_start - 1].get("role", "user") if compress_start > 0 else "user"
+            first_tail_role = messages[compress_end].get("role", "user") if compress_end < n_messages else "user"
+            # Pick a role that avoids consecutive same-role with both neighbors.
+            # Priority: avoid colliding with head (already committed), then tail.
+            if last_head_role in ("assistant", "tool"):
+                summary_role = "user"
+            else:
+                summary_role = "assistant"
+            # If the chosen role collides with the tail AND flipping wouldn't
+            # collide with the head, flip it.
+            if summary_role == first_tail_role:
+                flipped = "assistant" if summary_role == "user" else "user"
+                if flipped != last_head_role:
+                    summary_role = flipped
+                else:
+                    # Both roles would create consecutive same-role messages
+                    # (e.g. head=assistant, tail=user — neither role works).
+                    # Merge the summary into the first tail message instead
+                    # of inserting a standalone message that breaks alternation.
+                    _merge_summary_into_tail = True
+            if not _merge_summary_into_tail:
+                compressed.append({"role": summary_role, "content": summary})
+        else:
+            if not self.quiet_mode:
+                logger.warning("No summary model available — middle turns dropped without summary")
+
+        for i in range(compress_end, n_messages):
+            msg = messages[i].copy()
+            if _merge_summary_into_tail and i == compress_end:
+                original = msg.get("content") or ""
+                msg["content"] = summary + "\n\n" + original
+                _merge_summary_into_tail = False
+            compressed.append(msg)
+
+        self.compression_count += 1
+
+        compressed = self._sanitize_tool_pairs(compressed)
+
+        if not self.quiet_mode:
+            new_estimate = estimate_messages_tokens_rough(compressed)
+            saved_estimate = display_tokens - new_estimate
+            logger.info(
+                "Compressed: %d -> %d messages (~%d tokens saved)",
+                n_messages,
+                len(compressed),
+                saved_estimate,
+            )
+            logger.info("Compression #%d complete", self.compression_count)
+
+        return compressed

agent/context_references.py

@@ -0,0 +1,492 @@
+from __future__ import annotations
+
+import asyncio
+import inspect
+import json
+import mimetypes
+import os
+import re
+import subprocess
+from dataclasses import dataclass, field
+from pathlib import Path
+from typing import Awaitable, Callable
+
+from agent.model_metadata import estimate_tokens_rough
+
+REFERENCE_PATTERN = re.compile(
+    r"(?<![\w/])@(?:(?P<simple>diff|staged)\b|(?P<kind>file|folder|git|url):(?P<value>\S+))"
+)
+TRAILING_PUNCTUATION = ",.;!?"
+_SENSITIVE_HOME_DIRS = (".ssh", ".aws", ".gnupg", ".kube", ".docker", ".azure", ".config/gh")
+_SENSITIVE_HERMES_DIRS = (Path("skills") / ".hub",)
+_SENSITIVE_HOME_FILES = (
+    Path(".ssh") / "authorized_keys",
+    Path(".ssh") / "id_rsa",
+    Path(".ssh") / "id_ed25519",
+    Path(".ssh") / "config",
+    Path(".bashrc"),
+    Path(".zshrc"),
+    Path(".profile"),
+    Path(".bash_profile"),
+    Path(".zprofile"),
+    Path(".netrc"),
+    Path(".pgpass"),
+    Path(".npmrc"),
+    Path(".pypirc"),
+)
+
+
+@dataclass(frozen=True)
+class ContextReference:
+    raw: str
+    kind: str
+    target: str
+    start: int
+    end: int
+    line_start: int | None = None
+    line_end: int | None = None
+
+
+@dataclass
+class ContextReferenceResult:
+    message: str
+    original_message: str
+    references: list[ContextReference] = field(default_factory=list)
+    warnings: list[str] = field(default_factory=list)
+    injected_tokens: int = 0
+    expanded: bool = False
+    blocked: bool = False
+
+
+def parse_context_references(message: str) -> list[ContextReference]:
+    refs: list[ContextReference] = []
+    if not message:
+        return refs
+
+    for match in REFERENCE_PATTERN.finditer(message):
+        simple = match.group("simple")
+        if simple:
+            refs.append(
+                ContextReference(
+                    raw=match.group(0),
+                    kind=simple,
+                    target="",
+                    start=match.start(),
+                    end=match.end(),
+                )
+            )
+            continue
+
+        kind = match.group("kind")
+        value = _strip_trailing_punctuation(match.group("value") or "")
+        line_start = None
+        line_end = None
+        target = value
+
+        if kind == "file":
+            range_match = re.match(r"^(?P<path>.+?):(?P<start>\d+)(?:-(?P<end>\d+))?$", value)
+            if range_match:
+                target = range_match.group("path")
+                line_start = int(range_match.group("start"))
+                line_end = int(range_match.group("end") or range_match.group("start"))
+
+        refs.append(
+            ContextReference(
+                raw=match.group(0),
+                kind=kind,
+                target=target,
+                start=match.start(),
+                end=match.end(),
+                line_start=line_start,
+                line_end=line_end,
+            )
+        )
+
+    return refs
+
+
+def preprocess_context_references(
+    message: str,
+    *,
+    cwd: str | Path,
+    context_length: int,
+    url_fetcher: Callable[[str], str | Awaitable[str]] | None = None,
+    allowed_root: str | Path | None = None,
+) -> ContextReferenceResult:
+    coro = preprocess_context_references_async(
+        message,
+        cwd=cwd,
+        context_length=context_length,
+        url_fetcher=url_fetcher,
+        allowed_root=allowed_root,
+    )
+    # Safe for both CLI (no loop) and gateway (loop already running).
+    try:
+        loop = asyncio.get_running_loop()
+    except RuntimeError:
+        loop = None
+    if loop and loop.is_running():
+        import concurrent.futures
+        with concurrent.futures.ThreadPoolExecutor(max_workers=1) as pool:
+            return pool.submit(asyncio.run, coro).result()
+    return asyncio.run(coro)
+
+
+async def preprocess_context_references_async(
+    message: str,
+    *,
+    cwd: str | Path,
+    context_length: int,
+    url_fetcher: Callable[[str], str | Awaitable[str]] | None = None,
+    allowed_root: str | Path | None = None,
+) -> ContextReferenceResult:
+    refs = parse_context_references(message)
+    if not refs:
+        return ContextReferenceResult(message=message, original_message=message)
+
+    cwd_path = Path(cwd).expanduser().resolve()
+    # Default to the current working directory so @ references cannot escape
+    # the active workspace unless a caller explicitly widens the root.
+    allowed_root_path = (
+        Path(allowed_root).expanduser().resolve() if allowed_root is not None else cwd_path
+    )
+    warnings: list[str] = []
+    blocks: list[str] = []
+    injected_tokens = 0
+
+    for ref in refs:
+        warning, block = await _expand_reference(
+            ref,
+            cwd_path,
+            url_fetcher=url_fetcher,
+            allowed_root=allowed_root_path,
+        )
+        if warning:
+            warnings.append(warning)
+        if block:
+            blocks.append(block)
+            injected_tokens += estimate_tokens_rough(block)
+
+    hard_limit = max(1, int(context_length * 0.50))
+    soft_limit = max(1, int(context_length * 0.25))
+    if injected_tokens > hard_limit:
+        warnings.append(
+            f"@ context injection refused: {injected_tokens} tokens exceeds the 50% hard limit ({hard_limit})."
+        )
+        return ContextReferenceResult(
+            message=message,
+            original_message=message,
+            references=refs,
+            warnings=warnings,
+            injected_tokens=injected_tokens,
+            expanded=False,
+            blocked=True,
+        )
+
+    if injected_tokens > soft_limit:
+        warnings.append(
+            f"@ context injection warning: {injected_tokens} tokens exceeds the 25% soft limit ({soft_limit})."
+        )
+
+    stripped = _remove_reference_tokens(message, refs)
+    final = stripped
+    if warnings:
+        final = f"{final}\n\n--- Context Warnings ---\n" + "\n".join(f"- {warning}" for warning in warnings)
+    if blocks:
+        final = f"{final}\n\n--- Attached Context ---\n\n" + "\n\n".join(blocks)
+
+    return ContextReferenceResult(
+        message=final.strip(),
+        original_message=message,
+        references=refs,
+        warnings=warnings,
+        injected_tokens=injected_tokens,
+        expanded=bool(blocks or warnings),
+        blocked=False,
+    )
+
+
+async def _expand_reference(
+    ref: ContextReference,
+    cwd: Path,
+    *,
+    url_fetcher: Callable[[str], str | Awaitable[str]] | None = None,
+    allowed_root: Path | None = None,
+) -> tuple[str | None, str | None]:
+    try:
+        if ref.kind == "file":
+            return _expand_file_reference(ref, cwd, allowed_root=allowed_root)
+        if ref.kind == "folder":
+            return _expand_folder_reference(ref, cwd, allowed_root=allowed_root)
+        if ref.kind == "diff":
+            return _expand_git_reference(ref, cwd, ["diff"], "git diff")
+        if ref.kind == "staged":
+            return _expand_git_reference(ref, cwd, ["diff", "--staged"], "git diff --staged")
+        if ref.kind == "git":
+            count = max(1, min(int(ref.target or "1"), 10))
+            return _expand_git_reference(ref, cwd, ["log", f"-{count}", "-p"], f"git log -{count} -p")
+        if ref.kind == "url":
+            content = await _fetch_url_content(ref.target, url_fetcher=url_fetcher)
+            if not content:
+                return f"{ref.raw}: no content extracted", None
+            return None, f"🌐 {ref.raw} ({estimate_tokens_rough(content)} tokens)\n{content}"
+    except Exception as exc:
+        return f"{ref.raw}: {exc}", None
+
+    return f"{ref.raw}: unsupported reference type", None
+
+
+def _expand_file_reference(
+    ref: ContextReference,
+    cwd: Path,
+    *,
+    allowed_root: Path | None = None,
+) -> tuple[str | None, str | None]:
+    path = _resolve_path(cwd, ref.target, allowed_root=allowed_root)
+    _ensure_reference_path_allowed(path)
+    if not path.exists():
+        return f"{ref.raw}: file not found", None
+    if not path.is_file():
+        return f"{ref.raw}: path is not a file", None
+    if _is_binary_file(path):
+        return f"{ref.raw}: binary files are not supported", None
+
+    text = path.read_text(encoding="utf-8")
+    if ref.line_start is not None:
+        lines = text.splitlines()
+        start_idx = max(ref.line_start - 1, 0)
+        end_idx = min(ref.line_end or ref.line_start, len(lines))
+        text = "\n".join(lines[start_idx:end_idx])
+
+    lang = _code_fence_language(path)
+    label = ref.raw
+    return None, f"📄 {label} ({estimate_tokens_rough(text)} tokens)\n```{lang}\n{text}\n```"
+
+
+def _expand_folder_reference(
+    ref: ContextReference,
+    cwd: Path,
+    *,
+    allowed_root: Path | None = None,
+) -> tuple[str | None, str | None]:
+    path = _resolve_path(cwd, ref.target, allowed_root=allowed_root)
+    _ensure_reference_path_allowed(path)
+    if not path.exists():
+        return f"{ref.raw}: folder not found", None
+    if not path.is_dir():
+        return f"{ref.raw}: path is not a folder", None
+
+    listing = _build_folder_listing(path, cwd)
+    return None, f"📁 {ref.raw} ({estimate_tokens_rough(listing)} tokens)\n{listing}"
+
+
+def _expand_git_reference(
+    ref: ContextReference,
+    cwd: Path,
+    args: list[str],
+    label: str,
+) -> tuple[str | None, str | None]:
+    try:
+        result = subprocess.run(
+            ["git", *args],
+            cwd=cwd,
+            capture_output=True,
+            text=True,
+            timeout=30,
+        )
+    except subprocess.TimeoutExpired:
+        return f"{ref.raw}: git command timed out (30s)", None
+    if result.returncode != 0:
+        stderr = (result.stderr or "").strip() or "git command failed"
+        return f"{ref.raw}: {stderr}", None
+    content = result.stdout.strip()
+    if not content:
+        content = "(no output)"
+    return None, f"🧾 {label} ({estimate_tokens_rough(content)} tokens)\n```diff\n{content}\n```"
+
+
+async def _fetch_url_content(
+    url: str,
+    *,
+    url_fetcher: Callable[[str], str | Awaitable[str]] | None = None,
+) -> str:
+    fetcher = url_fetcher or _default_url_fetcher
+    content = fetcher(url)
+    if inspect.isawaitable(content):
+        content = await content
+    return str(content or "").strip()
+
+
+async def _default_url_fetcher(url: str) -> str:
+    from tools.web_tools import web_extract_tool
+
+    raw = await web_extract_tool([url], format="markdown", use_llm_processing=True)
+    payload = json.loads(raw)
+    docs = payload.get("data", {}).get("documents", [])
+    if not docs:
+        return ""
+    doc = docs[0]
+    return str(doc.get("content") or doc.get("raw_content") or "").strip()
+
+
+def _resolve_path(cwd: Path, target: str, *, allowed_root: Path | None = None) -> Path:
+    path = Path(os.path.expanduser(target))
+    if not path.is_absolute():
+        path = cwd / path
+    resolved = path.resolve()
+    if allowed_root is not None:
+        try:
+            resolved.relative_to(allowed_root)
+        except ValueError as exc:
+            raise ValueError("path is outside the allowed workspace") from exc
+    return resolved
+
+
+def _ensure_reference_path_allowed(path: Path) -> None:
+    home = Path(os.path.expanduser("~")).resolve()
+    hermes_home = Path(
+        os.getenv("HERMES_HOME", str(home / ".hermes"))
+    ).expanduser().resolve()
+
+    blocked_exact = {home / rel for rel in _SENSITIVE_HOME_FILES}
+    blocked_exact.add(hermes_home / ".env")
+    blocked_dirs = [home / rel for rel in _SENSITIVE_HOME_DIRS]
+    blocked_dirs.extend(hermes_home / rel for rel in _SENSITIVE_HERMES_DIRS)
+
+    if path in blocked_exact:
+        raise ValueError("path is a sensitive credential file and cannot be attached")
+
+    for blocked_dir in blocked_dirs:
+        try:
+            path.relative_to(blocked_dir)
+        except ValueError:
+            continue
+        raise ValueError("path is a sensitive credential or internal Hermes path and cannot be attached")
+
+
+def _strip_trailing_punctuation(value: str) -> str:
+    stripped = value.rstrip(TRAILING_PUNCTUATION)
+    while stripped.endswith((")", "]", "}")):
+        closer = stripped[-1]
+        opener = {")": "(", "]": "[", "}": "{"}[closer]
+        if stripped.count(closer) > stripped.count(opener):
+            stripped = stripped[:-1]
+            continue
+        break
+    return stripped
+
+
+def _remove_reference_tokens(message: str, refs: list[ContextReference]) -> str:
+    pieces: list[str] = []
+    cursor = 0
+    for ref in refs:
+        pieces.append(message[cursor:ref.start])
+        cursor = ref.end
+    pieces.append(message[cursor:])
+    text = "".join(pieces)
+    text = re.sub(r"\s{2,}", " ", text)
+    text = re.sub(r"\s+([,.;:!?])", r"\1", text)
+    return text.strip()
+
+
+def _is_binary_file(path: Path) -> bool:
+    mime, _ = mimetypes.guess_type(path.name)
+    if mime and not mime.startswith("text/") and not any(
+        path.name.endswith(ext) for ext in (".py", ".md", ".txt", ".json", ".yaml", ".yml", ".toml", ".js", ".ts")
+    ):
+        return True
+    chunk = path.read_bytes()[:4096]
+    return b"\x00" in chunk
+
+
+def _build_folder_listing(path: Path, cwd: Path, limit: int = 200) -> str:
+    lines = [f"{path.relative_to(cwd)}/"]
+    entries = _iter_visible_entries(path, cwd, limit=limit)
+    for entry in entries:
+        rel = entry.relative_to(cwd)
+        indent = "  " * max(len(rel.parts) - len(path.relative_to(cwd).parts) - 1, 0)
+        if entry.is_dir():
+            lines.append(f"{indent}- {entry.name}/")
+        else:
+            meta = _file_metadata(entry)
+            lines.append(f"{indent}- {entry.name} ({meta})")
+    if len(entries) >= limit:
+        lines.append("- ...")
+    return "\n".join(lines)
+
+
+def _iter_visible_entries(path: Path, cwd: Path, limit: int) -> list[Path]:
+    rg_entries = _rg_files(path, cwd, limit=limit)
+    if rg_entries is not None:
+        output: list[Path] = []
+        seen_dirs: set[Path] = set()
+        for rel in rg_entries:
+            full = cwd / rel
+            for parent in full.parents:
+                if parent == cwd or parent in seen_dirs or path not in {parent, *parent.parents}:
+                    continue
+                seen_dirs.add(parent)
+                output.append(parent)
+            output.append(full)
+        return sorted({p for p in output if p.exists()}, key=lambda p: (not p.is_dir(), str(p)))
+
+    output = []
+    for root, dirs, files in os.walk(path):
+        dirs[:] = sorted(d for d in dirs if not d.startswith(".") and d != "__pycache__")
+        files = sorted(f for f in files if not f.startswith("."))
+        root_path = Path(root)
+        for d in dirs:
+            output.append(root_path / d)
+            if len(output) >= limit:
+                return output
+        for f in files:
+            output.append(root_path / f)
+            if len(output) >= limit:
+                return output
+    return output
+
+
+def _rg_files(path: Path, cwd: Path, limit: int) -> list[Path] | None:
+    try:
+        result = subprocess.run(
+            ["rg", "--files", str(path.relative_to(cwd))],
+            cwd=cwd,
+            capture_output=True,
+            text=True,
+            timeout=10,
+        )
+    except FileNotFoundError:
+        return None
+    except subprocess.TimeoutExpired:
+        return None
+    if result.returncode != 0:
+        return None
+    files = [Path(line.strip()) for line in result.stdout.splitlines() if line.strip()]
+    return files[:limit]
+
+
+def _file_metadata(path: Path) -> str:
+    if _is_binary_file(path):
+        return f"{path.stat().st_size} bytes"
+    try:
+        line_count = path.read_text(encoding="utf-8").count("\n") + 1
+    except Exception:
+        return f"{path.stat().st_size} bytes"
+    return f"{line_count} lines"
+
+
+def _code_fence_language(path: Path) -> str:
+    mapping = {
+        ".py": "python",
+        ".js": "javascript",
+        ".ts": "typescript",
+        ".tsx": "tsx",
+        ".jsx": "jsx",
+        ".json": "json",
+        ".md": "markdown",
+        ".sh": "bash",
+        ".yml": "yaml",
+        ".yaml": "yaml",
+        ".toml": "toml",
+    }
+    return mapping.get(path.suffix.lower(), "")

agent/copilot_acp_client.py

@@ -0,0 +1,447 @@
+"""OpenAI-compatible shim that forwards Hermes requests to `copilot --acp`.
+
+This adapter lets Hermes treat the GitHub Copilot ACP server as a chat-style
+backend. Each request starts a short-lived ACP session, sends the formatted
+conversation as a single prompt, collects text chunks, and converts the result
+back into the minimal shape Hermes expects from an OpenAI client.
+"""
+
+from __future__ import annotations
+
+import json
+import os
+import queue
+import shlex
+import subprocess
+import threading
+import time
+from collections import deque
+from pathlib import Path
+from types import SimpleNamespace
+from typing import Any
+
+ACP_MARKER_BASE_URL = "acp://copilot"
+_DEFAULT_TIMEOUT_SECONDS = 900.0
+
+
+def _resolve_command() -> str:
+    return (
+        os.getenv("HERMES_COPILOT_ACP_COMMAND", "").strip()
+        or os.getenv("COPILOT_CLI_PATH", "").strip()
+        or "copilot"
+    )
+
+
+def _resolve_args() -> list[str]:
+    raw = os.getenv("HERMES_COPILOT_ACP_ARGS", "").strip()
+    if not raw:
+        return ["--acp", "--stdio"]
+    return shlex.split(raw)
+
+
+def _jsonrpc_error(message_id: Any, code: int, message: str) -> dict[str, Any]:
+    return {
+        "jsonrpc": "2.0",
+        "id": message_id,
+        "error": {
+            "code": code,
+            "message": message,
+        },
+    }
+
+
+def _format_messages_as_prompt(messages: list[dict[str, Any]], model: str | None = None) -> str:
+    sections: list[str] = [
+        "You are being used as the active ACP agent backend for Hermes.",
+        "Use your own ACP capabilities and respond directly in natural language.",
+        "Do not emit OpenAI tool-call JSON.",
+    ]
+    if model:
+        sections.append(f"Hermes requested model hint: {model}")
+
+    transcript: list[str] = []
+    for message in messages:
+        if not isinstance(message, dict):
+            continue
+        role = str(message.get("role") or "unknown").strip().lower()
+        if role == "tool":
+            role = "tool"
+        elif role not in {"system", "user", "assistant"}:
+            role = "context"
+
+        content = message.get("content")
+        rendered = _render_message_content(content)
+        if not rendered:
+            continue
+
+        label = {
+            "system": "System",
+            "user": "User",
+            "assistant": "Assistant",
+            "tool": "Tool",
+            "context": "Context",
+        }.get(role, role.title())
+        transcript.append(f"{label}:\n{rendered}")
+
+    if transcript:
+        sections.append("Conversation transcript:\n\n" + "\n\n".join(transcript))
+
+    sections.append("Continue the conversation from the latest user request.")
+    return "\n\n".join(section.strip() for section in sections if section and section.strip())
+
+
+def _render_message_content(content: Any) -> str:
+    if content is None:
+        return ""
+    if isinstance(content, str):
+        return content.strip()
+    if isinstance(content, dict):
+        if "text" in content:
+            return str(content.get("text") or "").strip()
+        if "content" in content and isinstance(content.get("content"), str):
+            return str(content.get("content") or "").strip()
+        return json.dumps(content, ensure_ascii=True)
+    if isinstance(content, list):
+        parts: list[str] = []
+        for item in content:
+            if isinstance(item, str):
+                parts.append(item)
+            elif isinstance(item, dict):
+                text = item.get("text")
+                if isinstance(text, str) and text.strip():
+                    parts.append(text.strip())
+        return "\n".join(parts).strip()
+    return str(content).strip()
+
+
+def _ensure_path_within_cwd(path_text: str, cwd: str) -> Path:
+    candidate = Path(path_text)
+    if not candidate.is_absolute():
+        raise PermissionError("ACP file-system paths must be absolute.")
+    resolved = candidate.resolve()
+    root = Path(cwd).resolve()
+    try:
+        resolved.relative_to(root)
+    except ValueError as exc:
+        raise PermissionError(f"Path '{resolved}' is outside the session cwd '{root}'.") from exc
+    return resolved
+
+
+class _ACPChatCompletions:
+    def __init__(self, client: "CopilotACPClient"):
+        self._client = client
+
+    def create(self, **kwargs: Any) -> Any:
+        return self._client._create_chat_completion(**kwargs)
+
+
+class _ACPChatNamespace:
+    def __init__(self, client: "CopilotACPClient"):
+        self.completions = _ACPChatCompletions(client)
+
+
+class CopilotACPClient:
+    """Minimal OpenAI-client-compatible facade for Copilot ACP."""
+
+    def __init__(
+        self,
+        *,
+        api_key: str | None = None,
+        base_url: str | None = None,
+        default_headers: dict[str, str] | None = None,
+        acp_command: str | None = None,
+        acp_args: list[str] | None = None,
+        acp_cwd: str | None = None,
+        command: str | None = None,
+        args: list[str] | None = None,
+        **_: Any,
+    ):
+        self.api_key = api_key or "copilot-acp"
+        self.base_url = base_url or ACP_MARKER_BASE_URL
+        self._default_headers = dict(default_headers or {})
+        self._acp_command = acp_command or command or _resolve_command()
+        self._acp_args = list(acp_args or args or _resolve_args())
+        self._acp_cwd = str(Path(acp_cwd or os.getcwd()).resolve())
+        self.chat = _ACPChatNamespace(self)
+        self.is_closed = False
+        self._active_process: subprocess.Popen[str] | None = None
+        self._active_process_lock = threading.Lock()
+
+    def close(self) -> None:
+        proc: subprocess.Popen[str] | None
+        with self._active_process_lock:
+            proc = self._active_process
+            self._active_process = None
+        self.is_closed = True
+        if proc is None:
+            return
+        try:
+            proc.terminate()
+            proc.wait(timeout=2)
+        except Exception:
+            try:
+                proc.kill()
+            except Exception:
+                pass
+
+    def _create_chat_completion(
+        self,
+        *,
+        model: str | None = None,
+        messages: list[dict[str, Any]] | None = None,
+        timeout: float | None = None,
+        **_: Any,
+    ) -> Any:
+        prompt_text = _format_messages_as_prompt(messages or [], model=model)
+        response_text, reasoning_text = self._run_prompt(
+            prompt_text,
+            timeout_seconds=float(timeout or _DEFAULT_TIMEOUT_SECONDS),
+        )
+
+        usage = SimpleNamespace(
+            prompt_tokens=0,
+            completion_tokens=0,
+            total_tokens=0,
+            prompt_tokens_details=SimpleNamespace(cached_tokens=0),
+        )
+        assistant_message = SimpleNamespace(
+            content=response_text,
+            tool_calls=[],
+            reasoning=reasoning_text or None,
+            reasoning_content=reasoning_text or None,
+            reasoning_details=None,
+        )
+        choice = SimpleNamespace(message=assistant_message, finish_reason="stop")
+        return SimpleNamespace(
+            choices=[choice],
+            usage=usage,
+            model=model or "copilot-acp",
+        )
+
+    def _run_prompt(self, prompt_text: str, *, timeout_seconds: float) -> tuple[str, str]:
+        try:
+            proc = subprocess.Popen(
+                [self._acp_command] + self._acp_args,
+                stdin=subprocess.PIPE,
+                stdout=subprocess.PIPE,
+                stderr=subprocess.PIPE,
+                text=True,
+                bufsize=1,
+                cwd=self._acp_cwd,
+            )
+        except FileNotFoundError as exc:
+            raise RuntimeError(
+                f"Could not start Copilot ACP command '{self._acp_command}'. "
+                "Install GitHub Copilot CLI or set HERMES_COPILOT_ACP_COMMAND/COPILOT_CLI_PATH."
+            ) from exc
+
+        if proc.stdin is None or proc.stdout is None:
+            proc.kill()
+            raise RuntimeError("Copilot ACP process did not expose stdin/stdout pipes.")
+
+        self.is_closed = False
+        with self._active_process_lock:
+            self._active_process = proc
+
+        inbox: queue.Queue[dict[str, Any]] = queue.Queue()
+        stderr_tail: deque[str] = deque(maxlen=40)
+
+        def _stdout_reader() -> None:
+            for line in proc.stdout:
+                try:
+                    inbox.put(json.loads(line))
+                except Exception:
+                    inbox.put({"raw": line.rstrip("\n")})
+
+        def _stderr_reader() -> None:
+            if proc.stderr is None:
+                return
+            for line in proc.stderr:
+                stderr_tail.append(line.rstrip("\n"))
+
+        out_thread = threading.Thread(target=_stdout_reader, daemon=True)
+        err_thread = threading.Thread(target=_stderr_reader, daemon=True)
+        out_thread.start()
+        err_thread.start()
+
+        next_id = 0
+
+        def _request(method: str, params: dict[str, Any], *, text_parts: list[str] | None = None, reasoning_parts: list[str] | None = None) -> Any:
+            nonlocal next_id
+            next_id += 1
+            request_id = next_id
+            payload = {
+                "jsonrpc": "2.0",
+                "id": request_id,
+                "method": method,
+                "params": params,
+            }
+            proc.stdin.write(json.dumps(payload) + "\n")
+            proc.stdin.flush()
+
+            deadline = time.time() + timeout_seconds
+            while time.time() < deadline:
+                if proc.poll() is not None:
+                    break
+                try:
+                    msg = inbox.get(timeout=0.1)
+                except queue.Empty:
+                    continue
+
+                if self._handle_server_message(
+                    msg,
+                    process=proc,
+                    cwd=self._acp_cwd,
+                    text_parts=text_parts,
+                    reasoning_parts=reasoning_parts,
+                ):
+                    continue
+
+                if msg.get("id") != request_id:
+                    continue
+                if "error" in msg:
+                    err = msg.get("error") or {}
+                    raise RuntimeError(
+                        f"Copilot ACP {method} failed: {err.get('message') or err}"
+                    )
+                return msg.get("result")
+
+            stderr_text = "\n".join(stderr_tail).strip()
+            if proc.poll() is not None and stderr_text:
+                raise RuntimeError(f"Copilot ACP process exited early: {stderr_text}")
+            raise TimeoutError(f"Timed out waiting for Copilot ACP response to {method}.")
+
+        try:
+            _request(
+                "initialize",
+                {
+                    "protocolVersion": 1,
+                    "clientCapabilities": {
+                        "fs": {
+                            "readTextFile": True,
+                            "writeTextFile": True,
+                        }
+                    },
+                    "clientInfo": {
+                        "name": "hermes-agent",
+                        "title": "Hermes Agent",
+                        "version": "0.0.0",
+                    },
+                },
+            )
+            session = _request(
+                "session/new",
+                {
+                    "cwd": self._acp_cwd,
+                    "mcpServers": [],
+                },
+            ) or {}
+            session_id = str(session.get("sessionId") or "").strip()
+            if not session_id:
+                raise RuntimeError("Copilot ACP did not return a sessionId.")
+
+            text_parts: list[str] = []
+            reasoning_parts: list[str] = []
+            _request(
+                "session/prompt",
+                {
+                    "sessionId": session_id,
+                    "prompt": [
+                        {
+                            "type": "text",
+                            "text": prompt_text,
+                        }
+                    ],
+                },
+                text_parts=text_parts,
+                reasoning_parts=reasoning_parts,
+            )
+            return "".join(text_parts), "".join(reasoning_parts)
+        finally:
+            self.close()
+
+    def _handle_server_message(
+        self,
+        msg: dict[str, Any],
+        *,
+        process: subprocess.Popen[str],
+        cwd: str,
+        text_parts: list[str] | None,
+        reasoning_parts: list[str] | None,
+    ) -> bool:
+        method = msg.get("method")
+        if not isinstance(method, str):
+            return False
+
+        if method == "session/update":
+            params = msg.get("params") or {}
+            update = params.get("update") or {}
+            kind = str(update.get("sessionUpdate") or "").strip()
+            content = update.get("content") or {}
+            chunk_text = ""
+            if isinstance(content, dict):
+                chunk_text = str(content.get("text") or "")
+            if kind == "agent_message_chunk" and chunk_text and text_parts is not None:
+                text_parts.append(chunk_text)
+            elif kind == "agent_thought_chunk" and chunk_text and reasoning_parts is not None:
+                reasoning_parts.append(chunk_text)
+            return True
+
+        if process.stdin is None:
+            return True
+
+        message_id = msg.get("id")
+        params = msg.get("params") or {}
+
+        if method == "session/request_permission":
+            response = {
+                "jsonrpc": "2.0",
+                "id": message_id,
+                "result": {
+                    "outcome": {
+                        "outcome": "allow_once",
+                    }
+                },
+            }
+        elif method == "fs/read_text_file":
+            try:
+                path = _ensure_path_within_cwd(str(params.get("path") or ""), cwd)
+                content = path.read_text() if path.exists() else ""
+                line = params.get("line")
+                limit = params.get("limit")
+                if isinstance(line, int) and line > 1:
+                    lines = content.splitlines(keepends=True)
+                    start = line - 1
+                    end = start + limit if isinstance(limit, int) and limit > 0 else None
+                    content = "".join(lines[start:end])
+                response = {
+                    "jsonrpc": "2.0",
+                    "id": message_id,
+                    "result": {
+                        "content": content,
+                    },
+                }
+            except Exception as exc:
+                response = _jsonrpc_error(message_id, -32602, str(exc))
+        elif method == "fs/write_text_file":
+            try:
+                path = _ensure_path_within_cwd(str(params.get("path") or ""), cwd)
+                path.parent.mkdir(parents=True, exist_ok=True)
+                path.write_text(str(params.get("content") or ""))
+                response = {
+                    "jsonrpc": "2.0",
+                    "id": message_id,
+                    "result": None,
+                }
+            except Exception as exc:
+                response = _jsonrpc_error(message_id, -32602, str(exc))
+        else:
+            response = _jsonrpc_error(
+                message_id,
+                -32601,
+                f"ACP client method '{method}' is not supported by Hermes yet.",
+            )
+
+        process.stdin.write(json.dumps(response) + "\n")
+        process.stdin.flush()
+        return True

agent/credential_pool.py

@@ -0,0 +1,848 @@
+"""Persistent multi-credential pool for same-provider failover."""
+
+from __future__ import annotations
+
+import logging
+import random
+import threading
+import time
+import uuid
+import os
+from dataclasses import dataclass, fields, replace
+from typing import Any, Dict, List, Optional, Set, Tuple
+
+from hermes_constants import OPENROUTER_BASE_URL
+import hermes_cli.auth as auth_mod
+from hermes_cli.auth import (
+    ACCESS_TOKEN_REFRESH_SKEW_SECONDS,
+    CODEX_ACCESS_TOKEN_REFRESH_SKEW_SECONDS,
+    DEFAULT_AGENT_KEY_MIN_TTL_SECONDS,
+    PROVIDER_REGISTRY,
+    _agent_key_is_usable,
+    _codex_access_token_is_expiring,
+    _decode_jwt_claims,
+    _is_expiring,
+    _load_auth_store,
+    _load_provider_state,
+    read_credential_pool,
+    write_credential_pool,
+)
+
+logger = logging.getLogger(__name__)
+
+
+def _load_config_safe() -> Optional[dict]:
+    """Load config.yaml, returning None on any error."""
+    try:
+        from hermes_cli.config import load_config
+
+        return load_config()
+    except Exception:
+        return None
+
+
+# --- Status and type constants ---
+
+STATUS_OK = "ok"
+STATUS_EXHAUSTED = "exhausted"
+
+AUTH_TYPE_OAUTH = "oauth"
+AUTH_TYPE_API_KEY = "api_key"
+
+SOURCE_MANUAL = "manual"
+
+STRATEGY_FILL_FIRST = "fill_first"
+STRATEGY_ROUND_ROBIN = "round_robin"
+STRATEGY_RANDOM = "random"
+STRATEGY_LEAST_USED = "least_used"
+SUPPORTED_POOL_STRATEGIES = {
+    STRATEGY_FILL_FIRST,
+    STRATEGY_ROUND_ROBIN,
+    STRATEGY_RANDOM,
+    STRATEGY_LEAST_USED,
+}
+
+# Cooldown before retrying an exhausted credential.
+# 429 (rate-limited) cools down faster since quotas reset frequently.
+# 402 (billing/quota) and other codes use a longer default.
+EXHAUSTED_TTL_429_SECONDS = 60 * 60          # 1 hour
+EXHAUSTED_TTL_DEFAULT_SECONDS = 24 * 60 * 60 # 24 hours
+
+# Pool key prefix for custom OpenAI-compatible endpoints.
+# Custom endpoints all share provider='custom' but are keyed by their
+# custom_providers name: 'custom:<normalized_name>'.
+CUSTOM_POOL_PREFIX = "custom:"
+
+
+# Fields that are only round-tripped through JSON — never used for logic as attributes.
+_EXTRA_KEYS = frozenset({
+    "token_type", "scope", "client_id", "portal_base_url", "obtained_at",
+    "expires_in", "agent_key_id", "agent_key_expires_in", "agent_key_reused",
+    "agent_key_obtained_at", "tls",
+})
+
+
+@dataclass
+class PooledCredential:
+    provider: str
+    id: str
+    label: str
+    auth_type: str
+    priority: int
+    source: str
+    access_token: str
+    refresh_token: Optional[str] = None
+    last_status: Optional[str] = None
+    last_status_at: Optional[float] = None
+    last_error_code: Optional[int] = None
+    base_url: Optional[str] = None
+    expires_at: Optional[str] = None
+    expires_at_ms: Optional[int] = None
+    last_refresh: Optional[str] = None
+    inference_base_url: Optional[str] = None
+    agent_key: Optional[str] = None
+    agent_key_expires_at: Optional[str] = None
+    request_count: int = 0
+    extra: Dict[str, Any] = None  # type: ignore[assignment]
+
+    def __post_init__(self):
+        if self.extra is None:
+            self.extra = {}
+
+    def __getattr__(self, name: str):
+        if name in _EXTRA_KEYS:
+            return self.extra.get(name)
+        raise AttributeError(f"'{type(self).__name__}' object has no attribute {name!r}")
+
+    @classmethod
+    def from_dict(cls, provider: str, payload: Dict[str, Any]) -> "PooledCredential":
+        field_names = {f.name for f in fields(cls) if f.name != "provider"}
+        data = {k: payload.get(k) for k in field_names if k in payload}
+        extra = {k: payload[k] for k in _EXTRA_KEYS if k in payload and payload[k] is not None}
+        data["extra"] = extra
+        data.setdefault("id", uuid.uuid4().hex[:6])
+        data.setdefault("label", payload.get("source", provider))
+        data.setdefault("auth_type", AUTH_TYPE_API_KEY)
+        data.setdefault("priority", 0)
+        data.setdefault("source", SOURCE_MANUAL)
+        data.setdefault("access_token", "")
+        return cls(provider=provider, **data)
+
+    def to_dict(self) -> Dict[str, Any]:
+        _ALWAYS_EMIT = {"last_status", "last_status_at", "last_error_code"}
+        result: Dict[str, Any] = {}
+        for field_def in fields(self):
+            if field_def.name in ("provider", "extra"):
+                continue
+            value = getattr(self, field_def.name)
+            if value is not None or field_def.name in _ALWAYS_EMIT:
+                result[field_def.name] = value
+        for k, v in self.extra.items():
+            if v is not None:
+                result[k] = v
+        return result
+
+    @property
+    def runtime_api_key(self) -> str:
+        if self.provider == "nous":
+            return str(self.agent_key or self.access_token or "")
+        return str(self.access_token or "")
+
+    @property
+    def runtime_base_url(self) -> Optional[str]:
+        if self.provider == "nous":
+            return self.inference_base_url or self.base_url
+        return self.base_url
+
+
+def label_from_token(token: str, fallback: str) -> str:
+    claims = _decode_jwt_claims(token)
+    for key in ("email", "preferred_username", "upn"):
+        value = claims.get(key)
+        if isinstance(value, str) and value.strip():
+            return value.strip()
+    return fallback
+
+
+def _next_priority(entries: List[PooledCredential]) -> int:
+    return max((entry.priority for entry in entries), default=-1) + 1
+
+
+def _is_manual_source(source: str) -> bool:
+    normalized = (source or "").strip().lower()
+    return normalized == SOURCE_MANUAL or normalized.startswith(f"{SOURCE_MANUAL}:")
+
+
+def _exhausted_ttl(error_code: Optional[int]) -> int:
+    """Return cooldown seconds based on the HTTP status that caused exhaustion."""
+    if error_code == 429:
+        return EXHAUSTED_TTL_429_SECONDS
+    return EXHAUSTED_TTL_DEFAULT_SECONDS
+
+
+def _normalize_custom_pool_name(name: str) -> str:
+    """Normalize a custom provider name for use as a pool key suffix."""
+    return name.strip().lower().replace(" ", "-")
+
+
+def _iter_custom_providers(config: Optional[dict] = None):
+    """Yield (normalized_name, entry_dict) for each valid custom_providers entry."""
+    if config is None:
+        config = _load_config_safe()
+    if config is None:
+        return
+    custom_providers = config.get("custom_providers")
+    if not isinstance(custom_providers, list):
+        return
+    for entry in custom_providers:
+        if not isinstance(entry, dict):
+            continue
+        name = entry.get("name")
+        if not isinstance(name, str):
+            continue
+        yield _normalize_custom_pool_name(name), entry
+
+
+def get_custom_provider_pool_key(base_url: str) -> Optional[str]:
+    """Look up the custom_providers list in config.yaml and return 'custom:<name>' for a matching base_url.
+
+    Returns None if no match is found.
+    """
+    if not base_url:
+        return None
+    normalized_url = base_url.strip().rstrip("/")
+    for norm_name, entry in _iter_custom_providers():
+        entry_url = str(entry.get("base_url") or "").strip().rstrip("/")
+        if entry_url and entry_url == normalized_url:
+            return f"{CUSTOM_POOL_PREFIX}{norm_name}"
+    return None
+
+
+def list_custom_pool_providers() -> List[str]:
+    """Return all 'custom:*' pool keys that have entries in auth.json."""
+    pool_data = read_credential_pool(None)
+    return sorted(
+        key for key in pool_data
+        if key.startswith(CUSTOM_POOL_PREFIX)
+        and isinstance(pool_data.get(key), list)
+        and pool_data[key]
+    )
+
+
+def _get_custom_provider_config(pool_key: str) -> Optional[Dict[str, Any]]:
+    """Return the custom_providers config entry matching a pool key like 'custom:together.ai'."""
+    if not pool_key.startswith(CUSTOM_POOL_PREFIX):
+        return None
+    suffix = pool_key[len(CUSTOM_POOL_PREFIX):]
+    for norm_name, entry in _iter_custom_providers():
+        if norm_name == suffix:
+            return entry
+    return None
+
+
+def get_pool_strategy(provider: str) -> str:
+    """Return the configured selection strategy for a provider."""
+    config = _load_config_safe()
+    if config is None:
+        return STRATEGY_FILL_FIRST
+
+    strategies = config.get("credential_pool_strategies")
+    if not isinstance(strategies, dict):
+        return STRATEGY_FILL_FIRST
+
+    strategy = str(strategies.get(provider, "") or "").strip().lower()
+    if strategy in SUPPORTED_POOL_STRATEGIES:
+        return strategy
+    return STRATEGY_FILL_FIRST
+
+
+class CredentialPool:
+    def __init__(self, provider: str, entries: List[PooledCredential]):
+        self.provider = provider
+        self._entries = sorted(entries, key=lambda entry: entry.priority)
+        self._current_id: Optional[str] = None
+        self._strategy = get_pool_strategy(provider)
+        self._lock = threading.Lock()
+
+    def has_credentials(self) -> bool:
+        return bool(self._entries)
+
+    def has_available(self) -> bool:
+        """True if at least one entry is not currently in exhaustion cooldown."""
+        return bool(self._available_entries())
+
+    def entries(self) -> List[PooledCredential]:
+        return list(self._entries)
+
+    def current(self) -> Optional[PooledCredential]:
+        if not self._current_id:
+            return None
+        return next((entry for entry in self._entries if entry.id == self._current_id), None)
+
+    def _replace_entry(self, old: PooledCredential, new: PooledCredential) -> None:
+        """Swap an entry in-place by id, preserving sort order."""
+        for idx, entry in enumerate(self._entries):
+            if entry.id == old.id:
+                self._entries[idx] = new
+                return
+
+    def _persist(self) -> None:
+        write_credential_pool(
+            self.provider,
+            [entry.to_dict() for entry in self._entries],
+        )
+
+    def _mark_exhausted(self, entry: PooledCredential, status_code: Optional[int]) -> PooledCredential:
+        updated = replace(
+            entry,
+            last_status=STATUS_EXHAUSTED,
+            last_status_at=time.time(),
+            last_error_code=status_code,
+        )
+        self._replace_entry(entry, updated)
+        self._persist()
+        return updated
+
+    def _refresh_entry(self, entry: PooledCredential, *, force: bool) -> Optional[PooledCredential]:
+        if entry.auth_type != AUTH_TYPE_OAUTH or not entry.refresh_token:
+            if force:
+                self._mark_exhausted(entry, None)
+            return None
+
+        try:
+            if self.provider == "anthropic":
+                from agent.anthropic_adapter import refresh_anthropic_oauth_pure
+
+                refreshed = refresh_anthropic_oauth_pure(
+                    entry.refresh_token,
+                    use_json=entry.source.endswith("hermes_pkce"),
+                )
+                updated = replace(
+                    entry,
+                    access_token=refreshed["access_token"],
+                    refresh_token=refreshed["refresh_token"],
+                    expires_at_ms=refreshed["expires_at_ms"],
+                )
+            elif self.provider == "openai-codex":
+                refreshed = auth_mod.refresh_codex_oauth_pure(
+                    entry.access_token,
+                    entry.refresh_token,
+                )
+                updated = replace(
+                    entry,
+                    access_token=refreshed["access_token"],
+                    refresh_token=refreshed["refresh_token"],
+                    last_refresh=refreshed.get("last_refresh"),
+                )
+            elif self.provider == "nous":
+                nous_state = {
+                    "access_token": entry.access_token,
+                    "refresh_token": entry.refresh_token,
+                    "client_id": entry.client_id,
+                    "portal_base_url": entry.portal_base_url,
+                    "inference_base_url": entry.inference_base_url,
+                    "token_type": entry.token_type,
+                    "scope": entry.scope,
+                    "obtained_at": entry.obtained_at,
+                    "expires_at": entry.expires_at,
+                    "agent_key": entry.agent_key,
+                    "agent_key_expires_at": entry.agent_key_expires_at,
+                    "tls": entry.tls,
+                }
+                refreshed = auth_mod.refresh_nous_oauth_from_state(
+                    nous_state,
+                    min_key_ttl_seconds=DEFAULT_AGENT_KEY_MIN_TTL_SECONDS,
+                    force_refresh=force,
+                    force_mint=force,
+                )
+                # Apply returned fields: dataclass fields via replace, extras via dict update
+                field_updates = {}
+                extra_updates = dict(entry.extra)
+                _field_names = {f.name for f in fields(entry)}
+                for k, v in refreshed.items():
+                    if k in _field_names:
+                        field_updates[k] = v
+                    elif k in _EXTRA_KEYS:
+                        extra_updates[k] = v
+                updated = replace(entry, extra=extra_updates, **field_updates)
+            else:
+                return entry
+        except Exception as exc:
+            logger.debug("Credential refresh failed for %s/%s: %s", self.provider, entry.id, exc)
+            self._mark_exhausted(entry, None)
+            return None
+
+        updated = replace(updated, last_status=STATUS_OK, last_status_at=None, last_error_code=None)
+        self._replace_entry(entry, updated)
+        self._persist()
+        return updated
+
+    def _entry_needs_refresh(self, entry: PooledCredential) -> bool:
+        if entry.auth_type != AUTH_TYPE_OAUTH:
+            return False
+        if self.provider == "anthropic":
+            if entry.expires_at_ms is None:
+                return False
+            return int(entry.expires_at_ms) <= int(time.time() * 1000) + 120_000
+        if self.provider == "openai-codex":
+            return _codex_access_token_is_expiring(
+                entry.access_token,
+                CODEX_ACCESS_TOKEN_REFRESH_SKEW_SECONDS,
+            )
+        if self.provider == "nous":
+            # Nous refresh/mint can require network access and should happen when
+            # runtime credentials are actually resolved, not merely when the pool
+            # is enumerated for listing, migration, or selection.
+            return False
+        return False
+
+    def mark_used(self, entry_id: Optional[str] = None) -> None:
+        """Increment request_count for tracking. Used by least_used strategy."""
+        target_id = entry_id or self._current_id
+        if not target_id:
+            return
+        with self._lock:
+            for idx, entry in enumerate(self._entries):
+                if entry.id == target_id:
+                    self._entries[idx] = replace(entry, request_count=entry.request_count + 1)
+                    return
+
+    def select(self) -> Optional[PooledCredential]:
+        with self._lock:
+            return self._select_unlocked()
+
+    def _available_entries(self, *, clear_expired: bool = False, refresh: bool = False) -> List[PooledCredential]:
+        """Return entries not currently in exhaustion cooldown.
+
+        When *clear_expired* is True, entries whose cooldown has elapsed are
+        reset to STATUS_OK and persisted.  When *refresh* is True, entries
+        that need a token refresh are refreshed (skipped on failure).
+        """
+        now = time.time()
+        cleared_any = False
+        available: List[PooledCredential] = []
+        for entry in self._entries:
+            if entry.last_status == STATUS_EXHAUSTED:
+                ttl = _exhausted_ttl(entry.last_error_code)
+                if entry.last_status_at and now - entry.last_status_at < ttl:
+                    continue
+                if clear_expired:
+                    cleared = replace(entry, last_status=STATUS_OK, last_status_at=None, last_error_code=None)
+                    self._replace_entry(entry, cleared)
+                    entry = cleared
+                    cleared_any = True
+            if refresh and self._entry_needs_refresh(entry):
+                refreshed = self._refresh_entry(entry, force=False)
+                if refreshed is None:
+                    continue
+                entry = refreshed
+            available.append(entry)
+        if cleared_any:
+            self._persist()
+        return available
+
+    def _select_unlocked(self) -> Optional[PooledCredential]:
+        available = self._available_entries(clear_expired=True, refresh=True)
+        if not available:
+            self._current_id = None
+            return None
+
+        if self._strategy == STRATEGY_RANDOM:
+            entry = random.choice(available)
+            self._current_id = entry.id
+            return entry
+
+        if self._strategy == STRATEGY_LEAST_USED and len(available) > 1:
+            entry = min(available, key=lambda e: e.request_count)
+            self._current_id = entry.id
+            return entry
+
+        if self._strategy == STRATEGY_ROUND_ROBIN and len(available) > 1:
+            entry = available[0]
+            rotated = [candidate for candidate in self._entries if candidate.id != entry.id]
+            rotated.append(replace(entry, priority=len(self._entries) - 1))
+            self._entries = [replace(candidate, priority=idx) for idx, candidate in enumerate(rotated)]
+            self._persist()
+            self._current_id = entry.id
+            return self.current() or entry
+
+        entry = available[0]
+        self._current_id = entry.id
+        return entry
+
+    def peek(self) -> Optional[PooledCredential]:
+        current = self.current()
+        if current is not None:
+            return current
+        available = self._available_entries()
+        return available[0] if available else None
+
+    def mark_exhausted_and_rotate(self, *, status_code: Optional[int]) -> Optional[PooledCredential]:
+        with self._lock:
+            entry = self.current() or self._select_unlocked()
+            if entry is None:
+                return None
+            self._mark_exhausted(entry, status_code)
+            self._current_id = None
+            return self._select_unlocked()
+
+    def try_refresh_current(self) -> Optional[PooledCredential]:
+        with self._lock:
+            return self._try_refresh_current_unlocked()
+
+    def _try_refresh_current_unlocked(self) -> Optional[PooledCredential]:
+        entry = self.current()
+        if entry is None:
+            return None
+        refreshed = self._refresh_entry(entry, force=True)
+        if refreshed is not None:
+            self._current_id = refreshed.id
+        return refreshed
+
+    def reset_statuses(self) -> int:
+        count = 0
+        new_entries = []
+        for entry in self._entries:
+            if entry.last_status or entry.last_status_at or entry.last_error_code:
+                new_entries.append(replace(entry, last_status=None, last_status_at=None, last_error_code=None))
+                count += 1
+            else:
+                new_entries.append(entry)
+        if count:
+            self._entries = new_entries
+            self._persist()
+        return count
+
+    def remove_index(self, index: int) -> Optional[PooledCredential]:
+        if index < 1 or index > len(self._entries):
+            return None
+        removed = self._entries.pop(index - 1)
+        self._entries = [
+            replace(entry, priority=new_priority)
+            for new_priority, entry in enumerate(self._entries)
+        ]
+        self._persist()
+        if self._current_id == removed.id:
+            self._current_id = None
+        return removed
+
+    def add_entry(self, entry: PooledCredential) -> PooledCredential:
+        entry = replace(entry, priority=_next_priority(self._entries))
+        self._entries.append(entry)
+        self._persist()
+        return entry
+
+
+def _upsert_entry(entries: List[PooledCredential], provider: str, source: str, payload: Dict[str, Any]) -> bool:
+    existing_idx = None
+    for idx, entry in enumerate(entries):
+        if entry.source == source:
+            existing_idx = idx
+            break
+
+    if existing_idx is None:
+        payload.setdefault("id", uuid.uuid4().hex[:6])
+        payload.setdefault("priority", _next_priority(entries))
+        payload.setdefault("label", payload.get("label") or source)
+        entries.append(PooledCredential.from_dict(provider, payload))
+        return True
+
+    existing = entries[existing_idx]
+    field_updates = {}
+    extra_updates = {}
+    _field_names = {f.name for f in fields(existing)}
+    for key, value in payload.items():
+        if key in {"id", "priority"} or value is None:
+            continue
+        if key == "label" and existing.label:
+            continue
+        if key in _field_names:
+            if getattr(existing, key) != value:
+                field_updates[key] = value
+        elif key in _EXTRA_KEYS:
+            if existing.extra.get(key) != value:
+                extra_updates[key] = value
+    if field_updates or extra_updates:
+        if extra_updates:
+            field_updates["extra"] = {**existing.extra, **extra_updates}
+        entries[existing_idx] = replace(existing, **field_updates)
+        return True
+    return False
+
+
+def _normalize_pool_priorities(provider: str, entries: List[PooledCredential]) -> bool:
+    if provider != "anthropic":
+        return False
+
+    source_rank = {
+        "env:ANTHROPIC_TOKEN": 0,
+        "env:CLAUDE_CODE_OAUTH_TOKEN": 1,
+        "hermes_pkce": 2,
+        "claude_code": 3,
+        "env:ANTHROPIC_API_KEY": 4,
+    }
+    manual_entries = sorted(
+        (entry for entry in entries if _is_manual_source(entry.source)),
+        key=lambda entry: entry.priority,
+    )
+    seeded_entries = sorted(
+        (entry for entry in entries if not _is_manual_source(entry.source)),
+        key=lambda entry: (
+            source_rank.get(entry.source, len(source_rank)),
+            entry.priority,
+            entry.label,
+        ),
+    )
+
+    ordered = [*manual_entries, *seeded_entries]
+    id_to_idx = {entry.id: idx for idx, entry in enumerate(entries)}
+    changed = False
+    for new_priority, entry in enumerate(ordered):
+        if entry.priority != new_priority:
+            entries[id_to_idx[entry.id]] = replace(entry, priority=new_priority)
+            changed = True
+    return changed
+
+
+def _seed_from_singletons(provider: str, entries: List[PooledCredential]) -> Tuple[bool, Set[str]]:
+    changed = False
+    active_sources: Set[str] = set()
+    auth_store = _load_auth_store()
+
+    if provider == "anthropic":
+        from agent.anthropic_adapter import read_claude_code_credentials, read_hermes_oauth_credentials
+
+        for source_name, creds in (
+            ("hermes_pkce", read_hermes_oauth_credentials()),
+            ("claude_code", read_claude_code_credentials()),
+        ):
+            if creds and creds.get("accessToken"):
+                active_sources.add(source_name)
+                changed |= _upsert_entry(
+                    entries,
+                    provider,
+                    source_name,
+                    {
+                        "source": source_name,
+                        "auth_type": AUTH_TYPE_OAUTH,
+                        "access_token": creds.get("accessToken", ""),
+                        "refresh_token": creds.get("refreshToken"),
+                        "expires_at_ms": creds.get("expiresAt"),
+                        "label": label_from_token(creds.get("accessToken", ""), source_name),
+                    },
+                )
+
+    elif provider == "nous":
+        state = _load_provider_state(auth_store, "nous")
+        if state:
+            active_sources.add("device_code")
+            changed |= _upsert_entry(
+                entries,
+                provider,
+                "device_code",
+                {
+                    "source": "device_code",
+                    "auth_type": AUTH_TYPE_OAUTH,
+                    "access_token": state.get("access_token", ""),
+                    "refresh_token": state.get("refresh_token"),
+                    "expires_at": state.get("expires_at"),
+                    "token_type": state.get("token_type"),
+                    "scope": state.get("scope"),
+                    "client_id": state.get("client_id"),
+                    "portal_base_url": state.get("portal_base_url"),
+                    "inference_base_url": state.get("inference_base_url"),
+                    "agent_key": state.get("agent_key"),
+                    "agent_key_expires_at": state.get("agent_key_expires_at"),
+                    "tls": state.get("tls") if isinstance(state.get("tls"), dict) else None,
+                    "label": label_from_token(state.get("access_token", ""), "device_code"),
+                },
+            )
+
+    elif provider == "openai-codex":
+        state = _load_provider_state(auth_store, "openai-codex")
+        tokens = state.get("tokens") if isinstance(state, dict) else None
+        if isinstance(tokens, dict) and tokens.get("access_token"):
+            active_sources.add("device_code")
+            changed |= _upsert_entry(
+                entries,
+                provider,
+                "device_code",
+                {
+                    "source": "device_code",
+                    "auth_type": AUTH_TYPE_OAUTH,
+                    "access_token": tokens.get("access_token", ""),
+                    "refresh_token": tokens.get("refresh_token"),
+                    "base_url": "https://chatgpt.com/backend-api/codex",
+                    "last_refresh": state.get("last_refresh"),
+                    "label": label_from_token(tokens.get("access_token", ""), "device_code"),
+                },
+            )
+
+    return changed, active_sources
+
+
+def _seed_from_env(provider: str, entries: List[PooledCredential]) -> Tuple[bool, Set[str]]:
+    changed = False
+    active_sources: Set[str] = set()
+    if provider == "openrouter":
+        token = os.getenv("OPENROUTER_API_KEY", "").strip()
+        if token:
+            source = "env:OPENROUTER_API_KEY"
+            active_sources.add(source)
+            changed |= _upsert_entry(
+                entries,
+                provider,
+                source,
+                {
+                    "source": source,
+                    "auth_type": AUTH_TYPE_API_KEY,
+                    "access_token": token,
+                    "base_url": OPENROUTER_BASE_URL,
+                    "label": "OPENROUTER_API_KEY",
+                },
+            )
+        return changed, active_sources
+
+    pconfig = PROVIDER_REGISTRY.get(provider)
+    if not pconfig or pconfig.auth_type != AUTH_TYPE_API_KEY:
+        return changed, active_sources
+
+    env_url = ""
+    if pconfig.base_url_env_var:
+        env_url = os.getenv(pconfig.base_url_env_var, "").strip().rstrip("/")
+
+    env_vars = list(pconfig.api_key_env_vars)
+    if provider == "anthropic":
+        env_vars = [
+            "ANTHROPIC_TOKEN",
+            "CLAUDE_CODE_OAUTH_TOKEN",
+            "ANTHROPIC_API_KEY",
+        ]
+
+    for env_var in env_vars:
+        token = os.getenv(env_var, "").strip()
+        if not token:
+            continue
+        source = f"env:{env_var}"
+        active_sources.add(source)
+        auth_type = AUTH_TYPE_OAUTH if provider == "anthropic" and not token.startswith("sk-ant-api") else AUTH_TYPE_API_KEY
+        base_url = env_url or pconfig.inference_base_url
+        changed |= _upsert_entry(
+            entries,
+            provider,
+            source,
+            {
+                "source": source,
+                "auth_type": auth_type,
+                "access_token": token,
+                "base_url": base_url,
+                "label": env_var,
+            },
+        )
+    return changed, active_sources
+
+
+def _prune_stale_seeded_entries(entries: List[PooledCredential], active_sources: Set[str]) -> bool:
+    retained = [
+        entry
+        for entry in entries
+        if _is_manual_source(entry.source)
+        or entry.source in active_sources
+        or not (
+            entry.source.startswith("env:")
+            or entry.source in {"claude_code", "hermes_pkce"}
+        )
+    ]
+    if len(retained) == len(entries):
+        return False
+    entries[:] = retained
+    return True
+
+
+def _seed_custom_pool(pool_key: str, entries: List[PooledCredential]) -> Tuple[bool, Set[str]]:
+    """Seed a custom endpoint pool from custom_providers config and model config."""
+    changed = False
+    active_sources: Set[str] = set()
+
+    # Seed from the custom_providers config entry's api_key field
+    cp_config = _get_custom_provider_config(pool_key)
+    if cp_config:
+        api_key = str(cp_config.get("api_key") or "").strip()
+        base_url = str(cp_config.get("base_url") or "").strip().rstrip("/")
+        name = str(cp_config.get("name") or "").strip()
+        if api_key:
+            source = f"config:{name}"
+            active_sources.add(source)
+            changed |= _upsert_entry(
+                entries,
+                pool_key,
+                source,
+                {
+                    "source": source,
+                    "auth_type": AUTH_TYPE_API_KEY,
+                    "access_token": api_key,
+                    "base_url": base_url,
+                    "label": name or source,
+                },
+            )
+
+    # Seed from model.api_key if model.provider=='custom' and model.base_url matches
+    try:
+        config = _load_config_safe()
+        model_cfg = config.get("model") if config else None
+        if isinstance(model_cfg, dict):
+            model_provider = str(model_cfg.get("provider") or "").strip().lower()
+            model_base_url = str(model_cfg.get("base_url") or "").strip().rstrip("/")
+            model_api_key = ""
+            for k in ("api_key", "api"):
+                v = model_cfg.get(k)
+                if isinstance(v, str) and v.strip():
+                    model_api_key = v.strip()
+                    break
+            if model_provider == "custom" and model_base_url and model_api_key:
+                # Check if this model's base_url matches our custom provider
+                matched_key = get_custom_provider_pool_key(model_base_url)
+                if matched_key == pool_key:
+                    source = "model_config"
+                    active_sources.add(source)
+                    changed |= _upsert_entry(
+                        entries,
+                        pool_key,
+                        source,
+                        {
+                            "source": source,
+                            "auth_type": AUTH_TYPE_API_KEY,
+                            "access_token": model_api_key,
+                            "base_url": model_base_url,
+                            "label": "model_config",
+                        },
+                    )
+    except Exception:
+        pass
+
+    return changed, active_sources
+
+
+def load_pool(provider: str) -> CredentialPool:
+    provider = (provider or "").strip().lower()
+    raw_entries = read_credential_pool(provider)
+    entries = [PooledCredential.from_dict(provider, payload) for payload in raw_entries]
+
+    if provider.startswith(CUSTOM_POOL_PREFIX):
+        # Custom endpoint pool — seed from custom_providers config and model config
+        custom_changed, custom_sources = _seed_custom_pool(provider, entries)
+        changed = custom_changed
+        changed |= _prune_stale_seeded_entries(entries, custom_sources)
+    else:
+        singleton_changed, singleton_sources = _seed_from_singletons(provider, entries)
+        env_changed, env_sources = _seed_from_env(provider, entries)
+        changed = singleton_changed or env_changed
+        changed |= _prune_stale_seeded_entries(entries, singleton_sources | env_sources)
+        changed |= _normalize_pool_priorities(provider, entries)
+
+    if changed:
+        write_credential_pool(
+            provider,
+            [entry.to_dict() for entry in sorted(entries, key=lambda item: item.priority)],
+        )
+    return CredentialPool(provider, entries)

agent/insights.py

@@ -0,0 +1,799 @@
+"""
+Session Insights Engine for Hermes Agent.
+
+Analyzes historical session data from the SQLite state database to produce
+comprehensive usage insights — token consumption, cost estimates, tool usage
+patterns, activity trends, model/platform breakdowns, and session metrics.
+
+Inspired by Claude Code's /insights command, adapted for Hermes Agent's
+multi-platform architecture with additional cost estimation and platform
+breakdown capabilities.
+
+Usage:
+    from agent.insights import InsightsEngine
+    engine = InsightsEngine(db)
+    report = engine.generate(days=30)
+    print(engine.format_terminal(report))
+"""
+
+import json
+import time
+from collections import Counter, defaultdict
+from datetime import datetime
+from typing import Any, Dict, List
+
+from agent.usage_pricing import (
+    CanonicalUsage,
+    DEFAULT_PRICING,
+    estimate_usage_cost,
+    format_duration_compact,
+    get_pricing,
+    has_known_pricing,
+)
+
+_DEFAULT_PRICING = DEFAULT_PRICING
+
+
+def _has_known_pricing(model_name: str, provider: str = None, base_url: str = None) -> bool:
+    """Check if a model has known pricing (vs unknown/custom endpoint)."""
+    return has_known_pricing(model_name, provider=provider, base_url=base_url)
+
+
+def _get_pricing(model_name: str) -> Dict[str, float]:
+    """Look up pricing for a model. Uses fuzzy matching on model name.
+
+    Returns _DEFAULT_PRICING (zero cost) for unknown/custom models —
+    we can't assume costs for self-hosted endpoints, local inference, etc.
+    """
+    return get_pricing(model_name)
+
+
+def _estimate_cost(
+    session_or_model: Dict[str, Any] | str,
+    input_tokens: int = 0,
+    output_tokens: int = 0,
+    *,
+    cache_read_tokens: int = 0,
+    cache_write_tokens: int = 0,
+    provider: str = None,
+    base_url: str = None,
+) -> tuple[float, str]:
+    """Estimate the USD cost for a session row or a model/token tuple."""
+    if isinstance(session_or_model, dict):
+        session = session_or_model
+        model = session.get("model") or ""
+        usage = CanonicalUsage(
+            input_tokens=session.get("input_tokens") or 0,
+            output_tokens=session.get("output_tokens") or 0,
+            cache_read_tokens=session.get("cache_read_tokens") or 0,
+            cache_write_tokens=session.get("cache_write_tokens") or 0,
+        )
+        provider = session.get("billing_provider")
+        base_url = session.get("billing_base_url")
+    else:
+        model = session_or_model or ""
+        usage = CanonicalUsage(
+            input_tokens=input_tokens,
+            output_tokens=output_tokens,
+            cache_read_tokens=cache_read_tokens,
+            cache_write_tokens=cache_write_tokens,
+        )
+    result = estimate_usage_cost(
+        model,
+        usage,
+        provider=provider,
+        base_url=base_url,
+    )
+    return float(result.amount_usd or 0.0), result.status
+
+
+def _format_duration(seconds: float) -> str:
+    """Format seconds into a human-readable duration string."""
+    return format_duration_compact(seconds)
+
+
+def _bar_chart(values: List[int], max_width: int = 20) -> List[str]:
+    """Create simple horizontal bar chart strings from values."""
+    peak = max(values) if values else 1
+    if peak == 0:
+        return ["" for _ in values]
+    return ["█" * max(1, int(v / peak * max_width)) if v > 0 else "" for v in values]
+
+
+class InsightsEngine:
+    """
+    Analyzes session history and produces usage insights.
+
+    Works directly with a SessionDB instance (or raw sqlite3 connection)
+    to query session and message data.
+    """
+
+    def __init__(self, db):
+        """
+        Initialize with a SessionDB instance.
+
+        Args:
+            db: A SessionDB instance (from hermes_state.py)
+        """
+        self.db = db
+        self._conn = db._conn
+
+    def generate(self, days: int = 30, source: str = None) -> Dict[str, Any]:
+        """
+        Generate a complete insights report.
+
+        Args:
+            days: Number of days to look back (default: 30)
+            source: Optional filter by source platform
+
+        Returns:
+            Dict with all computed insights
+        """
+        cutoff = time.time() - (days * 86400)
+
+        # Gather raw data
+        sessions = self._get_sessions(cutoff, source)
+        tool_usage = self._get_tool_usage(cutoff, source)
+        message_stats = self._get_message_stats(cutoff, source)
+
+        if not sessions:
+            return {
+                "days": days,
+                "source_filter": source,
+                "empty": True,
+                "overview": {},
+                "models": [],
+                "platforms": [],
+                "tools": [],
+                "activity": {},
+                "top_sessions": [],
+            }
+
+        # Compute insights
+        overview = self._compute_overview(sessions, message_stats)
+        models = self._compute_model_breakdown(sessions)
+        platforms = self._compute_platform_breakdown(sessions)
+        tools = self._compute_tool_breakdown(tool_usage)
+        activity = self._compute_activity_patterns(sessions)
+        top_sessions = self._compute_top_sessions(sessions)
+
+        return {
+            "days": days,
+            "source_filter": source,
+            "empty": False,
+            "generated_at": time.time(),
+            "overview": overview,
+            "models": models,
+            "platforms": platforms,
+            "tools": tools,
+            "activity": activity,
+            "top_sessions": top_sessions,
+        }
+
+    # =========================================================================
+    # Data gathering (SQL queries)
+    # =========================================================================
+
+    # Columns we actually need (skip system_prompt, model_config blobs)
+    _SESSION_COLS = ("id, source, model, started_at, ended_at, "
+                     "message_count, tool_call_count, input_tokens, output_tokens, "
+                     "cache_read_tokens, cache_write_tokens, billing_provider, "
+                     "billing_base_url, billing_mode, estimated_cost_usd, "
+                     "actual_cost_usd, cost_status, cost_source")
+
+    # Pre-computed query strings — f-string evaluated once at class definition,
+    # not at runtime, so no user-controlled value can alter the query structure.
+    _GET_SESSIONS_WITH_SOURCE = (
+        f"SELECT {_SESSION_COLS} FROM sessions"
+        " WHERE started_at >= ? AND source = ?"
+        " ORDER BY started_at DESC"
+    )
+    _GET_SESSIONS_ALL = (
+        f"SELECT {_SESSION_COLS} FROM sessions"
+        " WHERE started_at >= ?"
+        " ORDER BY started_at DESC"
+    )
+
+    def _get_sessions(self, cutoff: float, source: str = None) -> List[Dict]:
+        """Fetch sessions within the time window."""
+        if source:
+            cursor = self._conn.execute(self._GET_SESSIONS_WITH_SOURCE, (cutoff, source))
+        else:
+            cursor = self._conn.execute(self._GET_SESSIONS_ALL, (cutoff,))
+        return [dict(row) for row in cursor.fetchall()]
+
+    def _get_tool_usage(self, cutoff: float, source: str = None) -> List[Dict]:
+        """Get tool call counts from messages.
+
+        Uses two sources:
+        1. tool_name column on 'tool' role messages (set by gateway)
+        2. tool_calls JSON on 'assistant' role messages (covers CLI where
+           tool_name is not populated on tool responses)
+        """
+        tool_counts = Counter()
+
+        # Source 1: explicit tool_name on tool response messages
+        if source:
+            cursor = self._conn.execute(
+                """SELECT m.tool_name, COUNT(*) as count
+                   FROM messages m
+                   JOIN sessions s ON s.id = m.session_id
+                   WHERE s.started_at >= ? AND s.source = ?
+                     AND m.role = 'tool' AND m.tool_name IS NOT NULL
+                   GROUP BY m.tool_name
+                   ORDER BY count DESC""",
+                (cutoff, source),
+            )
+        else:
+            cursor = self._conn.execute(
+                """SELECT m.tool_name, COUNT(*) as count
+                   FROM messages m
+                   JOIN sessions s ON s.id = m.session_id
+                   WHERE s.started_at >= ?
+                     AND m.role = 'tool' AND m.tool_name IS NOT NULL
+                   GROUP BY m.tool_name
+                   ORDER BY count DESC""",
+                (cutoff,),
+            )
+        for row in cursor.fetchall():
+            tool_counts[row["tool_name"]] += row["count"]
+
+        # Source 2: extract from tool_calls JSON on assistant messages
+        # (covers CLI sessions where tool_name is NULL on tool responses)
+        if source:
+            cursor2 = self._conn.execute(
+                """SELECT m.tool_calls
+                   FROM messages m
+                   JOIN sessions s ON s.id = m.session_id
+                   WHERE s.started_at >= ? AND s.source = ?
+                     AND m.role = 'assistant' AND m.tool_calls IS NOT NULL""",
+                (cutoff, source),
+            )
+        else:
+            cursor2 = self._conn.execute(
+                """SELECT m.tool_calls
+                   FROM messages m
+                   JOIN sessions s ON s.id = m.session_id
+                   WHERE s.started_at >= ?
+                     AND m.role = 'assistant' AND m.tool_calls IS NOT NULL""",
+                (cutoff,),
+            )
+
+        tool_calls_counts = Counter()
+        for row in cursor2.fetchall():
+            try:
+                calls = row["tool_calls"]
+                if isinstance(calls, str):
+                    calls = json.loads(calls)
+                if isinstance(calls, list):
+                    for call in calls:
+                        func = call.get("function", {}) if isinstance(call, dict) else {}
+                        name = func.get("name")
+                        if name:
+                            tool_calls_counts[name] += 1
+            except (json.JSONDecodeError, TypeError, AttributeError):
+                continue
+
+        # Merge: prefer tool_name source, supplement with tool_calls source
+        # for tools not already counted
+        if not tool_counts and tool_calls_counts:
+            # No tool_name data at all — use tool_calls exclusively
+            tool_counts = tool_calls_counts
+        elif tool_counts and tool_calls_counts:
+            # Both sources have data — use whichever has the higher count per tool
+            # (they may overlap, so take the max to avoid double-counting)
+            all_tools = set(tool_counts) | set(tool_calls_counts)
+            merged = Counter()
+            for tool in all_tools:
+                merged[tool] = max(tool_counts.get(tool, 0), tool_calls_counts.get(tool, 0))
+            tool_counts = merged
+
+        # Convert to the expected format
+        return [
+            {"tool_name": name, "count": count}
+            for name, count in tool_counts.most_common()
+        ]
+
+    def _get_message_stats(self, cutoff: float, source: str = None) -> Dict:
+        """Get aggregate message statistics."""
+        if source:
+            cursor = self._conn.execute(
+                """SELECT
+                     COUNT(*) as total_messages,
+                     SUM(CASE WHEN m.role = 'user' THEN 1 ELSE 0 END) as user_messages,
+                     SUM(CASE WHEN m.role = 'assistant' THEN 1 ELSE 0 END) as assistant_messages,
+                     SUM(CASE WHEN m.role = 'tool' THEN 1 ELSE 0 END) as tool_messages
+                   FROM messages m
+                   JOIN sessions s ON s.id = m.session_id
+                   WHERE s.started_at >= ? AND s.source = ?""",
+                (cutoff, source),
+            )
+        else:
+            cursor = self._conn.execute(
+                """SELECT
+                     COUNT(*) as total_messages,
+                     SUM(CASE WHEN m.role = 'user' THEN 1 ELSE 0 END) as user_messages,
+                     SUM(CASE WHEN m.role = 'assistant' THEN 1 ELSE 0 END) as assistant_messages,
+                     SUM(CASE WHEN m.role = 'tool' THEN 1 ELSE 0 END) as tool_messages
+                   FROM messages m
+                   JOIN sessions s ON s.id = m.session_id
+                   WHERE s.started_at >= ?""",
+                (cutoff,),
+            )
+        row = cursor.fetchone()
+        return dict(row) if row else {
+            "total_messages": 0, "user_messages": 0,
+            "assistant_messages": 0, "tool_messages": 0,
+        }
+
+    # =========================================================================
+    # Computation
+    # =========================================================================
+
+    def _compute_overview(self, sessions: List[Dict], message_stats: Dict) -> Dict:
+        """Compute high-level overview statistics."""
+        total_input = sum(s.get("input_tokens") or 0 for s in sessions)
+        total_output = sum(s.get("output_tokens") or 0 for s in sessions)
+        total_cache_read = sum(s.get("cache_read_tokens") or 0 for s in sessions)
+        total_cache_write = sum(s.get("cache_write_tokens") or 0 for s in sessions)
+        total_tokens = total_input + total_output + total_cache_read + total_cache_write
+        total_tool_calls = sum(s.get("tool_call_count") or 0 for s in sessions)
+        total_messages = sum(s.get("message_count") or 0 for s in sessions)
+
+        # Cost estimation (weighted by model)
+        total_cost = 0.0
+        actual_cost = 0.0
+        models_with_pricing = set()
+        models_without_pricing = set()
+        unknown_cost_sessions = 0
+        included_cost_sessions = 0
+        for s in sessions:
+            model = s.get("model") or ""
+            estimated, status = _estimate_cost(s)
+            total_cost += estimated
+            actual_cost += s.get("actual_cost_usd") or 0.0
+            display = model.split("/")[-1] if "/" in model else (model or "unknown")
+            if status == "included":
+                included_cost_sessions += 1
+            elif status == "unknown":
+                unknown_cost_sessions += 1
+            if _has_known_pricing(model, s.get("billing_provider"), s.get("billing_base_url")):
+                models_with_pricing.add(display)
+            else:
+                models_without_pricing.add(display)
+
+        # Session duration stats (guard against negative durations from clock drift)
+        durations = []
+        for s in sessions:
+            start = s.get("started_at")
+            end = s.get("ended_at")
+            if start and end and end > start:
+                durations.append(end - start)
+
+        total_hours = sum(durations) / 3600 if durations else 0
+        avg_duration = sum(durations) / len(durations) if durations else 0
+
+        # Earliest and latest session
+        started_timestamps = [s["started_at"] for s in sessions if s.get("started_at")]
+        date_range_start = min(started_timestamps) if started_timestamps else None
+        date_range_end = max(started_timestamps) if started_timestamps else None
+
+        return {
+            "total_sessions": len(sessions),
+            "total_messages": total_messages,
+            "total_tool_calls": total_tool_calls,
+            "total_input_tokens": total_input,
+            "total_output_tokens": total_output,
+            "total_cache_read_tokens": total_cache_read,
+            "total_cache_write_tokens": total_cache_write,
+            "total_tokens": total_tokens,
+            "estimated_cost": total_cost,
+            "actual_cost": actual_cost,
+            "total_hours": total_hours,
+            "avg_session_duration": avg_duration,
+            "avg_messages_per_session": total_messages / len(sessions) if sessions else 0,
+            "avg_tokens_per_session": total_tokens / len(sessions) if sessions else 0,
+            "user_messages": message_stats.get("user_messages") or 0,
+            "assistant_messages": message_stats.get("assistant_messages") or 0,
+            "tool_messages": message_stats.get("tool_messages") or 0,
+            "date_range_start": date_range_start,
+            "date_range_end": date_range_end,
+            "models_with_pricing": sorted(models_with_pricing),
+            "models_without_pricing": sorted(models_without_pricing),
+            "unknown_cost_sessions": unknown_cost_sessions,
+            "included_cost_sessions": included_cost_sessions,
+        }
+
+    def _compute_model_breakdown(self, sessions: List[Dict]) -> List[Dict]:
+        """Break down usage by model."""
+        model_data = defaultdict(lambda: {
+            "sessions": 0, "input_tokens": 0, "output_tokens": 0,
+            "cache_read_tokens": 0, "cache_write_tokens": 0,
+            "total_tokens": 0, "tool_calls": 0, "cost": 0.0,
+        })
+
+        for s in sessions:
+            model = s.get("model") or "unknown"
+            # Normalize: strip provider prefix for display
+            display_model = model.split("/")[-1] if "/" in model else model
+            d = model_data[display_model]
+            d["sessions"] += 1
+            inp = s.get("input_tokens") or 0
+            out = s.get("output_tokens") or 0
+            cache_read = s.get("cache_read_tokens") or 0
+            cache_write = s.get("cache_write_tokens") or 0
+            d["input_tokens"] += inp
+            d["output_tokens"] += out
+            d["cache_read_tokens"] += cache_read
+            d["cache_write_tokens"] += cache_write
+            d["total_tokens"] += inp + out + cache_read + cache_write
+            d["tool_calls"] += s.get("tool_call_count") or 0
+            estimate, status = _estimate_cost(s)
+            d["cost"] += estimate
+            d["has_pricing"] = _has_known_pricing(model, s.get("billing_provider"), s.get("billing_base_url"))
+            d["cost_status"] = status
+
+        result = [
+            {"model": model, **data}
+            for model, data in model_data.items()
+        ]
+        # Sort by tokens first, fall back to session count when tokens are 0
+        result.sort(key=lambda x: (x["total_tokens"], x["sessions"]), reverse=True)
+        return result
+
+    def _compute_platform_breakdown(self, sessions: List[Dict]) -> List[Dict]:
+        """Break down usage by platform/source."""
+        platform_data = defaultdict(lambda: {
+            "sessions": 0, "messages": 0, "input_tokens": 0,
+            "output_tokens": 0, "cache_read_tokens": 0,
+            "cache_write_tokens": 0, "total_tokens": 0, "tool_calls": 0,
+        })
+
+        for s in sessions:
+            source = s.get("source") or "unknown"
+            d = platform_data[source]
+            d["sessions"] += 1
+            d["messages"] += s.get("message_count") or 0
+            inp = s.get("input_tokens") or 0
+            out = s.get("output_tokens") or 0
+            cache_read = s.get("cache_read_tokens") or 0
+            cache_write = s.get("cache_write_tokens") or 0
+            d["input_tokens"] += inp
+            d["output_tokens"] += out
+            d["cache_read_tokens"] += cache_read
+            d["cache_write_tokens"] += cache_write
+            d["total_tokens"] += inp + out + cache_read + cache_write
+            d["tool_calls"] += s.get("tool_call_count") or 0
+
+        result = [
+            {"platform": platform, **data}
+            for platform, data in platform_data.items()
+        ]
+        result.sort(key=lambda x: x["sessions"], reverse=True)
+        return result
+
+    def _compute_tool_breakdown(self, tool_usage: List[Dict]) -> List[Dict]:
+        """Process tool usage data into a ranked list with percentages."""
+        total_calls = sum(t["count"] for t in tool_usage) if tool_usage else 0
+        result = []
+        for t in tool_usage:
+            pct = (t["count"] / total_calls * 100) if total_calls else 0
+            result.append({
+                "tool": t["tool_name"],
+                "count": t["count"],
+                "percentage": pct,
+            })
+        return result
+
+    def _compute_activity_patterns(self, sessions: List[Dict]) -> Dict:
+        """Analyze activity patterns by day of week and hour."""
+        day_counts = Counter()  # 0=Monday ... 6=Sunday
+        hour_counts = Counter()
+        daily_counts = Counter()  # date string -> count
+
+        for s in sessions:
+            ts = s.get("started_at")
+            if not ts:
+                continue
+            dt = datetime.fromtimestamp(ts)
+            day_counts[dt.weekday()] += 1
+            hour_counts[dt.hour] += 1
+            daily_counts[dt.strftime("%Y-%m-%d")] += 1
+
+        day_names = ["Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun"]
+        day_breakdown = [
+            {"day": day_names[i], "count": day_counts.get(i, 0)}
+            for i in range(7)
+        ]
+
+        hour_breakdown = [
+            {"hour": i, "count": hour_counts.get(i, 0)}
+            for i in range(24)
+        ]
+
+        # Busiest day and hour
+        busiest_day = max(day_breakdown, key=lambda x: x["count"]) if day_breakdown else None
+        busiest_hour = max(hour_breakdown, key=lambda x: x["count"]) if hour_breakdown else None
+
+        # Active days (days with at least one session)
+        active_days = len(daily_counts)
+
+        # Streak calculation
+        if daily_counts:
+            all_dates = sorted(daily_counts.keys())
+            current_streak = 1
+            max_streak = 1
+            for i in range(1, len(all_dates)):
+                d1 = datetime.strptime(all_dates[i - 1], "%Y-%m-%d")
+                d2 = datetime.strptime(all_dates[i], "%Y-%m-%d")
+                if (d2 - d1).days == 1:
+                    current_streak += 1
+                    max_streak = max(max_streak, current_streak)
+                else:
+                    current_streak = 1
+        else:
+            max_streak = 0
+
+        return {
+            "by_day": day_breakdown,
+            "by_hour": hour_breakdown,
+            "busiest_day": busiest_day,
+            "busiest_hour": busiest_hour,
+            "active_days": active_days,
+            "max_streak": max_streak,
+        }
+
+    def _compute_top_sessions(self, sessions: List[Dict]) -> List[Dict]:
+        """Find notable sessions (longest, most messages, most tokens)."""
+        top = []
+
+        # Longest by duration
+        sessions_with_duration = [
+            s for s in sessions
+            if s.get("started_at") and s.get("ended_at")
+        ]
+        if sessions_with_duration:
+            longest = max(
+                sessions_with_duration,
+                key=lambda s: (s["ended_at"] - s["started_at"]),
+            )
+            dur = longest["ended_at"] - longest["started_at"]
+            top.append({
+                "label": "Longest session",
+                "session_id": longest["id"][:16],
+                "value": _format_duration(dur),
+                "date": datetime.fromtimestamp(longest["started_at"]).strftime("%b %d"),
+            })
+
+        # Most messages
+        most_msgs = max(sessions, key=lambda s: s.get("message_count") or 0)
+        if (most_msgs.get("message_count") or 0) > 0:
+            top.append({
+                "label": "Most messages",
+                "session_id": most_msgs["id"][:16],
+                "value": f"{most_msgs['message_count']} msgs",
+                "date": datetime.fromtimestamp(most_msgs["started_at"]).strftime("%b %d") if most_msgs.get("started_at") else "?",
+            })
+
+        # Most tokens
+        most_tokens = max(
+            sessions,
+            key=lambda s: (s.get("input_tokens") or 0) + (s.get("output_tokens") or 0),
+        )
+        token_total = (most_tokens.get("input_tokens") or 0) + (most_tokens.get("output_tokens") or 0)
+        if token_total > 0:
+            top.append({
+                "label": "Most tokens",
+                "session_id": most_tokens["id"][:16],
+                "value": f"{token_total:,} tokens",
+                "date": datetime.fromtimestamp(most_tokens["started_at"]).strftime("%b %d") if most_tokens.get("started_at") else "?",
+            })
+
+        # Most tool calls
+        most_tools = max(sessions, key=lambda s: s.get("tool_call_count") or 0)
+        if (most_tools.get("tool_call_count") or 0) > 0:
+            top.append({
+                "label": "Most tool calls",
+                "session_id": most_tools["id"][:16],
+                "value": f"{most_tools['tool_call_count']} calls",
+                "date": datetime.fromtimestamp(most_tools["started_at"]).strftime("%b %d") if most_tools.get("started_at") else "?",
+            })
+
+        return top
+
+    # =========================================================================
+    # Formatting
+    # =========================================================================
+
+    def format_terminal(self, report: Dict) -> str:
+        """Format the insights report for terminal display (CLI)."""
+        if report.get("empty"):
+            days = report.get("days", 30)
+            src = f" (source: {report['source_filter']})" if report.get("source_filter") else ""
+            return f"  No sessions found in the last {days} days{src}."
+
+        lines = []
+        o = report["overview"]
+        days = report["days"]
+        src_filter = report.get("source_filter")
+
+        # Header
+        lines.append("")
+        lines.append("  ╔══════════════════════════════════════════════════════════╗")
+        lines.append("  ║                    📊 Hermes Insights                    ║")
+        period_label = f"Last {days} days"
+        if src_filter:
+            period_label += f" ({src_filter})"
+        padding = 58 - len(period_label) - 2
+        left_pad = padding // 2
+        right_pad = padding - left_pad
+        lines.append(f"  ║{' ' * left_pad} {period_label} {' ' * right_pad}║")
+        lines.append("  ╚══════════════════════════════════════════════════════════╝")
+        lines.append("")
+
+        # Date range
+        if o.get("date_range_start") and o.get("date_range_end"):
+            start_str = datetime.fromtimestamp(o["date_range_start"]).strftime("%b %d, %Y")
+            end_str = datetime.fromtimestamp(o["date_range_end"]).strftime("%b %d, %Y")
+            lines.append(f"  Period: {start_str} — {end_str}")
+            lines.append("")
+
+        # Overview
+        lines.append("  📋 Overview")
+        lines.append("  " + "─" * 56)
+        lines.append(f"  Sessions:          {o['total_sessions']:<12}  Messages:        {o['total_messages']:,}")
+        lines.append(f"  Tool calls:        {o['total_tool_calls']:<12,}  User messages:   {o['user_messages']:,}")
+        lines.append(f"  Input tokens:      {o['total_input_tokens']:<12,}  Output tokens:   {o['total_output_tokens']:,}")
+        cache_total = o.get("total_cache_read_tokens", 0) + o.get("total_cache_write_tokens", 0)
+        if cache_total > 0:
+            lines.append(f"  Cache read:        {o['total_cache_read_tokens']:<12,}  Cache write:     {o['total_cache_write_tokens']:,}")
+        cost_str = f"${o['estimated_cost']:.2f}"
+        if o.get("models_without_pricing"):
+            cost_str += " *"
+        lines.append(f"  Total tokens:      {o['total_tokens']:<12,}  Est. cost:       {cost_str}")
+        if o["total_hours"] > 0:
+            lines.append(f"  Active time:       ~{_format_duration(o['total_hours'] * 3600):<11}  Avg session:     ~{_format_duration(o['avg_session_duration'])}")
+        lines.append(f"  Avg msgs/session:  {o['avg_messages_per_session']:.1f}")
+        lines.append("")
+
+        # Model breakdown
+        if report["models"]:
+            lines.append("  🤖 Models Used")
+            lines.append("  " + "─" * 56)
+            lines.append(f"  {'Model':<30} {'Sessions':>8} {'Tokens':>12} {'Cost':>8}")
+            for m in report["models"]:
+                model_name = m["model"][:28]
+                if m.get("has_pricing"):
+                    cost_cell = f"${m['cost']:>6.2f}"
+                else:
+                    cost_cell = "     N/A"
+                lines.append(f"  {model_name:<30} {m['sessions']:>8} {m['total_tokens']:>12,} {cost_cell}")
+            if o.get("models_without_pricing"):
+                lines.append("  * Cost N/A for custom/self-hosted models")
+            lines.append("")
+
+        # Platform breakdown
+        if len(report["platforms"]) > 1 or (report["platforms"] and report["platforms"][0]["platform"] != "cli"):
+            lines.append("  📱 Platforms")
+            lines.append("  " + "─" * 56)
+            lines.append(f"  {'Platform':<14} {'Sessions':>8} {'Messages':>10} {'Tokens':>14}")
+            for p in report["platforms"]:
+                lines.append(f"  {p['platform']:<14} {p['sessions']:>8} {p['messages']:>10,} {p['total_tokens']:>14,}")
+            lines.append("")
+
+        # Tool usage
+        if report["tools"]:
+            lines.append("  🔧 Top Tools")
+            lines.append("  " + "─" * 56)
+            lines.append(f"  {'Tool':<28} {'Calls':>8} {'%':>8}")
+            for t in report["tools"][:15]:  # Top 15
+                lines.append(f"  {t['tool']:<28} {t['count']:>8,} {t['percentage']:>7.1f}%")
+            if len(report["tools"]) > 15:
+                lines.append(f"  ... and {len(report['tools']) - 15} more tools")
+            lines.append("")
+
+        # Activity patterns
+        act = report.get("activity", {})
+        if act.get("by_day"):
+            lines.append("  📅 Activity Patterns")
+            lines.append("  " + "─" * 56)
+
+            # Day of week chart
+            day_values = [d["count"] for d in act["by_day"]]
+            bars = _bar_chart(day_values, max_width=15)
+            for i, d in enumerate(act["by_day"]):
+                bar = bars[i]
+                lines.append(f"  {d['day']}  {bar:<15} {d['count']}")
+
+            lines.append("")
+
+            # Peak hours (show top 5 busiest hours)
+            busy_hours = sorted(act["by_hour"], key=lambda x: x["count"], reverse=True)
+            busy_hours = [h for h in busy_hours if h["count"] > 0][:5]
+            if busy_hours:
+                hour_strs = []
+                for h in busy_hours:
+                    hr = h["hour"]
+                    ampm = "AM" if hr < 12 else "PM"
+                    display_hr = hr % 12 or 12
+                    hour_strs.append(f"{display_hr}{ampm} ({h['count']})")
+                lines.append(f"  Peak hours: {', '.join(hour_strs)}")
+
+            if act.get("active_days"):
+                lines.append(f"  Active days: {act['active_days']}")
+            if act.get("max_streak") and act["max_streak"] > 1:
+                lines.append(f"  Best streak: {act['max_streak']} consecutive days")
+            lines.append("")
+
+        # Notable sessions
+        if report.get("top_sessions"):
+            lines.append("  🏆 Notable Sessions")
+            lines.append("  " + "─" * 56)
+            for ts in report["top_sessions"]:
+                lines.append(f"  {ts['label']:<20} {ts['value']:<18} ({ts['date']}, {ts['session_id']})")
+            lines.append("")
+
+        return "\n".join(lines)
+
+    def format_gateway(self, report: Dict) -> str:
+        """Format the insights report for gateway/messaging (shorter)."""
+        if report.get("empty"):
+            days = report.get("days", 30)
+            return f"No sessions found in the last {days} days."
+
+        lines = []
+        o = report["overview"]
+        days = report["days"]
+
+        lines.append(f"📊 **Hermes Insights** — Last {days} days\n")
+
+        # Overview
+        lines.append(f"**Sessions:** {o['total_sessions']} | **Messages:** {o['total_messages']:,} | **Tool calls:** {o['total_tool_calls']:,}")
+        cache_total = o.get("total_cache_read_tokens", 0) + o.get("total_cache_write_tokens", 0)
+        if cache_total > 0:
+            lines.append(f"**Tokens:** {o['total_tokens']:,} (in: {o['total_input_tokens']:,} / out: {o['total_output_tokens']:,} / cache: {cache_total:,})")
+        else:
+            lines.append(f"**Tokens:** {o['total_tokens']:,} (in: {o['total_input_tokens']:,} / out: {o['total_output_tokens']:,})")
+        cost_note = ""
+        if o.get("models_without_pricing"):
+            cost_note = " _(excludes custom/self-hosted models)_"
+        lines.append(f"**Est. cost:** ${o['estimated_cost']:.2f}{cost_note}")
+        if o["total_hours"] > 0:
+            lines.append(f"**Active time:** ~{_format_duration(o['total_hours'] * 3600)} | **Avg session:** ~{_format_duration(o['avg_session_duration'])}")
+        lines.append("")
+
+        # Models (top 5)
+        if report["models"]:
+            lines.append("**🤖 Models:**")
+            for m in report["models"][:5]:
+                cost_str = f"${m['cost']:.2f}" if m.get("has_pricing") else "N/A"
+                lines.append(f"  {m['model'][:25]} — {m['sessions']} sessions, {m['total_tokens']:,} tokens, {cost_str}")
+            lines.append("")
+
+        # Platforms (if multi-platform)
+        if len(report["platforms"]) > 1:
+            lines.append("**📱 Platforms:**")
+            for p in report["platforms"]:
+                lines.append(f"  {p['platform']} — {p['sessions']} sessions, {p['messages']:,} msgs")
+            lines.append("")
+
+        # Tools (top 8)
+        if report["tools"]:
+            lines.append("**🔧 Top Tools:**")
+            for t in report["tools"][:8]:
+                lines.append(f"  {t['tool']} — {t['count']:,} calls ({t['percentage']:.1f}%)")
+            lines.append("")
+
+        # Activity summary
+        act = report.get("activity", {})
+        if act.get("busiest_day") and act.get("busiest_hour"):
+            hr = act["busiest_hour"]["hour"]
+            ampm = "AM" if hr < 12 else "PM"
+            display_hr = hr % 12 or 12
+            lines.append(f"**📅 Busiest:** {act['busiest_day']['day']}s ({act['busiest_day']['count']} sessions), {display_hr}{ampm} ({act['busiest_hour']['count']} sessions)")
+            if act.get("active_days"):
+                lines.append(f"**Active days:** {act['active_days']}", )
+            if act.get("max_streak", 0) > 1:
+                lines.append(f"**Best streak:** {act['max_streak']} consecutive days")
+
+        return "\n".join(lines)

agent/model_metadata.py

@@ -0,0 +1,931 @@
+"""Model metadata, context lengths, and token estimation utilities.
+
+Pure utility functions with no AIAgent dependency. Used by ContextCompressor
+and run_agent.py for pre-flight context checks.
+"""
+
+import logging
+import os
+import re
+import time
+from pathlib import Path
+from typing import Any, Dict, List, Optional
+from urllib.parse import urlparse
+
+import requests
+import yaml
+
+from hermes_constants import OPENROUTER_MODELS_URL
+
+logger = logging.getLogger(__name__)
+
+# Provider names that can appear as a "provider:" prefix before a model ID.
+# Only these are stripped — Ollama-style "model:tag" colons (e.g. "qwen3.5:27b")
+# are preserved so the full model name reaches cache lookups and server queries.
+_PROVIDER_PREFIXES: frozenset[str] = frozenset({
+    "openrouter", "nous", "openai-codex", "copilot", "copilot-acp",
+    "zai", "kimi-coding", "minimax", "minimax-cn", "anthropic", "deepseek",
+    "opencode-zen", "opencode-go", "ai-gateway", "kilocode", "alibaba",
+    "custom", "local",
+    # Common aliases
+    "glm", "z-ai", "z.ai", "zhipu", "github", "github-copilot",
+    "github-models", "kimi", "moonshot", "claude", "deep-seek",
+    "opencode", "zen", "go", "vercel", "kilo", "dashscope", "aliyun", "qwen",
+})
+
+
+_OLLAMA_TAG_PATTERN = re.compile(
+    r"^(\d+\.?\d*b|latest|stable|q\d|fp?\d|instruct|chat|coder|vision|text)",
+    re.IGNORECASE,
+)
+
+
+def _strip_provider_prefix(model: str) -> str:
+    """Strip a recognised provider prefix from a model string.
+
+    ``"local:my-model"`` → ``"my-model"``
+    ``"qwen3.5:27b"``   → ``"qwen3.5:27b"``  (unchanged — not a provider prefix)
+    ``"qwen:0.5b"``     → ``"qwen:0.5b"``    (unchanged — Ollama model:tag)
+    ``"deepseek:latest"``→ ``"deepseek:latest"``(unchanged — Ollama model:tag)
+    """
+    if ":" not in model or model.startswith("http"):
+        return model
+    prefix, suffix = model.split(":", 1)
+    prefix_lower = prefix.strip().lower()
+    if prefix_lower in _PROVIDER_PREFIXES:
+        # Don't strip if suffix looks like an Ollama tag (e.g. "7b", "latest", "q4_0")
+        if _OLLAMA_TAG_PATTERN.match(suffix.strip()):
+            return model
+        return suffix
+    return model
+
+_model_metadata_cache: Dict[str, Dict[str, Any]] = {}
+_model_metadata_cache_time: float = 0
+_MODEL_CACHE_TTL = 3600
+_endpoint_model_metadata_cache: Dict[str, Dict[str, Dict[str, Any]]] = {}
+_endpoint_model_metadata_cache_time: Dict[str, float] = {}
+_ENDPOINT_MODEL_CACHE_TTL = 300
+
+# Descending tiers for context length probing when the model is unknown.
+# We start at 128K (a safe default for most modern models) and step down
+# on context-length errors until one works.
+CONTEXT_PROBE_TIERS = [
+    128_000,
+    64_000,
+    32_000,
+    16_000,
+    8_000,
+]
+
+# Default context length when no detection method succeeds.
+DEFAULT_FALLBACK_CONTEXT = CONTEXT_PROBE_TIERS[0]
+
+# Thin fallback defaults — only broad model family patterns.
+# These fire only when provider is unknown AND models.dev/OpenRouter/Anthropic
+# all miss. Replaced the previous 80+ entry dict.
+# For provider-specific context lengths, models.dev is the primary source.
+DEFAULT_CONTEXT_LENGTHS = {
+    # Anthropic Claude 4.6 (1M context) — bare IDs only to avoid
+    # fuzzy-match collisions (e.g. "anthropic/claude-sonnet-4" is a
+    # substring of "anthropic/claude-sonnet-4.6").
+    # OpenRouter-prefixed models resolve via OpenRouter live API or models.dev.
+    "claude-opus-4-6": 1000000,
+    "claude-sonnet-4-6": 1000000,
+    "claude-opus-4.6": 1000000,
+    "claude-sonnet-4.6": 1000000,
+    # Catch-all for older Claude models (must sort after specific entries)
+    "claude": 200000,
+    # OpenAI
+    "gpt-4.1": 1047576,
+    "gpt-5": 128000,
+    "gpt-4": 128000,
+    # Google
+    "gemini": 1048576,
+    # DeepSeek
+    "deepseek": 128000,
+    # Meta
+    "llama": 131072,
+    # Qwen
+    "qwen": 131072,
+    # MiniMax
+    "minimax": 204800,
+    # GLM
+    "glm": 202752,
+    # Kimi
+    "kimi": 262144,
+    # Hugging Face Inference Providers — model IDs use org/name format
+    "Qwen/Qwen3.5-397B-A17B": 131072,
+    "Qwen/Qwen3.5-35B-A3B": 131072,
+    "deepseek-ai/DeepSeek-V3.2": 65536,
+    "moonshotai/Kimi-K2.5": 262144,
+    "moonshotai/Kimi-K2-Thinking": 262144,
+    "MiniMaxAI/MiniMax-M2.5": 204800,
+    "XiaomiMiMo/MiMo-V2-Flash": 32768,
+    "zai-org/GLM-5": 202752,
+}
+
+_CONTEXT_LENGTH_KEYS = (
+    "context_length",
+    "context_window",
+    "max_context_length",
+    "max_position_embeddings",
+    "max_model_len",
+    "max_input_tokens",
+    "max_sequence_length",
+    "max_seq_len",
+    "n_ctx_train",
+    "n_ctx",
+)
+
+_MAX_COMPLETION_KEYS = (
+    "max_completion_tokens",
+    "max_output_tokens",
+    "max_tokens",
+)
+
+# Local server hostnames / address patterns
+_LOCAL_HOSTS = ("localhost", "127.0.0.1", "::1", "0.0.0.0")
+
+
+def _normalize_base_url(base_url: str) -> str:
+    return (base_url or "").strip().rstrip("/")
+
+
+def _is_openrouter_base_url(base_url: str) -> bool:
+    return "openrouter.ai" in _normalize_base_url(base_url).lower()
+
+
+def _is_custom_endpoint(base_url: str) -> bool:
+    normalized = _normalize_base_url(base_url)
+    return bool(normalized) and not _is_openrouter_base_url(normalized)
+
+
+_URL_TO_PROVIDER: Dict[str, str] = {
+    "api.openai.com": "openai",
+    "chatgpt.com": "openai",
+    "api.anthropic.com": "anthropic",
+    "api.z.ai": "zai",
+    "api.moonshot.ai": "kimi-coding",
+    "api.kimi.com": "kimi-coding",
+    "api.minimax": "minimax",
+    "dashscope.aliyuncs.com": "alibaba",
+    "dashscope-intl.aliyuncs.com": "alibaba",
+    "openrouter.ai": "openrouter",
+    "generativelanguage.googleapis.com": "google",
+    "inference-api.nousresearch.com": "nous",
+    "api.deepseek.com": "deepseek",
+    "api.githubcopilot.com": "copilot",
+    "models.github.ai": "copilot",
+    "api.fireworks.ai": "fireworks",
+}
+
+
+def _infer_provider_from_url(base_url: str) -> Optional[str]:
+    """Infer the models.dev provider name from a base URL.
+
+    This allows context length resolution via models.dev for custom endpoints
+    like DashScope (Alibaba), Z.AI, Kimi, etc. without requiring the user to
+    explicitly set the provider name in config.
+    """
+    normalized = _normalize_base_url(base_url)
+    if not normalized:
+        return None
+    parsed = urlparse(normalized if "://" in normalized else f"https://{normalized}")
+    host = parsed.netloc.lower() or parsed.path.lower()
+    for url_part, provider in _URL_TO_PROVIDER.items():
+        if url_part in host:
+            return provider
+    return None
+
+
+def _is_known_provider_base_url(base_url: str) -> bool:
+    return _infer_provider_from_url(base_url) is not None
+
+
+def is_local_endpoint(base_url: str) -> bool:
+    """Return True if base_url points to a local machine (localhost / RFC-1918 / WSL)."""
+    normalized = _normalize_base_url(base_url)
+    if not normalized:
+        return False
+    url = normalized if "://" in normalized else f"http://{normalized}"
+    try:
+        parsed = urlparse(url)
+        host = parsed.hostname or ""
+    except Exception:
+        return False
+    if host in _LOCAL_HOSTS:
+        return True
+    # RFC-1918 private ranges and link-local
+    import ipaddress
+    try:
+        addr = ipaddress.ip_address(host)
+        return addr.is_private or addr.is_loopback or addr.is_link_local
+    except ValueError:
+        pass
+    # Bare IP that looks like a private range (e.g. 172.26.x.x for WSL)
+    parts = host.split(".")
+    if len(parts) == 4:
+        try:
+            first, second = int(parts[0]), int(parts[1])
+            if first == 10:
+                return True
+            if first == 172 and 16 <= second <= 31:
+                return True
+            if first == 192 and second == 168:
+                return True
+        except ValueError:
+            pass
+    return False
+
+
+def detect_local_server_type(base_url: str) -> Optional[str]:
+    """Detect which local server is running at base_url by probing known endpoints.
+
+    Returns one of: "ollama", "lm-studio", "vllm", "llamacpp", or None.
+    """
+    import httpx
+
+    normalized = _normalize_base_url(base_url)
+    server_url = normalized
+    if server_url.endswith("/v1"):
+        server_url = server_url[:-3]
+
+    try:
+        with httpx.Client(timeout=2.0) as client:
+            # LM Studio exposes /api/v1/models — check first (most specific)
+            try:
+                r = client.get(f"{server_url}/api/v1/models")
+                if r.status_code == 200:
+                    return "lm-studio"
+            except Exception:
+                pass
+            # Ollama exposes /api/tags and responds with {"models": [...]}
+            # LM Studio returns {"error": "Unexpected endpoint"} with status 200
+            # on this path, so we must verify the response contains "models".
+            try:
+                r = client.get(f"{server_url}/api/tags")
+                if r.status_code == 200:
+                    try:
+                        data = r.json()
+                        if "models" in data:
+                            return "ollama"
+                    except Exception:
+                        pass
+            except Exception:
+                pass
+            # llama.cpp exposes /v1/props (older builds used /props without the /v1 prefix)
+            try:
+                r = client.get(f"{server_url}/v1/props")
+                if r.status_code != 200:
+                    r = client.get(f"{server_url}/props")  # fallback for older builds
+                if r.status_code == 200 and "default_generation_settings" in r.text:
+                    return "llamacpp"
+            except Exception:
+                pass
+            # vLLM: /version
+            try:
+                r = client.get(f"{server_url}/version")
+                if r.status_code == 200:
+                    data = r.json()
+                    if "version" in data:
+                        return "vllm"
+            except Exception:
+                pass
+    except Exception:
+        pass
+
+    return None
+
+
+def _iter_nested_dicts(value: Any):
+    if isinstance(value, dict):
+        yield value
+        for nested in value.values():
+            yield from _iter_nested_dicts(nested)
+    elif isinstance(value, list):
+        for item in value:
+            yield from _iter_nested_dicts(item)
+
+
+def _coerce_reasonable_int(value: Any, minimum: int = 1024, maximum: int = 10_000_000) -> Optional[int]:
+    try:
+        if isinstance(value, bool):
+            return None
+        if isinstance(value, str):
+            value = value.strip().replace(",", "")
+        result = int(value)
+    except (TypeError, ValueError):
+        return None
+    if minimum <= result <= maximum:
+        return result
+    return None
+
+
+def _extract_first_int(payload: Dict[str, Any], keys: tuple[str, ...]) -> Optional[int]:
+    keyset = {key.lower() for key in keys}
+    for mapping in _iter_nested_dicts(payload):
+        for key, value in mapping.items():
+            if str(key).lower() not in keyset:
+                continue
+            coerced = _coerce_reasonable_int(value)
+            if coerced is not None:
+                return coerced
+    return None
+
+
+def _extract_context_length(payload: Dict[str, Any]) -> Optional[int]:
+    return _extract_first_int(payload, _CONTEXT_LENGTH_KEYS)
+
+
+def _extract_max_completion_tokens(payload: Dict[str, Any]) -> Optional[int]:
+    return _extract_first_int(payload, _MAX_COMPLETION_KEYS)
+
+
+def _extract_pricing(payload: Dict[str, Any]) -> Dict[str, Any]:
+    alias_map = {
+        "prompt": ("prompt", "input", "input_cost_per_token", "prompt_token_cost"),
+        "completion": ("completion", "output", "output_cost_per_token", "completion_token_cost"),
+        "request": ("request", "request_cost"),
+        "cache_read": ("cache_read", "cached_prompt", "input_cache_read", "cache_read_cost_per_token"),
+        "cache_write": ("cache_write", "cache_creation", "input_cache_write", "cache_write_cost_per_token"),
+    }
+    for mapping in _iter_nested_dicts(payload):
+        normalized = {str(key).lower(): value for key, value in mapping.items()}
+        if not any(any(alias in normalized for alias in aliases) for aliases in alias_map.values()):
+            continue
+        pricing: Dict[str, Any] = {}
+        for target, aliases in alias_map.items():
+            for alias in aliases:
+                if alias in normalized and normalized[alias] not in (None, ""):
+                    pricing[target] = normalized[alias]
+                    break
+        if pricing:
+            return pricing
+    return {}
+
+
+def _add_model_aliases(cache: Dict[str, Dict[str, Any]], model_id: str, entry: Dict[str, Any]) -> None:
+    cache[model_id] = entry
+    if "/" in model_id:
+        bare_model = model_id.split("/", 1)[1]
+        cache.setdefault(bare_model, entry)
+
+
+def fetch_model_metadata(force_refresh: bool = False) -> Dict[str, Dict[str, Any]]:
+    """Fetch model metadata from OpenRouter (cached for 1 hour)."""
+    global _model_metadata_cache, _model_metadata_cache_time
+
+    if not force_refresh and _model_metadata_cache and (time.time() - _model_metadata_cache_time) < _MODEL_CACHE_TTL:
+        return _model_metadata_cache
+
+    try:
+        response = requests.get(OPENROUTER_MODELS_URL, timeout=10)
+        response.raise_for_status()
+        data = response.json()
+
+        cache = {}
+        for model in data.get("data", []):
+            model_id = model.get("id", "")
+            entry = {
+                "context_length": model.get("context_length", 128000),
+                "max_completion_tokens": model.get("top_provider", {}).get("max_completion_tokens", 4096),
+                "name": model.get("name", model_id),
+                "pricing": model.get("pricing", {}),
+            }
+            _add_model_aliases(cache, model_id, entry)
+            canonical = model.get("canonical_slug", "")
+            if canonical and canonical != model_id:
+                _add_model_aliases(cache, canonical, entry)
+
+        _model_metadata_cache = cache
+        _model_metadata_cache_time = time.time()
+        logger.debug("Fetched metadata for %s models from OpenRouter", len(cache))
+        return cache
+
+    except Exception as e:
+        logging.warning(f"Failed to fetch model metadata from OpenRouter: {e}")
+        return _model_metadata_cache or {}
+
+
+def fetch_endpoint_model_metadata(
+    base_url: str,
+    api_key: str = "",
+    force_refresh: bool = False,
+) -> Dict[str, Dict[str, Any]]:
+    """Fetch model metadata from an OpenAI-compatible ``/models`` endpoint.
+
+    This is used for explicit custom endpoints where hardcoded global model-name
+    defaults are unreliable. Results are cached in memory per base URL.
+    """
+    normalized = _normalize_base_url(base_url)
+    if not normalized or _is_openrouter_base_url(normalized):
+        return {}
+
+    if not force_refresh:
+        cached = _endpoint_model_metadata_cache.get(normalized)
+        cached_at = _endpoint_model_metadata_cache_time.get(normalized, 0)
+        if cached is not None and (time.time() - cached_at) < _ENDPOINT_MODEL_CACHE_TTL:
+            return cached
+
+    candidates = [normalized]
+    if normalized.endswith("/v1"):
+        alternate = normalized[:-3].rstrip("/")
+    else:
+        alternate = normalized + "/v1"
+    if alternate and alternate not in candidates:
+        candidates.append(alternate)
+
+    headers = {"Authorization": f"Bearer {api_key}"} if api_key else {}
+    last_error: Optional[Exception] = None
+
+    for candidate in candidates:
+        url = candidate.rstrip("/") + "/models"
+        try:
+            response = requests.get(url, headers=headers, timeout=10)
+            response.raise_for_status()
+            payload = response.json()
+            cache: Dict[str, Dict[str, Any]] = {}
+            for model in payload.get("data", []):
+                if not isinstance(model, dict):
+                    continue
+                model_id = model.get("id")
+                if not model_id:
+                    continue
+                entry: Dict[str, Any] = {"name": model.get("name", model_id)}
+                context_length = _extract_context_length(model)
+                if context_length is not None:
+                    entry["context_length"] = context_length
+                max_completion_tokens = _extract_max_completion_tokens(model)
+                if max_completion_tokens is not None:
+                    entry["max_completion_tokens"] = max_completion_tokens
+                pricing = _extract_pricing(model)
+                if pricing:
+                    entry["pricing"] = pricing
+                _add_model_aliases(cache, model_id, entry)
+
+            # If this is a llama.cpp server, query /props for actual allocated context
+            is_llamacpp = any(
+                m.get("owned_by") == "llamacpp"
+                for m in payload.get("data", []) if isinstance(m, dict)
+            )
+            if is_llamacpp:
+                try:
+                    # Try /v1/props first (current llama.cpp); fall back to /props for older builds
+                    base = candidate.rstrip("/").replace("/v1", "")
+                    props_resp = requests.get(base + "/v1/props", headers=headers, timeout=5)
+                    if not props_resp.ok:
+                        props_resp = requests.get(base + "/props", headers=headers, timeout=5)
+                    if props_resp.ok:
+                        props = props_resp.json()
+                        gen_settings = props.get("default_generation_settings", {})
+                        n_ctx = gen_settings.get("n_ctx")
+                        model_alias = props.get("model_alias", "")
+                        if n_ctx and model_alias and model_alias in cache:
+                            cache[model_alias]["context_length"] = n_ctx
+                except Exception:
+                    pass
+
+            _endpoint_model_metadata_cache[normalized] = cache
+            _endpoint_model_metadata_cache_time[normalized] = time.time()
+            return cache
+        except Exception as exc:
+            last_error = exc
+
+    if last_error:
+        logger.debug("Failed to fetch model metadata from %s/models: %s", normalized, last_error)
+    _endpoint_model_metadata_cache[normalized] = {}
+    _endpoint_model_metadata_cache_time[normalized] = time.time()
+    return {}
+
+
+def _get_context_cache_path() -> Path:
+    """Return path to the persistent context length cache file."""
+    hermes_home = Path(os.environ.get("HERMES_HOME", Path.home() / ".hermes"))
+    return hermes_home / "context_length_cache.yaml"
+
+
+def _load_context_cache() -> Dict[str, int]:
+    """Load the model+provider -> context_length cache from disk."""
+    path = _get_context_cache_path()
+    if not path.exists():
+        return {}
+    try:
+        with open(path) as f:
+            data = yaml.safe_load(f) or {}
+        return data.get("context_lengths", {})
+    except Exception as e:
+        logger.debug("Failed to load context length cache: %s", e)
+        return {}
+
+
+def save_context_length(model: str, base_url: str, length: int) -> None:
+    """Persist a discovered context length for a model+provider combo.
+
+    Cache key is ``model@base_url`` so the same model name served from
+    different providers can have different limits.
+    """
+    key = f"{model}@{base_url}"
+    cache = _load_context_cache()
+    if cache.get(key) == length:
+        return  # already stored
+    cache[key] = length
+    path = _get_context_cache_path()
+    try:
+        path.parent.mkdir(parents=True, exist_ok=True)
+        with open(path, "w") as f:
+            yaml.dump({"context_lengths": cache}, f, default_flow_style=False)
+        logger.info("Cached context length %s -> %s tokens", key, f"{length:,}")
+    except Exception as e:
+        logger.debug("Failed to save context length cache: %s", e)
+
+
+def get_cached_context_length(model: str, base_url: str) -> Optional[int]:
+    """Look up a previously discovered context length for model+provider."""
+    key = f"{model}@{base_url}"
+    cache = _load_context_cache()
+    return cache.get(key)
+
+
+def get_next_probe_tier(current_length: int) -> Optional[int]:
+    """Return the next lower probe tier, or None if already at minimum."""
+    for tier in CONTEXT_PROBE_TIERS:
+        if tier < current_length:
+            return tier
+    return None
+
+
+def parse_context_limit_from_error(error_msg: str) -> Optional[int]:
+    """Try to extract the actual context limit from an API error message.
+
+    Many providers include the limit in their error text, e.g.:
+      - "maximum context length is 32768 tokens"
+      - "context_length_exceeded: 131072"
+      - "Maximum context size 32768 exceeded"
+      - "model's max context length is 65536"
+    """
+    error_lower = error_msg.lower()
+    # Pattern: look for numbers near context-related keywords
+    patterns = [
+        r'(?:max(?:imum)?|limit)\s*(?:context\s*)?(?:length|size|window)?\s*(?:is|of|:)?\s*(\d{4,})',
+        r'context\s*(?:length|size|window)\s*(?:is|of|:)?\s*(\d{4,})',
+        r'(\d{4,})\s*(?:token)?\s*(?:context|limit)',
+        r'>\s*(\d{4,})\s*(?:max|limit|token)',  # "250000 tokens > 200000 maximum"
+        r'(\d{4,})\s*(?:max(?:imum)?)\b',  # "200000 maximum"
+    ]
+    for pattern in patterns:
+        match = re.search(pattern, error_lower)
+        if match:
+            limit = int(match.group(1))
+            # Sanity check: must be a reasonable context length
+            if 1024 <= limit <= 10_000_000:
+                return limit
+    return None
+
+
+def _model_id_matches(candidate_id: str, lookup_model: str) -> bool:
+    """Return True if *candidate_id* (from server) matches *lookup_model* (configured).
+
+    Supports two forms:
+    - Exact match:  "nvidia-nemotron-super-49b-v1" == "nvidia-nemotron-super-49b-v1"
+    - Slug match:   "nvidia/nvidia-nemotron-super-49b-v1" matches "nvidia-nemotron-super-49b-v1"
+                    (the part after the last "/" equals lookup_model)
+
+    This covers LM Studio's native API which stores models as "publisher/slug"
+    while users typically configure only the slug after the "local:" prefix.
+    """
+    if candidate_id == lookup_model:
+        return True
+    # Slug match: basename of candidate equals the lookup name
+    if "/" in candidate_id and candidate_id.rsplit("/", 1)[1] == lookup_model:
+        return True
+    return False
+
+
+def _query_local_context_length(model: str, base_url: str) -> Optional[int]:
+    """Query a local server for the model's context length."""
+    import httpx
+
+    # Strip recognised provider prefix (e.g., "local:model-name" → "model-name").
+    # Ollama "model:tag" colons (e.g. "qwen3.5:27b") are intentionally preserved.
+    model = _strip_provider_prefix(model)
+
+    # Strip /v1 suffix to get the server root
+    server_url = base_url.rstrip("/")
+    if server_url.endswith("/v1"):
+        server_url = server_url[:-3]
+
+    try:
+        server_type = detect_local_server_type(base_url)
+    except Exception:
+        server_type = None
+
+    try:
+        with httpx.Client(timeout=3.0) as client:
+            # Ollama: /api/show returns model details with context info
+            if server_type == "ollama":
+                resp = client.post(f"{server_url}/api/show", json={"name": model})
+                if resp.status_code == 200:
+                    data = resp.json()
+                    # Check model_info for context length
+                    model_info = data.get("model_info", {})
+                    for key, value in model_info.items():
+                        if "context_length" in key and isinstance(value, (int, float)):
+                            return int(value)
+                    # Check parameters string for num_ctx
+                    params = data.get("parameters", "")
+                    if "num_ctx" in params:
+                        for line in params.split("\n"):
+                            if "num_ctx" in line:
+                                parts = line.strip().split()
+                                if len(parts) >= 2:
+                                    try:
+                                        return int(parts[-1])
+                                    except ValueError:
+                                        pass
+
+            # LM Studio native API: /api/v1/models returns max_context_length.
+            # This is more reliable than the OpenAI-compat /v1/models which
+            # doesn't include context window information for LM Studio servers.
+            # Use _model_id_matches for fuzzy matching: LM Studio stores models as
+            # "publisher/slug" but users configure only "slug" after "local:" prefix.
+            if server_type == "lm-studio":
+                resp = client.get(f"{server_url}/api/v1/models")
+                if resp.status_code == 200:
+                    data = resp.json()
+                    for m in data.get("models", []):
+                        if _model_id_matches(m.get("key", ""), model) or _model_id_matches(m.get("id", ""), model):
+                            # Prefer loaded instance context (actual runtime value)
+                            for inst in m.get("loaded_instances", []):
+                                cfg = inst.get("config", {})
+                                ctx = cfg.get("context_length")
+                                if ctx and isinstance(ctx, (int, float)):
+                                    return int(ctx)
+                            # Fall back to max_context_length (theoretical model max)
+                            ctx = m.get("max_context_length") or m.get("context_length")
+                            if ctx and isinstance(ctx, (int, float)):
+                                return int(ctx)
+
+            # LM Studio / vLLM / llama.cpp: try /v1/models/{model}
+            resp = client.get(f"{server_url}/v1/models/{model}")
+            if resp.status_code == 200:
+                data = resp.json()
+                # vLLM returns max_model_len
+                ctx = data.get("max_model_len") or data.get("context_length") or data.get("max_tokens")
+                if ctx and isinstance(ctx, (int, float)):
+                    return int(ctx)
+
+            # Try /v1/models and find the model in the list.
+            # Use _model_id_matches to handle "publisher/slug" vs bare "slug".
+            resp = client.get(f"{server_url}/v1/models")
+            if resp.status_code == 200:
+                data = resp.json()
+                models_list = data.get("data", [])
+                for m in models_list:
+                    if _model_id_matches(m.get("id", ""), model):
+                        ctx = m.get("max_model_len") or m.get("context_length") or m.get("max_tokens")
+                        if ctx and isinstance(ctx, (int, float)):
+                            return int(ctx)
+    except Exception:
+        pass
+
+    return None
+
+
+def _normalize_model_version(model: str) -> str:
+    """Normalize version separators for matching.
+
+    Nous uses dashes: claude-opus-4-6, claude-sonnet-4-5
+    OpenRouter uses dots: claude-opus-4.6, claude-sonnet-4.5
+    Normalize both to dashes for comparison.
+    """
+    return model.replace(".", "-")
+
+
+def _query_anthropic_context_length(model: str, base_url: str, api_key: str) -> Optional[int]:
+    """Query Anthropic's /v1/models endpoint for context length.
+
+    Only works with regular ANTHROPIC_API_KEY (sk-ant-api*).
+    OAuth tokens (sk-ant-oat*) from Claude Code return 401.
+    """
+    if not api_key or api_key.startswith("sk-ant-oat"):
+        return None  # OAuth tokens can't access /v1/models
+    try:
+        base = base_url.rstrip("/")
+        if base.endswith("/v1"):
+            base = base[:-3]
+        url = f"{base}/v1/models?limit=1000"
+        headers = {
+            "x-api-key": api_key,
+            "anthropic-version": "2023-06-01",
+        }
+        resp = requests.get(url, headers=headers, timeout=10)
+        if resp.status_code != 200:
+            return None
+        data = resp.json()
+        for m in data.get("data", []):
+            if m.get("id") == model:
+                ctx = m.get("max_input_tokens")
+                if isinstance(ctx, int) and ctx > 0:
+                    return ctx
+    except Exception as e:
+        logger.debug("Anthropic /v1/models query failed: %s", e)
+    return None
+
+
+def _resolve_nous_context_length(model: str) -> Optional[int]:
+    """Resolve Nous Portal model context length via OpenRouter metadata.
+
+    Nous model IDs are bare (e.g. 'claude-opus-4-6') while OpenRouter uses
+    prefixed IDs (e.g. 'anthropic/claude-opus-4.6'). Try suffix matching
+    with version normalization (dot↔dash).
+    """
+    metadata = fetch_model_metadata()  # OpenRouter cache
+    # Exact match first
+    if model in metadata:
+        return metadata[model].get("context_length")
+
+    normalized = _normalize_model_version(model).lower()
+
+    for or_id, entry in metadata.items():
+        bare = or_id.split("/", 1)[1] if "/" in or_id else or_id
+        if bare.lower() == model.lower() or _normalize_model_version(bare).lower() == normalized:
+            return entry.get("context_length")
+
+    # Partial prefix match for cases like gemini-3-flash → gemini-3-flash-preview
+    # Require match to be at a word boundary (followed by -, :, or end of string)
+    model_lower = model.lower()
+    for or_id, entry in metadata.items():
+        bare = or_id.split("/", 1)[1] if "/" in or_id else or_id
+        for candidate, query in [(bare.lower(), model_lower), (_normalize_model_version(bare).lower(), normalized)]:
+            if candidate.startswith(query) and (
+                len(candidate) == len(query) or candidate[len(query)] in "-:."
+            ):
+                return entry.get("context_length")
+
+    return None
+
+
+def get_model_context_length(
+    model: str,
+    base_url: str = "",
+    api_key: str = "",
+    config_context_length: int | None = None,
+    provider: str = "",
+) -> int:
+    """Get the context length for a model.
+
+    Resolution order:
+    0. Explicit config override (model.context_length or custom_providers per-model)
+    1. Persistent cache (previously discovered via probing)
+    2. Active endpoint metadata (/models for explicit custom endpoints)
+    3. Local server query (for local endpoints)
+    4. Anthropic /v1/models API (API-key users only, not OAuth)
+    5. OpenRouter live API metadata
+    6. Nous suffix-match via OpenRouter cache
+    7. models.dev registry lookup (provider-aware)
+    8. Thin hardcoded defaults (broad family patterns)
+    9. Default fallback (128K)
+    """
+    # 0. Explicit config override — user knows best
+    if config_context_length is not None and isinstance(config_context_length, int) and config_context_length > 0:
+        return config_context_length
+
+    # Normalise provider-prefixed model names (e.g. "local:model-name" →
+    # "model-name") so cache lookups and server queries use the bare ID that
+    # local servers actually know about.  Ollama "model:tag" colons are preserved.
+    model = _strip_provider_prefix(model)
+
+    # 1. Check persistent cache (model+provider)
+    if base_url:
+        cached = get_cached_context_length(model, base_url)
+        if cached is not None:
+            return cached
+
+    # 2. Active endpoint metadata for truly custom/unknown endpoints.
+    # Known providers (Copilot, OpenAI, Anthropic, etc.) skip this — their
+    # /models endpoint may report a provider-imposed limit (e.g. Copilot
+    # returns 128k) instead of the model's full context (400k).  models.dev
+    # has the correct per-provider values and is checked at step 5+.
+    if _is_custom_endpoint(base_url) and not _is_known_provider_base_url(base_url):
+        endpoint_metadata = fetch_endpoint_model_metadata(base_url, api_key=api_key)
+        matched = endpoint_metadata.get(model)
+        if not matched:
+            # Single-model servers: if only one model is loaded, use it
+            if len(endpoint_metadata) == 1:
+                matched = next(iter(endpoint_metadata.values()))
+            else:
+                # Fuzzy match: substring in either direction
+                for key, entry in endpoint_metadata.items():
+                    if model in key or key in model:
+                        matched = entry
+                        break
+        if matched:
+            context_length = matched.get("context_length")
+            if isinstance(context_length, int):
+                return context_length
+        if not _is_known_provider_base_url(base_url):
+            # 3. Try querying local server directly
+            if is_local_endpoint(base_url):
+                local_ctx = _query_local_context_length(model, base_url)
+                if local_ctx and local_ctx > 0:
+                    save_context_length(model, base_url, local_ctx)
+                    return local_ctx
+            logger.info(
+                "Could not detect context length for model %r at %s — "
+                "defaulting to %s tokens (probe-down). Set model.context_length "
+                "in config.yaml to override.",
+                model, base_url, f"{DEFAULT_FALLBACK_CONTEXT:,}",
+            )
+            return DEFAULT_FALLBACK_CONTEXT
+
+    # 4. Anthropic /v1/models API (only for regular API keys, not OAuth)
+    if provider == "anthropic" or (
+        base_url and "api.anthropic.com" in base_url
+    ):
+        ctx = _query_anthropic_context_length(model, base_url or "https://api.anthropic.com", api_key)
+        if ctx:
+            return ctx
+
+    # 5. Provider-aware lookups (before generic OpenRouter cache)
+    # These are provider-specific and take priority over the generic OR cache,
+    # since the same model can have different context limits per provider
+    # (e.g. claude-opus-4.6 is 1M on Anthropic but 128K on GitHub Copilot).
+    # If provider is generic (openrouter/custom/empty), try to infer from URL.
+    effective_provider = provider
+    if not effective_provider or effective_provider in ("openrouter", "custom"):
+        if base_url:
+            inferred = _infer_provider_from_url(base_url)
+            if inferred:
+                effective_provider = inferred
+
+    if effective_provider == "nous":
+        ctx = _resolve_nous_context_length(model)
+        if ctx:
+            return ctx
+    if effective_provider:
+        from agent.models_dev import lookup_models_dev_context
+        ctx = lookup_models_dev_context(effective_provider, model)
+        if ctx:
+            return ctx
+
+    # 6. OpenRouter live API metadata (provider-unaware fallback)
+    metadata = fetch_model_metadata()
+    if model in metadata:
+        return metadata[model].get("context_length", 128000)
+
+    # 8. Hardcoded defaults (fuzzy match — longest key first for specificity)
+    # Only check `default_model in model` (is the key a substring of the input).
+    # The reverse (`model in default_model`) causes shorter names like
+    # "claude-sonnet-4" to incorrectly match "claude-sonnet-4-6" and return 1M.
+    model_lower = model.lower()
+    for default_model, length in sorted(
+        DEFAULT_CONTEXT_LENGTHS.items(), key=lambda x: len(x[0]), reverse=True
+    ):
+        if default_model in model_lower:
+            return length
+
+    # 9. Query local server as last resort
+    if base_url and is_local_endpoint(base_url):
+        local_ctx = _query_local_context_length(model, base_url)
+        if local_ctx and local_ctx > 0:
+            save_context_length(model, base_url, local_ctx)
+            return local_ctx
+
+    # 10. Default fallback — 128K
+    return DEFAULT_FALLBACK_CONTEXT
+
+
+def estimate_tokens_rough(text: str) -> int:
+    """Rough token estimate (~4 chars/token) for pre-flight checks."""
+    if not text:
+        return 0
+    return len(text) // 4
+
+
+def estimate_messages_tokens_rough(messages: List[Dict[str, Any]]) -> int:
+    """Rough token estimate for a message list (pre-flight only)."""
+    total_chars = sum(len(str(msg)) for msg in messages)
+    return total_chars // 4
+
+
+def estimate_request_tokens_rough(
+    messages: List[Dict[str, Any]],
+    *,
+    system_prompt: str = "",
+    tools: Optional[List[Dict[str, Any]]] = None,
+) -> int:
+    """Rough token estimate for a full chat-completions request.
+
+    Includes the major payload buckets Hermes sends to providers:
+    system prompt, conversation messages, and tool schemas.  With 50+
+    tools enabled, schemas alone can add 20-30K tokens — a significant
+    blind spot when only counting messages.
+    """
+    total_chars = 0
+    if system_prompt:
+        total_chars += len(system_prompt)
+    if messages:
+        total_chars += sum(len(str(msg)) for msg in messages)
+    if tools:
+        total_chars += len(str(tools))
+    return total_chars // 4

agent/models_dev.py

@@ -0,0 +1,172 @@
+"""Models.dev registry integration for provider-aware context length detection.
+
+Fetches model metadata from https://models.dev/api.json — a community-maintained
+database of 3800+ models across 100+ providers, including per-provider context
+windows, pricing, and capabilities.
+
+Data is cached in memory (1hr TTL) and on disk (~/.hermes/models_dev_cache.json)
+to avoid cold-start network latency.
+"""
+
+import json
+import logging
+import os
+import time
+from pathlib import Path
+from typing import Any, Dict, Optional
+
+from utils import atomic_json_write
+
+import requests
+
+logger = logging.getLogger(__name__)
+
+MODELS_DEV_URL = "https://models.dev/api.json"
+_MODELS_DEV_CACHE_TTL = 3600  # 1 hour in-memory
+
+# In-memory cache
+_models_dev_cache: Dict[str, Any] = {}
+_models_dev_cache_time: float = 0
+
+# Provider ID mapping: Hermes provider names → models.dev provider IDs
+PROVIDER_TO_MODELS_DEV: Dict[str, str] = {
+    "openrouter": "openrouter",
+    "anthropic": "anthropic",
+    "zai": "zai",
+    "kimi-coding": "kimi-for-coding",
+    "minimax": "minimax",
+    "minimax-cn": "minimax-cn",
+    "deepseek": "deepseek",
+    "alibaba": "alibaba",
+    "copilot": "github-copilot",
+    "ai-gateway": "vercel",
+    "opencode-zen": "opencode",
+    "opencode-go": "opencode-go",
+    "kilocode": "kilo",
+    "fireworks": "fireworks-ai",
+}
+
+
+def _get_cache_path() -> Path:
+    """Return path to disk cache file."""
+    env_val = os.environ.get("HERMES_HOME", "")
+    hermes_home = Path(env_val) if env_val else Path.home() / ".hermes"
+    return hermes_home / "models_dev_cache.json"
+
+
+def _load_disk_cache() -> Dict[str, Any]:
+    """Load models.dev data from disk cache."""
+    try:
+        cache_path = _get_cache_path()
+        if cache_path.exists():
+            with open(cache_path, encoding="utf-8") as f:
+                return json.load(f)
+    except Exception as e:
+        logger.debug("Failed to load models.dev disk cache: %s", e)
+    return {}
+
+
+def _save_disk_cache(data: Dict[str, Any]) -> None:
+    """Save models.dev data to disk cache atomically."""
+    try:
+        cache_path = _get_cache_path()
+        atomic_json_write(cache_path, data, indent=None, separators=(",", ":"))
+    except Exception as e:
+        logger.debug("Failed to save models.dev disk cache: %s", e)
+
+
+def fetch_models_dev(force_refresh: bool = False) -> Dict[str, Any]:
+    """Fetch models.dev registry. In-memory cache (1hr) + disk fallback.
+
+    Returns the full registry dict keyed by provider ID, or empty dict on failure.
+    """
+    global _models_dev_cache, _models_dev_cache_time
+
+    # Check in-memory cache
+    if (
+        not force_refresh
+        and _models_dev_cache
+        and (time.time() - _models_dev_cache_time) < _MODELS_DEV_CACHE_TTL
+    ):
+        return _models_dev_cache
+
+    # Try network fetch
+    try:
+        response = requests.get(MODELS_DEV_URL, timeout=15)
+        response.raise_for_status()
+        data = response.json()
+        if isinstance(data, dict) and len(data) > 0:
+            _models_dev_cache = data
+            _models_dev_cache_time = time.time()
+            _save_disk_cache(data)
+            logger.debug(
+                "Fetched models.dev registry: %d providers, %d total models",
+                len(data),
+                sum(len(p.get("models", {})) for p in data.values() if isinstance(p, dict)),
+            )
+            return data
+    except Exception as e:
+        logger.debug("Failed to fetch models.dev: %s", e)
+
+    # Fall back to disk cache — use a short TTL (5 min) so we retry
+    # the network fetch soon instead of serving stale data for a full hour.
+    if not _models_dev_cache:
+        _models_dev_cache = _load_disk_cache()
+        if _models_dev_cache:
+            _models_dev_cache_time = time.time() - _MODELS_DEV_CACHE_TTL + 300
+            logger.debug("Loaded models.dev from disk cache (%d providers)", len(_models_dev_cache))
+
+    return _models_dev_cache
+
+
+def lookup_models_dev_context(provider: str, model: str) -> Optional[int]:
+    """Look up context_length for a provider+model combo in models.dev.
+
+    Returns the context window in tokens, or None if not found.
+    Handles case-insensitive matching and filters out context=0 entries.
+    """
+    mdev_provider_id = PROVIDER_TO_MODELS_DEV.get(provider)
+    if not mdev_provider_id:
+        return None
+
+    data = fetch_models_dev()
+    provider_data = data.get(mdev_provider_id)
+    if not isinstance(provider_data, dict):
+        return None
+
+    models = provider_data.get("models", {})
+    if not isinstance(models, dict):
+        return None
+
+    # Exact match
+    entry = models.get(model)
+    if entry:
+        ctx = _extract_context(entry)
+        if ctx:
+            return ctx
+
+    # Case-insensitive match
+    model_lower = model.lower()
+    for mid, mdata in models.items():
+        if mid.lower() == model_lower:
+            ctx = _extract_context(mdata)
+            if ctx:
+                return ctx
+
+    return None
+
+
+def _extract_context(entry: Dict[str, Any]) -> Optional[int]:
+    """Extract context_length from a models.dev model entry.
+
+    Returns None for invalid/zero values (some audio/image models have context=0).
+    """
+    if not isinstance(entry, dict):
+        return None
+    limit = entry.get("limit")
+    if not isinstance(limit, dict):
+        return None
+    ctx = limit.get("context")
+    if isinstance(ctx, (int, float)) and ctx > 0:
+        return int(ctx)
+    return None

agent/prompt_builder.py

@@ -0,0 +1,816 @@
+"""System prompt assembly -- identity, platform hints, skills index, context files.
+
+All functions are stateless. AIAgent._build_system_prompt() calls these to
+assemble pieces, then combines them with memory and ephemeral prompts.
+"""
+
+import json
+import logging
+import os
+import re
+import threading
+from collections import OrderedDict
+from pathlib import Path
+
+from hermes_constants import get_hermes_home
+from typing import Optional
+
+from agent.skill_utils import (
+    extract_skill_conditions,
+    extract_skill_description,
+    get_all_skills_dirs,
+    get_disabled_skill_names,
+    iter_skill_index_files,
+    parse_frontmatter,
+    skill_matches_platform,
+)
+from utils import atomic_json_write
+
+logger = logging.getLogger(__name__)
+
+# ---------------------------------------------------------------------------
+# Context file scanning — detect prompt injection in AGENTS.md, .cursorrules,
+# SOUL.md before they get injected into the system prompt.
+# ---------------------------------------------------------------------------
+
+_CONTEXT_THREAT_PATTERNS = [
+    (r'ignore\s+(previous|all|above|prior)\s+instructions', "prompt_injection"),
+    (r'do\s+not\s+tell\s+the\s+user', "deception_hide"),
+    (r'system\s+prompt\s+override', "sys_prompt_override"),
+    (r'disregard\s+(your|all|any)\s+(instructions|rules|guidelines)', "disregard_rules"),
+    (r'act\s+as\s+(if|though)\s+you\s+(have\s+no|don\'t\s+have)\s+(restrictions|limits|rules)', "bypass_restrictions"),
+    (r'<!--[^>]*(?:ignore|override|system|secret|hidden)[^>]*-->', "html_comment_injection"),
+    (r'<\s*div\s+style\s*=\s*["\'].*display\s*:\s*none', "hidden_div"),
+    (r'translate\s+.*\s+into\s+.*\s+and\s+(execute|run|eval)', "translate_execute"),
+    (r'curl\s+[^\n]*\$\{?\w*(KEY|TOKEN|SECRET|PASSWORD|CREDENTIAL|API)', "exfil_curl"),
+    (r'cat\s+[^\n]*(\.env|credentials|\.netrc|\.pgpass)', "read_secrets"),
+]
+
+_CONTEXT_INVISIBLE_CHARS = {
+    '\u200b', '\u200c', '\u200d', '\u2060', '\ufeff',
+    '\u202a', '\u202b', '\u202c', '\u202d', '\u202e',
+}
+
+
+def _scan_context_content(content: str, filename: str) -> str:
+    """Scan context file content for injection. Returns sanitized content."""
+    findings = []
+
+    # Check invisible unicode
+    for char in _CONTEXT_INVISIBLE_CHARS:
+        if char in content:
+            findings.append(f"invisible unicode U+{ord(char):04X}")
+
+    # Check threat patterns
+    for pattern, pid in _CONTEXT_THREAT_PATTERNS:
+        if re.search(pattern, content, re.IGNORECASE):
+            findings.append(pid)
+
+    if findings:
+        logger.warning("Context file %s blocked: %s", filename, ", ".join(findings))
+        return f"[BLOCKED: {filename} contained potential prompt injection ({', '.join(findings)}). Content not loaded.]"
+
+    return content
+
+
+def _find_git_root(start: Path) -> Optional[Path]:
+    """Walk *start* and its parents looking for a ``.git`` directory.
+
+    Returns the directory containing ``.git``, or ``None`` if we hit the
+    filesystem root without finding one.
+    """
+    current = start.resolve()
+    for parent in [current, *current.parents]:
+        if (parent / ".git").exists():
+            return parent
+    return None
+
+
+_HERMES_MD_NAMES = (".hermes.md", "HERMES.md")
+
+
+def _find_hermes_md(cwd: Path) -> Optional[Path]:
+    """Discover the nearest ``.hermes.md`` or ``HERMES.md``.
+
+    Search order: *cwd* first, then each parent directory up to (and
+    including) the git repository root.  Returns the first match, or
+    ``None`` if nothing is found.
+    """
+    stop_at = _find_git_root(cwd)
+    current = cwd.resolve()
+
+    for directory in [current, *current.parents]:
+        for name in _HERMES_MD_NAMES:
+            candidate = directory / name
+            if candidate.is_file():
+                return candidate
+        # Stop walking at the git root (or filesystem root).
+        if stop_at and directory == stop_at:
+            break
+    return None
+
+
+def _strip_yaml_frontmatter(content: str) -> str:
+    """Remove optional YAML frontmatter (``---`` delimited) from *content*.
+
+    The frontmatter may contain structured config (model overrides, tool
+    settings) that will be handled separately in a future PR.  For now we
+    strip it so only the human-readable markdown body is injected into the
+    system prompt.
+    """
+    if content.startswith("---"):
+        end = content.find("\n---", 3)
+        if end != -1:
+            # Skip past the closing --- and any trailing newline
+            body = content[end + 4:].lstrip("\n")
+            return body if body else content
+    return content
+
+
+# =========================================================================
+# Constants
+# =========================================================================
+
+DEFAULT_AGENT_IDENTITY = (
+    "You are Hermes Agent, an intelligent AI assistant created by Nous Research. "
+    "You are helpful, knowledgeable, and direct. You assist users with a wide "
+    "range of tasks including answering questions, writing and editing code, "
+    "analyzing information, creative work, and executing actions via your tools. "
+    "You communicate clearly, admit uncertainty when appropriate, and prioritize "
+    "being genuinely useful over being verbose unless otherwise directed below. "
+    "Be targeted and efficient in your exploration and investigations."
+)
+
+MEMORY_GUIDANCE = (
+    "You have persistent memory across sessions. Save durable facts using the memory "
+    "tool: user preferences, environment details, tool quirks, and stable conventions. "
+    "Memory is injected into every turn, so keep it compact and focused on facts that "
+    "will still matter later.\n"
+    "Prioritize what reduces future user steering — the most valuable memory is one "
+    "that prevents the user from having to correct or remind you again. "
+    "User preferences and recurring corrections matter more than procedural task details.\n"
+    "Do NOT save task progress, session outcomes, completed-work logs, or temporary TODO "
+    "state to memory; use session_search to recall those from past transcripts. "
+    "If you've discovered a new way to do something, solved a problem that could be "
+    "necessary later, save it as a skill with the skill tool."
+)
+
+SESSION_SEARCH_GUIDANCE = (
+    "When the user references something from a past conversation or you suspect "
+    "relevant cross-session context exists, use session_search to recall it before "
+    "asking them to repeat themselves."
+)
+
+SKILLS_GUIDANCE = (
+    "After completing a complex task (5+ tool calls), fixing a tricky error, "
+    "or discovering a non-trivial workflow, save the approach as a "
+    "skill with skill_manage so you can reuse it next time.\n"
+    "When using a skill and finding it outdated, incomplete, or wrong, "
+    "patch it immediately with skill_manage(action='patch') — don't wait to be asked. "
+    "Skills that aren't maintained become liabilities."
+)
+
+TOOL_USE_ENFORCEMENT_GUIDANCE = (
+    "# Tool-use enforcement\n"
+    "You MUST use your tools to take action — do not describe what you would do "
+    "or plan to do without actually doing it. When you say you will perform an "
+    "action (e.g. 'I will run the tests', 'Let me check the file', 'I will create "
+    "the project'), you MUST immediately make the corresponding tool call in the same "
+    "response. Never end your turn with a promise of future action — execute it now.\n"
+    "Keep working until the task is actually complete. Do not stop with a summary of "
+    "what you plan to do next time. If you have tools available that can accomplish "
+    "the task, use them instead of telling the user what you would do.\n"
+    "Every response should either (a) contain tool calls that make progress, or "
+    "(b) deliver a final result to the user. Responses that only describe intentions "
+    "without acting are not acceptable."
+)
+
+# Model name substrings that trigger tool-use enforcement guidance.
+# Add new patterns here when a model family needs explicit steering.
+TOOL_USE_ENFORCEMENT_MODELS = ("gpt", "codex")
+
+PLATFORM_HINTS = {
+    "whatsapp": (
+        "You are on a text messaging communication platform, WhatsApp. "
+        "Please do not use markdown as it does not render. "
+        "You can send media files natively: to deliver a file to the user, "
+        "include MEDIA:/absolute/path/to/file in your response. The file "
+        "will be sent as a native WhatsApp attachment — images (.jpg, .png, "
+        ".webp) appear as photos, videos (.mp4, .mov) play inline, and other "
+        "files arrive as downloadable documents. You can also include image "
+        "URLs in markdown format ![alt](url) and they will be sent as photos."
+    ),
+    "telegram": (
+        "You are on a text messaging communication platform, Telegram. "
+        "Please do not use markdown as it does not render. "
+        "You can send media files natively: to deliver a file to the user, "
+        "include MEDIA:/absolute/path/to/file in your response. Images "
+        "(.png, .jpg, .webp) appear as photos, audio (.ogg) sends as voice "
+        "bubbles, and videos (.mp4) play inline. You can also include image "
+        "URLs in markdown format ![alt](url) and they will be sent as native photos."
+    ),
+    "discord": (
+        "You are in a Discord server or group chat communicating with your user. "
+        "You can send media files natively: include MEDIA:/absolute/path/to/file "
+        "in your response. Images (.png, .jpg, .webp) are sent as photo "
+        "attachments, audio as file attachments. You can also include image URLs "
+        "in markdown format ![alt](url) and they will be sent as attachments."
+    ),
+    "slack": (
+        "You are in a Slack workspace communicating with your user. "
+        "You can send media files natively: include MEDIA:/absolute/path/to/file "
+        "in your response. Images (.png, .jpg, .webp) are uploaded as photo "
+        "attachments, audio as file attachments. You can also include image URLs "
+        "in markdown format ![alt](url) and they will be uploaded as attachments."
+    ),
+    "signal": (
+        "You are on a text messaging communication platform, Signal. "
+        "Please do not use markdown as it does not render. "
+        "You can send media files natively: to deliver a file to the user, "
+        "include MEDIA:/absolute/path/to/file in your response. Images "
+        "(.png, .jpg, .webp) appear as photos, audio as attachments, and other "
+        "files arrive as downloadable documents. You can also include image "
+        "URLs in markdown format ![alt](url) and they will be sent as photos."
+    ),
+    "email": (
+        "You are communicating via email. Write clear, well-structured responses "
+        "suitable for email. Use plain text formatting (no markdown). "
+        "Keep responses concise but complete. You can send file attachments — "
+        "include MEDIA:/absolute/path/to/file in your response. The subject line "
+        "is preserved for threading. Do not include greetings or sign-offs unless "
+        "contextually appropriate."
+    ),
+    "cron": (
+        "You are running as a scheduled cron job. There is no user present — you "
+        "cannot ask questions, request clarification, or wait for follow-up. Execute "
+        "the task fully and autonomously, making reasonable decisions where needed. "
+        "Your final response is automatically delivered to the job's configured "
+        "destination — put the primary content directly in your response."
+    ),
+    "cli": (
+        "You are a CLI AI Agent. Try not to use markdown but simple text "
+        "renderable inside a terminal."
+    ),
+    "sms": (
+        "You are communicating via SMS. Keep responses concise and use plain text "
+        "only — no markdown, no formatting. SMS messages are limited to ~1600 "
+        "characters, so be brief and direct."
+    ),
+}
+
+CONTEXT_FILE_MAX_CHARS = 20_000
+CONTEXT_TRUNCATE_HEAD_RATIO = 0.7
+CONTEXT_TRUNCATE_TAIL_RATIO = 0.2
+
+
+# =========================================================================
+# Skills prompt cache
+# =========================================================================
+
+_SKILLS_PROMPT_CACHE_MAX = 8
+_SKILLS_PROMPT_CACHE: OrderedDict[tuple, str] = OrderedDict()
+_SKILLS_PROMPT_CACHE_LOCK = threading.Lock()
+_SKILLS_SNAPSHOT_VERSION = 1
+
+
+def _skills_prompt_snapshot_path() -> Path:
+    return get_hermes_home() / ".skills_prompt_snapshot.json"
+
+
+def clear_skills_system_prompt_cache(*, clear_snapshot: bool = False) -> None:
+    """Drop the in-process skills prompt cache (and optionally the disk snapshot)."""
+    with _SKILLS_PROMPT_CACHE_LOCK:
+        _SKILLS_PROMPT_CACHE.clear()
+    if clear_snapshot:
+        try:
+            _skills_prompt_snapshot_path().unlink(missing_ok=True)
+        except OSError as e:
+            logger.debug("Could not remove skills prompt snapshot: %s", e)
+
+
+def _build_skills_manifest(skills_dir: Path) -> dict[str, list[int]]:
+    """Build an mtime/size manifest of all SKILL.md and DESCRIPTION.md files."""
+    manifest: dict[str, list[int]] = {}
+    for filename in ("SKILL.md", "DESCRIPTION.md"):
+        for path in iter_skill_index_files(skills_dir, filename):
+            try:
+                st = path.stat()
+            except OSError:
+                continue
+            manifest[str(path.relative_to(skills_dir))] = [st.st_mtime_ns, st.st_size]
+    return manifest
+
+
+def _load_skills_snapshot(skills_dir: Path) -> Optional[dict]:
+    """Load the disk snapshot if it exists and its manifest still matches."""
+    snapshot_path = _skills_prompt_snapshot_path()
+    if not snapshot_path.exists():
+        return None
+    try:
+        snapshot = json.loads(snapshot_path.read_text(encoding="utf-8"))
+    except Exception:
+        return None
+    if not isinstance(snapshot, dict):
+        return None
+    if snapshot.get("version") != _SKILLS_SNAPSHOT_VERSION:
+        return None
+    if snapshot.get("manifest") != _build_skills_manifest(skills_dir):
+        return None
+    return snapshot
+
+
+def _write_skills_snapshot(
+    skills_dir: Path,
+    manifest: dict[str, list[int]],
+    skill_entries: list[dict],
+    category_descriptions: dict[str, str],
+) -> None:
+    """Persist skill metadata to disk for fast cold-start reuse."""
+    payload = {
+        "version": _SKILLS_SNAPSHOT_VERSION,
+        "manifest": manifest,
+        "skills": skill_entries,
+        "category_descriptions": category_descriptions,
+    }
+    try:
+        atomic_json_write(_skills_prompt_snapshot_path(), payload)
+    except Exception as e:
+        logger.debug("Could not write skills prompt snapshot: %s", e)
+
+
+def _build_snapshot_entry(
+    skill_file: Path,
+    skills_dir: Path,
+    frontmatter: dict,
+    description: str,
+) -> dict:
+    """Build a serialisable metadata dict for one skill."""
+    rel_path = skill_file.relative_to(skills_dir)
+    parts = rel_path.parts
+    if len(parts) >= 2:
+        skill_name = parts[-2]
+        category = "/".join(parts[:-2]) if len(parts) > 2 else parts[0]
+    else:
+        category = "general"
+        skill_name = skill_file.parent.name
+
+    platforms = frontmatter.get("platforms") or []
+    if isinstance(platforms, str):
+        platforms = [platforms]
+
+    return {
+        "skill_name": skill_name,
+        "category": category,
+        "frontmatter_name": str(frontmatter.get("name", skill_name)),
+        "description": description,
+        "platforms": [str(p).strip() for p in platforms if str(p).strip()],
+        "conditions": extract_skill_conditions(frontmatter),
+    }
+
+
+# =========================================================================
+# Skills index
+# =========================================================================
+
+def _parse_skill_file(skill_file: Path) -> tuple[bool, dict, str]:
+    """Read a SKILL.md once and return platform compatibility, frontmatter, and description.
+
+    Returns (is_compatible, frontmatter, description). On any error, returns
+    (True, {}, "") to err on the side of showing the skill.
+    """
+    try:
+        raw = skill_file.read_text(encoding="utf-8")[:2000]
+        frontmatter, _ = parse_frontmatter(raw)
+
+        if not skill_matches_platform(frontmatter):
+            return False, frontmatter, ""
+
+        return True, frontmatter, extract_skill_description(frontmatter)
+    except Exception as e:
+        logger.debug("Failed to parse skill file %s: %s", skill_file, e)
+        return True, {}, ""
+
+
+def _read_skill_conditions(skill_file: Path) -> dict:
+    """Extract conditional activation fields from SKILL.md frontmatter."""
+    try:
+        raw = skill_file.read_text(encoding="utf-8")[:2000]
+        frontmatter, _ = parse_frontmatter(raw)
+        return extract_skill_conditions(frontmatter)
+    except Exception as e:
+        logger.debug("Failed to read skill conditions from %s: %s", skill_file, e)
+        return {}
+
+
+def _skill_should_show(
+    conditions: dict,
+    available_tools: "set[str] | None",
+    available_toolsets: "set[str] | None",
+) -> bool:
+    """Return False if the skill's conditional activation rules exclude it."""
+    if available_tools is None and available_toolsets is None:
+        return True  # No filtering info — show everything (backward compat)
+
+    at = available_tools or set()
+    ats = available_toolsets or set()
+
+    # fallback_for: hide when the primary tool/toolset IS available
+    for ts in conditions.get("fallback_for_toolsets", []):
+        if ts in ats:
+            return False
+    for t in conditions.get("fallback_for_tools", []):
+        if t in at:
+            return False
+
+    # requires: hide when a required tool/toolset is NOT available
+    for ts in conditions.get("requires_toolsets", []):
+        if ts not in ats:
+            return False
+    for t in conditions.get("requires_tools", []):
+        if t not in at:
+            return False
+
+    return True
+
+
+def build_skills_system_prompt(
+    available_tools: "set[str] | None" = None,
+    available_toolsets: "set[str] | None" = None,
+) -> str:
+    """Build a compact skill index for the system prompt.
+
+    Two-layer cache:
+      1. In-process LRU dict keyed by (skills_dir, tools, toolsets)
+      2. Disk snapshot (``.skills_prompt_snapshot.json``) validated by
+         mtime/size manifest — survives process restarts
+
+    Falls back to a full filesystem scan when both layers miss.
+
+    External skill directories (``skills.external_dirs`` in config.yaml) are
+    scanned alongside the local ``~/.hermes/skills/`` directory.  External dirs
+    are read-only — they appear in the index but new skills are always created
+    in the local dir.  Local skills take precedence when names collide.
+    """
+    hermes_home = get_hermes_home()
+    skills_dir = hermes_home / "skills"
+    external_dirs = get_all_skills_dirs()[1:]  # skip local (index 0)
+
+    if not skills_dir.exists() and not external_dirs:
+        return ""
+
+    # ── Layer 1: in-process LRU cache ─────────────────────────────────
+    cache_key = (
+        str(skills_dir.resolve()),
+        tuple(str(d) for d in external_dirs),
+        tuple(sorted(str(t) for t in (available_tools or set()))),
+        tuple(sorted(str(ts) for ts in (available_toolsets or set()))),
+    )
+    with _SKILLS_PROMPT_CACHE_LOCK:
+        cached = _SKILLS_PROMPT_CACHE.get(cache_key)
+        if cached is not None:
+            _SKILLS_PROMPT_CACHE.move_to_end(cache_key)
+            return cached
+
+    disabled = get_disabled_skill_names()
+
+    # ── Layer 2: disk snapshot ────────────────────────────────────────
+    snapshot = _load_skills_snapshot(skills_dir)
+
+    skills_by_category: dict[str, list[tuple[str, str]]] = {}
+    category_descriptions: dict[str, str] = {}
+
+    if snapshot is not None:
+        # Fast path: use pre-parsed metadata from disk
+        for entry in snapshot.get("skills", []):
+            if not isinstance(entry, dict):
+                continue
+            skill_name = entry.get("skill_name") or ""
+            category = entry.get("category") or "general"
+            frontmatter_name = entry.get("frontmatter_name") or skill_name
+            platforms = entry.get("platforms") or []
+            if not skill_matches_platform({"platforms": platforms}):
+                continue
+            if frontmatter_name in disabled or skill_name in disabled:
+                continue
+            if not _skill_should_show(
+                entry.get("conditions") or {},
+                available_tools,
+                available_toolsets,
+            ):
+                continue
+            skills_by_category.setdefault(category, []).append(
+                (skill_name, entry.get("description", ""))
+            )
+        category_descriptions = {
+            str(k): str(v)
+            for k, v in (snapshot.get("category_descriptions") or {}).items()
+        }
+    else:
+        # Cold path: full filesystem scan + write snapshot for next time
+        skill_entries: list[dict] = []
+        for skill_file in iter_skill_index_files(skills_dir, "SKILL.md"):
+            is_compatible, frontmatter, desc = _parse_skill_file(skill_file)
+            entry = _build_snapshot_entry(skill_file, skills_dir, frontmatter, desc)
+            skill_entries.append(entry)
+            if not is_compatible:
+                continue
+            skill_name = entry["skill_name"]
+            if entry["frontmatter_name"] in disabled or skill_name in disabled:
+                continue
+            if not _skill_should_show(
+                extract_skill_conditions(frontmatter),
+                available_tools,
+                available_toolsets,
+            ):
+                continue
+            skills_by_category.setdefault(entry["category"], []).append(
+                (skill_name, entry["description"])
+            )
+
+        # Read category-level DESCRIPTION.md files
+        for desc_file in iter_skill_index_files(skills_dir, "DESCRIPTION.md"):
+            try:
+                content = desc_file.read_text(encoding="utf-8")
+                fm, _ = parse_frontmatter(content)
+                cat_desc = fm.get("description")
+                if not cat_desc:
+                    continue
+                rel = desc_file.relative_to(skills_dir)
+                cat = "/".join(rel.parts[:-1]) if len(rel.parts) > 1 else "general"
+                category_descriptions[cat] = str(cat_desc).strip().strip("'\"")
+            except Exception as e:
+                logger.debug("Could not read skill description %s: %s", desc_file, e)
+
+        _write_skills_snapshot(
+            skills_dir,
+            _build_skills_manifest(skills_dir),
+            skill_entries,
+            category_descriptions,
+        )
+
+    # ── External skill directories ─────────────────────────────────────
+    # Scan external dirs directly (no snapshot caching — they're read-only
+    # and typically small).  Local skills already in skills_by_category take
+    # precedence: we track seen names and skip duplicates from external dirs.
+    seen_skill_names: set[str] = set()
+    for cat_skills in skills_by_category.values():
+        for name, _desc in cat_skills:
+            seen_skill_names.add(name)
+
+    for ext_dir in external_dirs:
+        if not ext_dir.exists():
+            continue
+        for skill_file in iter_skill_index_files(ext_dir, "SKILL.md"):
+            try:
+                is_compatible, frontmatter, desc = _parse_skill_file(skill_file)
+                if not is_compatible:
+                    continue
+                entry = _build_snapshot_entry(skill_file, ext_dir, frontmatter, desc)
+                skill_name = entry["skill_name"]
+                if skill_name in seen_skill_names:
+                    continue
+                if entry["frontmatter_name"] in disabled or skill_name in disabled:
+                    continue
+                if not _skill_should_show(
+                    extract_skill_conditions(frontmatter),
+                    available_tools,
+                    available_toolsets,
+                ):
+                    continue
+                seen_skill_names.add(skill_name)
+                skills_by_category.setdefault(entry["category"], []).append(
+                    (skill_name, entry["description"])
+                )
+            except Exception as e:
+                logger.debug("Error reading external skill %s: %s", skill_file, e)
+
+        # External category descriptions
+        for desc_file in iter_skill_index_files(ext_dir, "DESCRIPTION.md"):
+            try:
+                content = desc_file.read_text(encoding="utf-8")
+                fm, _ = parse_frontmatter(content)
+                cat_desc = fm.get("description")
+                if not cat_desc:
+                    continue
+                rel = desc_file.relative_to(ext_dir)
+                cat = "/".join(rel.parts[:-1]) if len(rel.parts) > 1 else "general"
+                category_descriptions.setdefault(cat, str(cat_desc).strip().strip("'\""))
+            except Exception as e:
+                logger.debug("Could not read external skill description %s: %s", desc_file, e)
+
+    if not skills_by_category:
+        result = ""
+    else:
+        index_lines = []
+        for category in sorted(skills_by_category.keys()):
+            cat_desc = category_descriptions.get(category, "")
+            if cat_desc:
+                index_lines.append(f"  {category}: {cat_desc}")
+            else:
+                index_lines.append(f"  {category}:")
+            # Deduplicate and sort skills within each category
+            seen = set()
+            for name, desc in sorted(skills_by_category[category], key=lambda x: x[0]):
+                if name in seen:
+                    continue
+                seen.add(name)
+                if desc:
+                    index_lines.append(f"    - {name}: {desc}")
+                else:
+                    index_lines.append(f"    - {name}")
+
+        result = (
+            "## Skills (mandatory)\n"
+            "Before replying, scan the skills below. If one clearly matches your task, "
+            "load it with skill_view(name) and follow its instructions. "
+            "If a skill has issues, fix it with skill_manage(action='patch').\n"
+            "After difficult/iterative tasks, offer to save as a skill. "
+            "If a skill you loaded was missing steps, had wrong commands, or needed "
+            "pitfalls you discovered, update it before finishing.\n"
+            "\n"
+            "<available_skills>\n"
+            + "\n".join(index_lines) + "\n"
+            "</available_skills>\n"
+            "\n"
+            "If none match, proceed normally without loading a skill."
+        )
+
+    # ── Store in LRU cache ────────────────────────────────────────────
+    with _SKILLS_PROMPT_CACHE_LOCK:
+        _SKILLS_PROMPT_CACHE[cache_key] = result
+        _SKILLS_PROMPT_CACHE.move_to_end(cache_key)
+        while len(_SKILLS_PROMPT_CACHE) > _SKILLS_PROMPT_CACHE_MAX:
+            _SKILLS_PROMPT_CACHE.popitem(last=False)
+
+    return result
+
+
+# =========================================================================
+# Context files (SOUL.md, AGENTS.md, .cursorrules)
+# =========================================================================
+
+def _truncate_content(content: str, filename: str, max_chars: int = CONTEXT_FILE_MAX_CHARS) -> str:
+    """Head/tail truncation with a marker in the middle."""
+    if len(content) <= max_chars:
+        return content
+    head_chars = int(max_chars * CONTEXT_TRUNCATE_HEAD_RATIO)
+    tail_chars = int(max_chars * CONTEXT_TRUNCATE_TAIL_RATIO)
+    head = content[:head_chars]
+    tail = content[-tail_chars:]
+    marker = f"\n\n[...truncated {filename}: kept {head_chars}+{tail_chars} of {len(content)} chars. Use file tools to read the full file.]\n\n"
+    return head + marker + tail
+
+
+def load_soul_md() -> Optional[str]:
+    """Load SOUL.md from HERMES_HOME and return its content, or None.
+
+    Used as the agent identity (slot #1 in the system prompt).  When this
+    returns content, ``build_context_files_prompt`` should be called with
+    ``skip_soul=True`` so SOUL.md isn't injected twice.
+    """
+    try:
+        from hermes_cli.config import ensure_hermes_home
+        ensure_hermes_home()
+    except Exception as e:
+        logger.debug("Could not ensure HERMES_HOME before loading SOUL.md: %s", e)
+
+    soul_path = get_hermes_home() / "SOUL.md"
+    if not soul_path.exists():
+        return None
+    try:
+        content = soul_path.read_text(encoding="utf-8").strip()
+        if not content:
+            return None
+        content = _scan_context_content(content, "SOUL.md")
+        content = _truncate_content(content, "SOUL.md")
+        return content
+    except Exception as e:
+        logger.debug("Could not read SOUL.md from %s: %s", soul_path, e)
+        return None
+
+
+def _load_hermes_md(cwd_path: Path) -> str:
+    """.hermes.md / HERMES.md — walk to git root."""
+    hermes_md_path = _find_hermes_md(cwd_path)
+    if not hermes_md_path:
+        return ""
+    try:
+        content = hermes_md_path.read_text(encoding="utf-8").strip()
+        if not content:
+            return ""
+        content = _strip_yaml_frontmatter(content)
+        rel = hermes_md_path.name
+        try:
+            rel = str(hermes_md_path.relative_to(cwd_path))
+        except ValueError:
+            pass
+        content = _scan_context_content(content, rel)
+        result = f"## {rel}\n\n{content}"
+        return _truncate_content(result, ".hermes.md")
+    except Exception as e:
+        logger.debug("Could not read %s: %s", hermes_md_path, e)
+        return ""
+
+
+def _load_agents_md(cwd_path: Path) -> str:
+    """AGENTS.md — top-level only (no recursive walk)."""
+    for name in ["AGENTS.md", "agents.md"]:
+        candidate = cwd_path / name
+        if candidate.exists():
+            try:
+                content = candidate.read_text(encoding="utf-8").strip()
+                if content:
+                    content = _scan_context_content(content, name)
+                    result = f"## {name}\n\n{content}"
+                    return _truncate_content(result, "AGENTS.md")
+            except Exception as e:
+                logger.debug("Could not read %s: %s", candidate, e)
+    return ""
+
+
+def _load_claude_md(cwd_path: Path) -> str:
+    """CLAUDE.md / claude.md — cwd only."""
+    for name in ["CLAUDE.md", "claude.md"]:
+        candidate = cwd_path / name
+        if candidate.exists():
+            try:
+                content = candidate.read_text(encoding="utf-8").strip()
+                if content:
+                    content = _scan_context_content(content, name)
+                    result = f"## {name}\n\n{content}"
+                    return _truncate_content(result, "CLAUDE.md")
+            except Exception as e:
+                logger.debug("Could not read %s: %s", candidate, e)
+    return ""
+
+
+def _load_cursorrules(cwd_path: Path) -> str:
+    """.cursorrules + .cursor/rules/*.mdc — cwd only."""
+    cursorrules_content = ""
+    cursorrules_file = cwd_path / ".cursorrules"
+    if cursorrules_file.exists():
+        try:
+            content = cursorrules_file.read_text(encoding="utf-8").strip()
+            if content:
+                content = _scan_context_content(content, ".cursorrules")
+                cursorrules_content += f"## .cursorrules\n\n{content}\n\n"
+        except Exception as e:
+            logger.debug("Could not read .cursorrules: %s", e)
+
+    cursor_rules_dir = cwd_path / ".cursor" / "rules"
+    if cursor_rules_dir.exists() and cursor_rules_dir.is_dir():
+        mdc_files = sorted(cursor_rules_dir.glob("*.mdc"))
+        for mdc_file in mdc_files:
+            try:
+                content = mdc_file.read_text(encoding="utf-8").strip()
+                if content:
+                    content = _scan_context_content(content, f".cursor/rules/{mdc_file.name}")
+                    cursorrules_content += f"## .cursor/rules/{mdc_file.name}\n\n{content}\n\n"
+            except Exception as e:
+                logger.debug("Could not read %s: %s", mdc_file, e)
+
+    if not cursorrules_content:
+        return ""
+    return _truncate_content(cursorrules_content, ".cursorrules")
+
+
+def build_context_files_prompt(cwd: Optional[str] = None, skip_soul: bool = False) -> str:
+    """Discover and load context files for the system prompt.
+
+    Priority (first found wins — only ONE project context type is loaded):
+      1. .hermes.md / HERMES.md  (walk to git root)
+      2. AGENTS.md / agents.md   (cwd only)
+      3. CLAUDE.md / claude.md   (cwd only)
+      4. .cursorrules / .cursor/rules/*.mdc  (cwd only)
+
+    SOUL.md from HERMES_HOME is independent and always included when present.
+    Each context source is capped at 20,000 chars.
+
+    When *skip_soul* is True, SOUL.md is not included here (it was already
+    loaded via ``load_soul_md()`` for the identity slot).
+    """
+    if cwd is None:
+        cwd = os.getcwd()
+
+    cwd_path = Path(cwd).resolve()
+    sections = []
+
+    # Priority-based project context: first match wins
+    project_context = (
+        _load_hermes_md(cwd_path)
+        or _load_agents_md(cwd_path)
+        or _load_claude_md(cwd_path)
+        or _load_cursorrules(cwd_path)
+    )
+    if project_context:
+        sections.append(project_context)
+
+    # SOUL.md from HERMES_HOME only — skip when already loaded as identity
+    if not skip_soul:
+        soul_content = load_soul_md()
+        if soul_content:
+            sections.append(soul_content)
+
+    if not sections:
+        return ""
+    return "# Project Context\n\nThe following project context files have been loaded and should be followed:\n\n" + "\n".join(sections)

agent/prompt_caching.py

@@ -0,0 +1,72 @@
+"""Anthropic prompt caching (system_and_3 strategy).
+
+Reduces input token costs by ~75% on multi-turn conversations by caching
+the conversation prefix. Uses 4 cache_control breakpoints (Anthropic max):
+  1. System prompt (stable across all turns)
+  2-4. Last 3 non-system messages (rolling window)
+
+Pure functions -- no class state, no AIAgent dependency.
+"""
+
+import copy
+from typing import Any, Dict, List
+
+
+def _apply_cache_marker(msg: dict, cache_marker: dict, native_anthropic: bool = False) -> None:
+    """Add cache_control to a single message, handling all format variations."""
+    role = msg.get("role", "")
+    content = msg.get("content")
+
+    if role == "tool":
+        if native_anthropic:
+            msg["cache_control"] = cache_marker
+        return
+
+    if content is None or content == "":
+        msg["cache_control"] = cache_marker
+        return
+
+    if isinstance(content, str):
+        msg["content"] = [
+            {"type": "text", "text": content, "cache_control": cache_marker}
+        ]
+        return
+
+    if isinstance(content, list) and content:
+        last = content[-1]
+        if isinstance(last, dict):
+            last["cache_control"] = cache_marker
+
+
+def apply_anthropic_cache_control(
+    api_messages: List[Dict[str, Any]],
+    cache_ttl: str = "5m",
+    native_anthropic: bool = False,
+) -> List[Dict[str, Any]]:
+    """Apply system_and_3 caching strategy to messages for Anthropic models.
+
+    Places up to 4 cache_control breakpoints: system prompt + last 3 non-system messages.
+
+    Returns:
+        Deep copy of messages with cache_control breakpoints injected.
+    """
+    messages = copy.deepcopy(api_messages)
+    if not messages:
+        return messages
+
+    marker = {"type": "ephemeral"}
+    if cache_ttl == "1h":
+        marker["ttl"] = "1h"
+
+    breakpoints_used = 0
+
+    if messages[0].get("role") == "system":
+        _apply_cache_marker(messages[0], marker, native_anthropic=native_anthropic)
+        breakpoints_used += 1
+
+    remaining = 4 - breakpoints_used
+    non_sys = [i for i in range(len(messages)) if messages[i].get("role") != "system"]
+    for idx in non_sys[-remaining:]:
+        _apply_cache_marker(messages[idx], marker, native_anthropic=native_anthropic)
+
+    return messages

agent/redact.py

@@ -0,0 +1,176 @@
+"""Regex-based secret redaction for logs and tool output.
+
+Applies pattern matching to mask API keys, tokens, and credentials
+before they reach log files, verbose output, or gateway logs.
+
+Short tokens (< 18 chars) are fully masked. Longer tokens preserve
+the first 6 and last 4 characters for debuggability.
+"""
+
+import logging
+import os
+import re
+
+logger = logging.getLogger(__name__)
+
+# Snapshot at import time so runtime env mutations (e.g. LLM-generated
+# `export HERMES_REDACT_SECRETS=false`) cannot disable redaction mid-session.
+_REDACT_ENABLED = os.getenv("HERMES_REDACT_SECRETS", "").lower() not in ("0", "false", "no", "off")
+
+# Known API key prefixes -- match the prefix + contiguous token chars
+_PREFIX_PATTERNS = [
+    r"sk-[A-Za-z0-9_-]{10,}",           # OpenAI / OpenRouter / Anthropic (sk-ant-*)
+    r"ghp_[A-Za-z0-9]{10,}",            # GitHub PAT (classic)
+    r"github_pat_[A-Za-z0-9_]{10,}",    # GitHub PAT (fine-grained)
+    r"gho_[A-Za-z0-9]{10,}",            # GitHub OAuth access token
+    r"ghu_[A-Za-z0-9]{10,}",            # GitHub user-to-server token
+    r"ghs_[A-Za-z0-9]{10,}",            # GitHub server-to-server token
+    r"ghr_[A-Za-z0-9]{10,}",            # GitHub refresh token
+    r"xox[baprs]-[A-Za-z0-9-]{10,}",    # Slack tokens
+    r"AIza[A-Za-z0-9_-]{30,}",          # Google API keys
+    r"pplx-[A-Za-z0-9]{10,}",           # Perplexity
+    r"fal_[A-Za-z0-9_-]{10,}",          # Fal.ai
+    r"fc-[A-Za-z0-9]{10,}",             # Firecrawl
+    r"bb_live_[A-Za-z0-9_-]{10,}",      # BrowserBase
+    r"gAAAA[A-Za-z0-9_=-]{20,}",        # Codex encrypted tokens
+    r"AKIA[A-Z0-9]{16}",                # AWS Access Key ID
+    r"sk_live_[A-Za-z0-9]{10,}",        # Stripe secret key (live)
+    r"sk_test_[A-Za-z0-9]{10,}",        # Stripe secret key (test)
+    r"rk_live_[A-Za-z0-9]{10,}",        # Stripe restricted key
+    r"SG\.[A-Za-z0-9_-]{10,}",          # SendGrid API key
+    r"hf_[A-Za-z0-9]{10,}",             # HuggingFace token
+    r"r8_[A-Za-z0-9]{10,}",             # Replicate API token
+    r"npm_[A-Za-z0-9]{10,}",            # npm access token
+    r"pypi-[A-Za-z0-9_-]{10,}",         # PyPI API token
+    r"dop_v1_[A-Za-z0-9]{10,}",         # DigitalOcean PAT
+    r"doo_v1_[A-Za-z0-9]{10,}",         # DigitalOcean OAuth
+    r"am_[A-Za-z0-9_-]{10,}",           # AgentMail API key
+    r"sk_[A-Za-z0-9_]{10,}",            # ElevenLabs TTS key (sk_ underscore, not sk- dash)
+    r"tvly-[A-Za-z0-9]{10,}",           # Tavily search API key
+    r"exa_[A-Za-z0-9]{10,}",            # Exa search API key
+]
+
+# ENV assignment patterns: KEY=value where KEY contains a secret-like name
+_SECRET_ENV_NAMES = r"(?:API_?KEY|TOKEN|SECRET|PASSWORD|PASSWD|CREDENTIAL|AUTH)"
+_ENV_ASSIGN_RE = re.compile(
+    rf"([A-Z_]*{_SECRET_ENV_NAMES}[A-Z_]*)\s*=\s*(['\"]?)(\S+)\2",
+    re.IGNORECASE,
+)
+
+# JSON field patterns: "apiKey": "value", "token": "value", etc.
+_JSON_KEY_NAMES = r"(?:api_?[Kk]ey|token|secret|password|access_token|refresh_token|auth_token|bearer|secret_value|raw_secret|secret_input|key_material)"
+_JSON_FIELD_RE = re.compile(
+    rf'("{_JSON_KEY_NAMES}")\s*:\s*"([^"]+)"',
+    re.IGNORECASE,
+)
+
+# Authorization headers
+_AUTH_HEADER_RE = re.compile(
+    r"(Authorization:\s*Bearer\s+)(\S+)",
+    re.IGNORECASE,
+)
+
+# Telegram bot tokens: bot<digits>:<token> or <digits>:<token>,
+# where token part is restricted to [-A-Za-z0-9_] and length >= 30
+_TELEGRAM_RE = re.compile(
+    r"(bot)?(\d{8,}):([-A-Za-z0-9_]{30,})",
+)
+
+# Private key blocks: -----BEGIN RSA PRIVATE KEY----- ... -----END RSA PRIVATE KEY-----
+_PRIVATE_KEY_RE = re.compile(
+    r"-----BEGIN[A-Z ]*PRIVATE KEY-----[\s\S]*?-----END[A-Z ]*PRIVATE KEY-----"
+)
+
+# Database connection strings: protocol://user:PASSWORD@host
+# Catches postgres, mysql, mongodb, redis, amqp URLs and redacts the password
+_DB_CONNSTR_RE = re.compile(
+    r"((?:postgres(?:ql)?|mysql|mongodb(?:\+srv)?|redis|amqp)://[^:]+:)([^@]+)(@)",
+    re.IGNORECASE,
+)
+
+# E.164 phone numbers: +<country><number>, 7-15 digits
+# Negative lookahead prevents matching hex strings or identifiers
+_SIGNAL_PHONE_RE = re.compile(r"(\+[1-9]\d{6,14})(?![A-Za-z0-9])")
+
+# Compile known prefix patterns into one alternation
+_PREFIX_RE = re.compile(
+    r"(?<![A-Za-z0-9_-])(" + "|".join(_PREFIX_PATTERNS) + r")(?![A-Za-z0-9_-])"
+)
+
+
+def _mask_token(token: str) -> str:
+    """Mask a token, preserving prefix for long tokens."""
+    if len(token) < 18:
+        return "***"
+    return f"{token[:6]}...{token[-4:]}"
+
+
+def redact_sensitive_text(text: str) -> str:
+    """Apply all redaction patterns to a block of text.
+
+    Safe to call on any string -- non-matching text passes through unchanged.
+    Disabled when security.redact_secrets is false in config.yaml.
+    """
+    if text is None:
+        return None
+    if not isinstance(text, str):
+        text = str(text)
+    if not text:
+        return text
+    if not _REDACT_ENABLED:
+        return text
+
+    # Known prefixes (sk-, ghp_, etc.)
+    text = _PREFIX_RE.sub(lambda m: _mask_token(m.group(1)), text)
+
+    # ENV assignments: OPENAI_API_KEY=sk-abc...
+    def _redact_env(m):
+        name, quote, value = m.group(1), m.group(2), m.group(3)
+        return f"{name}={quote}{_mask_token(value)}{quote}"
+    text = _ENV_ASSIGN_RE.sub(_redact_env, text)
+
+    # JSON fields: "apiKey": "value"
+    def _redact_json(m):
+        key, value = m.group(1), m.group(2)
+        return f'{key}: "{_mask_token(value)}"'
+    text = _JSON_FIELD_RE.sub(_redact_json, text)
+
+    # Authorization headers
+    text = _AUTH_HEADER_RE.sub(
+        lambda m: m.group(1) + _mask_token(m.group(2)),
+        text,
+    )
+
+    # Telegram bot tokens
+    def _redact_telegram(m):
+        prefix = m.group(1) or ""
+        digits = m.group(2)
+        return f"{prefix}{digits}:***"
+    text = _TELEGRAM_RE.sub(_redact_telegram, text)
+
+    # Private key blocks
+    text = _PRIVATE_KEY_RE.sub("[REDACTED PRIVATE KEY]", text)
+
+    # Database connection string passwords
+    text = _DB_CONNSTR_RE.sub(lambda m: f"{m.group(1)}***{m.group(3)}", text)
+
+    # E.164 phone numbers (Signal, WhatsApp)
+    def _redact_phone(m):
+        phone = m.group(1)
+        if len(phone) <= 8:
+            return phone[:2] + "****" + phone[-2:]
+        return phone[:4] + "****" + phone[-4:]
+    text = _SIGNAL_PHONE_RE.sub(_redact_phone, text)
+
+    return text
+
+
+class RedactingFormatter(logging.Formatter):
+    """Log formatter that redacts secrets from all log messages."""
+
+    def __init__(self, fmt=None, datefmt=None, style='%', **kwargs):
+        super().__init__(fmt, datefmt, style, **kwargs)
+
+    def format(self, record: logging.LogRecord) -> str:
+        original = super().format(record)
+        return redact_sensitive_text(original)

agent/skill_commands.py

@@ -0,0 +1,297 @@
+"""Shared slash command helpers for skills and built-in prompt-style modes.
+
+Shared between CLI (cli.py) and gateway (gateway/run.py) so both surfaces
+can invoke skills via /skill-name commands and prompt-only built-ins like
+/plan.
+"""
+
+import json
+import logging
+import re
+from datetime import datetime
+from pathlib import Path
+from typing import Any, Dict, Optional
+
+logger = logging.getLogger(__name__)
+
+_skill_commands: Dict[str, Dict[str, Any]] = {}
+_PLAN_SLUG_RE = re.compile(r"[^a-z0-9]+")
+
+
+def build_plan_path(
+    user_instruction: str = "",
+    *,
+    now: datetime | None = None,
+) -> Path:
+    """Return the default workspace-relative markdown path for a /plan invocation.
+
+    Relative paths are intentional: file tools are task/backend-aware and resolve
+    them against the active working directory for local, docker, ssh, modal,
+    daytona, and similar terminal backends. That keeps the plan with the active
+    workspace instead of the Hermes host's global home directory.
+    """
+    slug_source = (user_instruction or "").strip().splitlines()[0] if user_instruction else ""
+    slug = _PLAN_SLUG_RE.sub("-", slug_source.lower()).strip("-")
+    if slug:
+        slug = "-".join(part for part in slug.split("-")[:8] if part)[:48].strip("-")
+    slug = slug or "conversation-plan"
+    timestamp = (now or datetime.now()).strftime("%Y-%m-%d_%H%M%S")
+    return Path(".hermes") / "plans" / f"{timestamp}-{slug}.md"
+
+
+def _load_skill_payload(skill_identifier: str, task_id: str | None = None) -> tuple[dict[str, Any], Path | None, str] | None:
+    """Load a skill by name/path and return (loaded_payload, skill_dir, display_name)."""
+    raw_identifier = (skill_identifier or "").strip()
+    if not raw_identifier:
+        return None
+
+    try:
+        from tools.skills_tool import SKILLS_DIR, skill_view
+
+        identifier_path = Path(raw_identifier).expanduser()
+        if identifier_path.is_absolute():
+            try:
+                normalized = str(identifier_path.resolve().relative_to(SKILLS_DIR.resolve()))
+            except Exception:
+                normalized = raw_identifier
+        else:
+            normalized = raw_identifier.lstrip("/")
+
+        loaded_skill = json.loads(skill_view(normalized, task_id=task_id))
+    except Exception:
+        return None
+
+    if not loaded_skill.get("success"):
+        return None
+
+    skill_name = str(loaded_skill.get("name") or normalized)
+    skill_path = str(loaded_skill.get("path") or "")
+    skill_dir = None
+    if skill_path:
+        try:
+            skill_dir = SKILLS_DIR / Path(skill_path).parent
+        except Exception:
+            skill_dir = None
+
+    return loaded_skill, skill_dir, skill_name
+
+
+def _build_skill_message(
+    loaded_skill: dict[str, Any],
+    skill_dir: Path | None,
+    activation_note: str,
+    user_instruction: str = "",
+    runtime_note: str = "",
+) -> str:
+    """Format a loaded skill into a user/system message payload."""
+    from tools.skills_tool import SKILLS_DIR
+
+    content = str(loaded_skill.get("content") or "")
+
+    parts = [activation_note, "", content.strip()]
+
+    if loaded_skill.get("setup_skipped"):
+        parts.extend(
+            [
+                "",
+                "[Skill setup note: Required environment setup was skipped. Continue loading the skill and explain any reduced functionality if it matters.]",
+            ]
+        )
+    elif loaded_skill.get("gateway_setup_hint"):
+        parts.extend(
+            [
+                "",
+                f"[Skill setup note: {loaded_skill['gateway_setup_hint']}]",
+            ]
+        )
+    elif loaded_skill.get("setup_needed") and loaded_skill.get("setup_note"):
+        parts.extend(
+            [
+                "",
+                f"[Skill setup note: {loaded_skill['setup_note']}]",
+            ]
+        )
+
+    supporting = []
+    linked_files = loaded_skill.get("linked_files") or {}
+    for entries in linked_files.values():
+        if isinstance(entries, list):
+            supporting.extend(entries)
+
+    if not supporting and skill_dir:
+        for subdir in ("references", "templates", "scripts", "assets"):
+            subdir_path = skill_dir / subdir
+            if subdir_path.exists():
+                for f in sorted(subdir_path.rglob("*")):
+                    if f.is_file():
+                        rel = str(f.relative_to(skill_dir))
+                        supporting.append(rel)
+
+    if supporting and skill_dir:
+        try:
+            skill_view_target = str(skill_dir.relative_to(SKILLS_DIR))
+        except ValueError:
+            # Skill is from an external dir — use the skill name instead
+            skill_view_target = skill_dir.name
+        parts.append("")
+        parts.append("[This skill has supporting files you can load with the skill_view tool:]")
+        for sf in supporting:
+            parts.append(f"- {sf}")
+        parts.append(
+            f'\nTo view any of these, use: skill_view(name="{skill_view_target}", file_path="<path>")'
+        )
+
+    if user_instruction:
+        parts.append("")
+        parts.append(f"The user has provided the following instruction alongside the skill invocation: {user_instruction}")
+
+    if runtime_note:
+        parts.append("")
+        parts.append(f"[Runtime note: {runtime_note}]")
+
+    return "\n".join(parts)
+
+
+def scan_skill_commands() -> Dict[str, Dict[str, Any]]:
+    """Scan ~/.hermes/skills/ and return a mapping of /command -> skill info.
+
+    Returns:
+        Dict mapping "/skill-name" to {name, description, skill_md_path, skill_dir}.
+    """
+    global _skill_commands
+    _skill_commands = {}
+    try:
+        from tools.skills_tool import SKILLS_DIR, _parse_frontmatter, skill_matches_platform, _get_disabled_skill_names
+        from agent.skill_utils import get_external_skills_dirs
+        disabled = _get_disabled_skill_names()
+        seen_names: set = set()
+
+        # Scan local dir first, then external dirs
+        dirs_to_scan = []
+        if SKILLS_DIR.exists():
+            dirs_to_scan.append(SKILLS_DIR)
+        dirs_to_scan.extend(get_external_skills_dirs())
+
+        for scan_dir in dirs_to_scan:
+            for skill_md in scan_dir.rglob("SKILL.md"):
+                if any(part in ('.git', '.github', '.hub') for part in skill_md.parts):
+                    continue
+                try:
+                    content = skill_md.read_text(encoding='utf-8')
+                    frontmatter, body = _parse_frontmatter(content)
+                    # Skip skills incompatible with the current OS platform
+                    if not skill_matches_platform(frontmatter):
+                        continue
+                    name = frontmatter.get('name', skill_md.parent.name)
+                    if name in seen_names:
+                        continue
+                    # Respect user's disabled skills config
+                    if name in disabled:
+                        continue
+                    description = frontmatter.get('description', '')
+                    if not description:
+                        for line in body.strip().split('\n'):
+                            line = line.strip()
+                            if line and not line.startswith('#'):
+                                description = line[:80]
+                                break
+                    seen_names.add(name)
+                    cmd_name = name.lower().replace(' ', '-').replace('_', '-')
+                    _skill_commands[f"/{cmd_name}"] = {
+                        "name": name,
+                        "description": description or f"Invoke the {name} skill",
+                        "skill_md_path": str(skill_md),
+                        "skill_dir": str(skill_md.parent),
+                    }
+                except Exception:
+                    continue
+    except Exception:
+        pass
+    return _skill_commands
+
+
+def get_skill_commands() -> Dict[str, Dict[str, Any]]:
+    """Return the current skill commands mapping (scan first if empty)."""
+    if not _skill_commands:
+        scan_skill_commands()
+    return _skill_commands
+
+
+def build_skill_invocation_message(
+    cmd_key: str,
+    user_instruction: str = "",
+    task_id: str | None = None,
+    runtime_note: str = "",
+) -> Optional[str]:
+    """Build the user message content for a skill slash command invocation.
+
+    Args:
+        cmd_key: The command key including leading slash (e.g., "/gif-search").
+        user_instruction: Optional text the user typed after the command.
+
+    Returns:
+        The formatted message string, or None if the skill wasn't found.
+    """
+    commands = get_skill_commands()
+    skill_info = commands.get(cmd_key)
+    if not skill_info:
+        return None
+
+    loaded = _load_skill_payload(skill_info["skill_dir"], task_id=task_id)
+    if not loaded:
+        return f"[Failed to load skill: {skill_info['name']}]"
+
+    loaded_skill, skill_dir, skill_name = loaded
+    activation_note = (
+        f'[SYSTEM: The user has invoked the "{skill_name}" skill, indicating they want '
+        "you to follow its instructions. The full skill content is loaded below.]"
+    )
+    return _build_skill_message(
+        loaded_skill,
+        skill_dir,
+        activation_note,
+        user_instruction=user_instruction,
+        runtime_note=runtime_note,
+    )
+
+
+def build_preloaded_skills_prompt(
+    skill_identifiers: list[str],
+    task_id: str | None = None,
+) -> tuple[str, list[str], list[str]]:
+    """Load one or more skills for session-wide CLI preloading.
+
+    Returns (prompt_text, loaded_skill_names, missing_identifiers).
+    """
+    prompt_parts: list[str] = []
+    loaded_names: list[str] = []
+    missing: list[str] = []
+
+    seen: set[str] = set()
+    for raw_identifier in skill_identifiers:
+        identifier = (raw_identifier or "").strip()
+        if not identifier or identifier in seen:
+            continue
+        seen.add(identifier)
+
+        loaded = _load_skill_payload(identifier, task_id=task_id)
+        if not loaded:
+            missing.append(identifier)
+            continue
+
+        loaded_skill, skill_dir, skill_name = loaded
+        activation_note = (
+            f'[SYSTEM: The user launched this CLI session with the "{skill_name}" skill '
+            "preloaded. Treat its instructions as active guidance for the duration of this "
+            "session unless the user overrides them.]"
+        )
+        prompt_parts.append(
+            _build_skill_message(
+                loaded_skill,
+                skill_dir,
+                activation_note,
+            )
+        )
+        loaded_names.append(skill_name)
+
+    return "\n\n".join(prompt_parts), loaded_names, missing

agent/skill_utils.py

@@ -0,0 +1,270 @@
+"""Lightweight skill metadata utilities shared by prompt_builder and skills_tool.
+
+This module intentionally avoids importing the tool registry, CLI config, or any
+heavy dependency chain.  It is safe to import at module level without triggering
+tool registration or provider resolution.
+"""
+
+import logging
+import os
+import re
+import sys
+from pathlib import Path
+from typing import Any, Dict, List, Optional, Set, Tuple
+
+from hermes_constants import get_hermes_home
+
+logger = logging.getLogger(__name__)
+
+# ── Platform mapping ──────────────────────────────────────────────────────
+
+PLATFORM_MAP = {
+    "macos": "darwin",
+    "linux": "linux",
+    "windows": "win32",
+}
+
+EXCLUDED_SKILL_DIRS = frozenset((".git", ".github", ".hub"))
+
+# ── Lazy YAML loader ─────────────────────────────────────────────────────
+
+_yaml_load_fn = None
+
+
+def yaml_load(content: str):
+    """Parse YAML with lazy import and CSafeLoader preference."""
+    global _yaml_load_fn
+    if _yaml_load_fn is None:
+        import yaml
+
+        loader = getattr(yaml, "CSafeLoader", None) or yaml.SafeLoader
+
+        def _load(value: str):
+            return yaml.load(value, Loader=loader)
+
+        _yaml_load_fn = _load
+    return _yaml_load_fn(content)
+
+
+# ── Frontmatter parsing ──────────────────────────────────────────────────
+
+
+def parse_frontmatter(content: str) -> Tuple[Dict[str, Any], str]:
+    """Parse YAML frontmatter from a markdown string.
+
+    Uses yaml with CSafeLoader for full YAML support (nested metadata, lists)
+    with a fallback to simple key:value splitting for robustness.
+
+    Returns:
+        (frontmatter_dict, remaining_body)
+    """
+    frontmatter: Dict[str, Any] = {}
+    body = content
+
+    if not content.startswith("---"):
+        return frontmatter, body
+
+    end_match = re.search(r"\n---\s*\n", content[3:])
+    if not end_match:
+        return frontmatter, body
+
+    yaml_content = content[3 : end_match.start() + 3]
+    body = content[end_match.end() + 3 :]
+
+    try:
+        parsed = yaml_load(yaml_content)
+        if isinstance(parsed, dict):
+            frontmatter = parsed
+    except Exception:
+        # Fallback: simple key:value parsing for malformed YAML
+        for line in yaml_content.strip().split("\n"):
+            if ":" not in line:
+                continue
+            key, value = line.split(":", 1)
+            frontmatter[key.strip()] = value.strip()
+
+    return frontmatter, body
+
+
+# ── Platform matching ─────────────────────────────────────────────────────
+
+
+def skill_matches_platform(frontmatter: Dict[str, Any]) -> bool:
+    """Return True when the skill is compatible with the current OS.
+
+    Skills declare platform requirements via a top-level ``platforms`` list
+    in their YAML frontmatter::
+
+        platforms: [macos]          # macOS only
+        platforms: [macos, linux]   # macOS and Linux
+
+    If the field is absent or empty the skill is compatible with **all**
+    platforms (backward-compatible default).
+    """
+    platforms = frontmatter.get("platforms")
+    if not platforms:
+        return True
+    if not isinstance(platforms, list):
+        platforms = [platforms]
+    current = sys.platform
+    for platform in platforms:
+        normalized = str(platform).lower().strip()
+        mapped = PLATFORM_MAP.get(normalized, normalized)
+        if current.startswith(mapped):
+            return True
+    return False
+
+
+# ── Disabled skills ───────────────────────────────────────────────────────
+
+
+def get_disabled_skill_names() -> Set[str]:
+    """Read disabled skill names from config.yaml.
+
+    Resolves platform from ``HERMES_PLATFORM`` env var, falls back to
+    the global disabled list.  Reads the config file directly (no CLI
+    config imports) to stay lightweight.
+    """
+    config_path = get_hermes_home() / "config.yaml"
+    if not config_path.exists():
+        return set()
+    try:
+        parsed = yaml_load(config_path.read_text(encoding="utf-8"))
+    except Exception as e:
+        logger.debug("Could not read skill config %s: %s", config_path, e)
+        return set()
+    if not isinstance(parsed, dict):
+        return set()
+
+    skills_cfg = parsed.get("skills")
+    if not isinstance(skills_cfg, dict):
+        return set()
+
+    resolved_platform = os.getenv("HERMES_PLATFORM")
+    if resolved_platform:
+        platform_disabled = (skills_cfg.get("platform_disabled") or {}).get(
+            resolved_platform
+        )
+        if platform_disabled is not None:
+            return _normalize_string_set(platform_disabled)
+    return _normalize_string_set(skills_cfg.get("disabled"))
+
+
+def _normalize_string_set(values) -> Set[str]:
+    if values is None:
+        return set()
+    if isinstance(values, str):
+        values = [values]
+    return {str(v).strip() for v in values if str(v).strip()}
+
+
+# ── External skills directories ──────────────────────────────────────────
+
+
+def get_external_skills_dirs() -> List[Path]:
+    """Read ``skills.external_dirs`` from config.yaml and return validated paths.
+
+    Each entry is expanded (``~`` and ``${VAR}``) and resolved to an absolute
+    path.  Only directories that actually exist are returned.  Duplicates and
+    paths that resolve to the local ``~/.hermes/skills/`` are silently skipped.
+    """
+    config_path = get_hermes_home() / "config.yaml"
+    if not config_path.exists():
+        return []
+    try:
+        parsed = yaml_load(config_path.read_text(encoding="utf-8"))
+    except Exception:
+        return []
+    if not isinstance(parsed, dict):
+        return []
+
+    skills_cfg = parsed.get("skills")
+    if not isinstance(skills_cfg, dict):
+        return []
+
+    raw_dirs = skills_cfg.get("external_dirs")
+    if not raw_dirs:
+        return []
+    if isinstance(raw_dirs, str):
+        raw_dirs = [raw_dirs]
+    if not isinstance(raw_dirs, list):
+        return []
+
+    local_skills = (get_hermes_home() / "skills").resolve()
+    seen: Set[Path] = set()
+    result: List[Path] = []
+
+    for entry in raw_dirs:
+        entry = str(entry).strip()
+        if not entry:
+            continue
+        # Expand ~ and environment variables
+        expanded = os.path.expanduser(os.path.expandvars(entry))
+        p = Path(expanded).resolve()
+        if p == local_skills:
+            continue
+        if p in seen:
+            continue
+        if p.is_dir():
+            seen.add(p)
+            result.append(p)
+        else:
+            logger.debug("External skills dir does not exist, skipping: %s", p)
+
+    return result
+
+
+def get_all_skills_dirs() -> List[Path]:
+    """Return all skill directories: local ``~/.hermes/skills/`` first, then external.
+
+    The local dir is always first (and always included even if it doesn't exist
+    yet — callers handle that).  External dirs follow in config order.
+    """
+    dirs = [get_hermes_home() / "skills"]
+    dirs.extend(get_external_skills_dirs())
+    return dirs
+
+
+# ── Condition extraction ──────────────────────────────────────────────────
+
+
+def extract_skill_conditions(frontmatter: Dict[str, Any]) -> Dict[str, List]:
+    """Extract conditional activation fields from parsed frontmatter."""
+    hermes = (frontmatter.get("metadata") or {}).get("hermes") or {}
+    return {
+        "fallback_for_toolsets": hermes.get("fallback_for_toolsets", []),
+        "requires_toolsets": hermes.get("requires_toolsets", []),
+        "fallback_for_tools": hermes.get("fallback_for_tools", []),
+        "requires_tools": hermes.get("requires_tools", []),
+    }
+
+
+# ── Description extraction ────────────────────────────────────────────────
+
+
+def extract_skill_description(frontmatter: Dict[str, Any]) -> str:
+    """Extract a truncated description from parsed frontmatter."""
+    raw_desc = frontmatter.get("description", "")
+    if not raw_desc:
+        return ""
+    desc = str(raw_desc).strip().strip("'\"")
+    if len(desc) > 60:
+        return desc[:57] + "..."
+    return desc
+
+
+# ── File iteration ────────────────────────────────────────────────────────
+
+
+def iter_skill_index_files(skills_dir: Path, filename: str):
+    """Walk skills_dir yielding sorted paths matching *filename*.
+
+    Excludes ``.git``, ``.github``, ``.hub`` directories.
+    """
+    matches = []
+    for root, dirs, files in os.walk(skills_dir):
+        dirs[:] = [d for d in dirs if d not in EXCLUDED_SKILL_DIRS]
+        if filename in files:
+            matches.append(Path(root) / filename)
+    for path in sorted(matches, key=lambda p: str(p.relative_to(skills_dir))):
+        yield path

agent/smart_model_routing.py

@@ -0,0 +1,198 @@
+"""Helpers for optional cheap-vs-strong model routing."""
+
+from __future__ import annotations
+
+import os
+import re
+from typing import Any, Dict, Optional
+
+_COMPLEX_KEYWORDS = {
+    "debug",
+    "debugging",
+    "implement",
+    "implementation",
+    "refactor",
+    "patch",
+    "traceback",
+    "stacktrace",
+    "exception",
+    "error",
+    "analyze",
+    "analysis",
+    "investigate",
+    "architecture",
+    "design",
+    "compare",
+    "benchmark",
+    "optimize",
+    "optimise",
+    "review",
+    "terminal",
+    "shell",
+    "tool",
+    "tools",
+    "pytest",
+    "test",
+    "tests",
+    "plan",
+    "planning",
+    "delegate",
+    "subagent",
+    "cron",
+    "docker",
+    "kubernetes",
+}
+
+_URL_RE = re.compile(r"https?://|www\.", re.IGNORECASE)
+
+
+def _coerce_bool(value: Any, default: bool = False) -> bool:
+    if value is None:
+        return default
+    if isinstance(value, bool):
+        return value
+    if isinstance(value, str):
+        return value.strip().lower() in {"1", "true", "yes", "on"}
+    return bool(value)
+
+
+def _coerce_int(value: Any, default: int) -> int:
+    try:
+        return int(value)
+    except (TypeError, ValueError):
+        return default
+
+
+def choose_cheap_model_route(user_message: str, routing_config: Optional[Dict[str, Any]]) -> Optional[Dict[str, Any]]:
+    """Return the configured cheap-model route when a message looks simple.
+
+    Conservative by design: if the message has signs of code/tool/debugging/
+    long-form work, keep the primary model.
+    """
+    cfg = routing_config or {}
+    if not _coerce_bool(cfg.get("enabled"), False):
+        return None
+
+    cheap_model = cfg.get("cheap_model") or {}
+    if not isinstance(cheap_model, dict):
+        return None
+    provider = str(cheap_model.get("provider") or "").strip().lower()
+    model = str(cheap_model.get("model") or "").strip()
+    if not provider or not model:
+        return None
+
+    text = (user_message or "").strip()
+    if not text:
+        return None
+
+    max_chars = _coerce_int(cfg.get("max_simple_chars"), 160)
+    max_words = _coerce_int(cfg.get("max_simple_words"), 28)
+
+    if len(text) > max_chars:
+        return None
+    if len(text.split()) > max_words:
+        return None
+    if text.count("\n") > 1:
+        return None
+    if "```" in text or "`" in text:
+        return None
+    if _URL_RE.search(text):
+        return None
+
+    lowered = text.lower()
+    words = {token.strip(".,:;!?()[]{}\"'`") for token in lowered.split()}
+    if words & _COMPLEX_KEYWORDS:
+        return None
+
+    route = dict(cheap_model)
+    route["provider"] = provider
+    route["model"] = model
+    route["routing_reason"] = "simple_turn"
+    return route
+
+
+def resolve_turn_route(user_message: str, routing_config: Optional[Dict[str, Any]], primary: Dict[str, Any]) -> Dict[str, Any]:
+    """Resolve the effective model/runtime for one turn.
+
+    Returns a dict with model/runtime/signature/label fields.
+    """
+    route = choose_cheap_model_route(user_message, routing_config)
+    if not route:
+        return {
+            "model": primary.get("model"),
+            "runtime": {
+                "api_key": primary.get("api_key"),
+                "base_url": primary.get("base_url"),
+                "provider": primary.get("provider"),
+                "api_mode": primary.get("api_mode"),
+                "command": primary.get("command"),
+                "args": list(primary.get("args") or []),
+                "credential_pool": primary.get("credential_pool"),
+            },
+            "label": None,
+            "signature": (
+                primary.get("model"),
+                primary.get("provider"),
+                primary.get("base_url"),
+                primary.get("api_mode"),
+                primary.get("command"),
+                tuple(primary.get("args") or ()),
+            ),
+        }
+
+    from hermes_cli.runtime_provider import resolve_runtime_provider
+
+    explicit_api_key = None
+    api_key_env = str(route.get("api_key_env") or "").strip()
+    if api_key_env:
+        explicit_api_key = os.getenv(api_key_env) or None
+
+    try:
+        runtime = resolve_runtime_provider(
+            requested=route.get("provider"),
+            explicit_api_key=explicit_api_key,
+            explicit_base_url=route.get("base_url"),
+        )
+    except Exception:
+        return {
+            "model": primary.get("model"),
+            "runtime": {
+                "api_key": primary.get("api_key"),
+                "base_url": primary.get("base_url"),
+                "provider": primary.get("provider"),
+                "api_mode": primary.get("api_mode"),
+                "command": primary.get("command"),
+                "args": list(primary.get("args") or []),
+                "credential_pool": primary.get("credential_pool"),
+            },
+            "label": None,
+            "signature": (
+                primary.get("model"),
+                primary.get("provider"),
+                primary.get("base_url"),
+                primary.get("api_mode"),
+                primary.get("command"),
+                tuple(primary.get("args") or ()),
+            ),
+        }
+
+    return {
+        "model": route.get("model"),
+        "runtime": {
+            "api_key": runtime.get("api_key"),
+            "base_url": runtime.get("base_url"),
+            "provider": runtime.get("provider"),
+            "api_mode": runtime.get("api_mode"),
+            "command": runtime.get("command"),
+            "args": list(runtime.get("args") or []),
+        },
+        "label": f"smart route → {route.get('model')} ({runtime.get('provider')})",
+        "signature": (
+            route.get("model"),
+            runtime.get("provider"),
+            runtime.get("base_url"),
+            runtime.get("api_mode"),
+            runtime.get("command"),
+            tuple(runtime.get("args") or ()),
+        ),
+    }

agent/title_generator.py

@@ -0,0 +1,125 @@
+"""Auto-generate short session titles from the first user/assistant exchange.
+
+Runs asynchronously after the first response is delivered so it never
+adds latency to the user-facing reply.
+"""
+
+import logging
+import threading
+from typing import Optional
+
+from agent.auxiliary_client import call_llm
+
+logger = logging.getLogger(__name__)
+
+_TITLE_PROMPT = (
+    "Generate a short, descriptive title (3-7 words) for a conversation that starts with the "
+    "following exchange. The title should capture the main topic or intent. "
+    "Return ONLY the title text, nothing else. No quotes, no punctuation at the end, no prefixes."
+)
+
+
+def generate_title(user_message: str, assistant_response: str, timeout: float = 30.0) -> Optional[str]:
+    """Generate a session title from the first exchange.
+
+    Uses the auxiliary LLM client (cheapest/fastest available model).
+    Returns the title string or None on failure.
+    """
+    # Truncate long messages to keep the request small
+    user_snippet = user_message[:500] if user_message else ""
+    assistant_snippet = assistant_response[:500] if assistant_response else ""
+
+    messages = [
+        {"role": "system", "content": _TITLE_PROMPT},
+        {"role": "user", "content": f"User: {user_snippet}\n\nAssistant: {assistant_snippet}"},
+    ]
+
+    try:
+        response = call_llm(
+            task="compression",  # reuse compression task config (cheap/fast model)
+            messages=messages,
+            max_tokens=30,
+            temperature=0.3,
+            timeout=timeout,
+        )
+        title = (response.choices[0].message.content or "").strip()
+        # Clean up: remove quotes, trailing punctuation, prefixes like "Title: "
+        title = title.strip('"\'')
+        if title.lower().startswith("title:"):
+            title = title[6:].strip()
+        # Enforce reasonable length
+        if len(title) > 80:
+            title = title[:77] + "..."
+        return title if title else None
+    except Exception as e:
+        logger.debug("Title generation failed: %s", e)
+        return None
+
+
+def auto_title_session(
+    session_db,
+    session_id: str,
+    user_message: str,
+    assistant_response: str,
+) -> None:
+    """Generate and set a session title if one doesn't already exist.
+
+    Called in a background thread after the first exchange completes.
+    Silently skips if:
+    - session_db is None
+    - session already has a title (user-set or previously auto-generated)
+    - title generation fails
+    """
+    if not session_db or not session_id:
+        return
+
+    # Check if title already exists (user may have set one via /title before first response)
+    try:
+        existing = session_db.get_session_title(session_id)
+        if existing:
+            return
+    except Exception:
+        return
+
+    title = generate_title(user_message, assistant_response)
+    if not title:
+        return
+
+    try:
+        session_db.set_session_title(session_id, title)
+        logger.debug("Auto-generated session title: %s", title)
+    except Exception as e:
+        logger.debug("Failed to set auto-generated title: %s", e)
+
+
+def maybe_auto_title(
+    session_db,
+    session_id: str,
+    user_message: str,
+    assistant_response: str,
+    conversation_history: list,
+) -> None:
+    """Fire-and-forget title generation after the first exchange.
+
+    Only generates a title when:
+    - This appears to be the first user→assistant exchange
+    - No title is already set
+    """
+    if not session_db or not session_id or not user_message or not assistant_response:
+        return
+
+    # Count user messages in history to detect first exchange.
+    # conversation_history includes the exchange that just happened,
+    # so for a first exchange we expect exactly 1 user message
+    # (or 2 counting system). Be generous: generate on first 2 exchanges.
+    user_msg_count = sum(1 for m in (conversation_history or []) if m.get("role") == "user")
+    if user_msg_count > 2:
+        return
+
+    thread = threading.Thread(
+        target=auto_title_session,
+        args=(session_db, session_id, user_message, assistant_response),
+        daemon=True,
+        name="auto-title",
+    )
+    thread.start()

agent/trajectory.py

@@ -0,0 +1,56 @@
+"""Trajectory saving utilities and static helpers.
+
+_convert_to_trajectory_format stays as an AIAgent method (batch_runner.py
+calls agent._convert_to_trajectory_format). Only the static helpers and
+the file-write logic live here.
+"""
+
+import json
+import logging
+from datetime import datetime
+from typing import Any, Dict, List
+
+logger = logging.getLogger(__name__)
+
+
+def convert_scratchpad_to_think(content: str) -> str:
+    """Convert <REASONING_SCRATCHPAD> tags to <think> tags."""
+    if not content or "<REASONING_SCRATCHPAD>" not in content:
+        return content
+    return content.replace("<REASONING_SCRATCHPAD>", "<think>").replace("</REASONING_SCRATCHPAD>", "</think>")
+
+
+def has_incomplete_scratchpad(content: str) -> bool:
+    """Check if content has an opening <REASONING_SCRATCHPAD> without a closing tag."""
+    if not content:
+        return False
+    return "<REASONING_SCRATCHPAD>" in content and "</REASONING_SCRATCHPAD>" not in content
+
+
+def save_trajectory(trajectory: List[Dict[str, Any]], model: str,
+                    completed: bool, filename: str = None):
+    """Append a trajectory entry to a JSONL file.
+
+    Args:
+        trajectory: The ShareGPT-format conversation list.
+        model: Model name for metadata.
+        completed: Whether the conversation completed successfully.
+        filename: Override output filename. Defaults to trajectory_samples.jsonl
+                  or failed_trajectories.jsonl based on ``completed``.
+    """
+    if filename is None:
+        filename = "trajectory_samples.jsonl" if completed else "failed_trajectories.jsonl"
+
+    entry = {
+        "conversations": trajectory,
+        "timestamp": datetime.now().isoformat(),
+        "model": model,
+        "completed": completed,
+    }
+
+    try:
+        with open(filename, "a", encoding="utf-8") as f:
+            f.write(json.dumps(entry, ensure_ascii=False) + "\n")
+        logger.info("Trajectory saved to %s", filename)
+    except Exception as e:
+        logger.warning("Failed to save trajectory: %s", e)

agent/usage_pricing.py

@@ -0,0 +1,656 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+from datetime import datetime, timezone
+from decimal import Decimal
+from typing import Any, Dict, Literal, Optional
+
+from agent.model_metadata import fetch_endpoint_model_metadata, fetch_model_metadata
+
+DEFAULT_PRICING = {"input": 0.0, "output": 0.0}
+
+_ZERO = Decimal("0")
+_ONE_MILLION = Decimal("1000000")
+
+CostStatus = Literal["actual", "estimated", "included", "unknown"]
+CostSource = Literal[
+    "provider_cost_api",
+    "provider_generation_api",
+    "provider_models_api",
+    "official_docs_snapshot",
+    "user_override",
+    "custom_contract",
+    "none",
+]
+
+
+@dataclass(frozen=True)
+class CanonicalUsage:
+    input_tokens: int = 0
+    output_tokens: int = 0
+    cache_read_tokens: int = 0
+    cache_write_tokens: int = 0
+    reasoning_tokens: int = 0
+    request_count: int = 1
+    raw_usage: Optional[dict[str, Any]] = None
+
+    @property
+    def prompt_tokens(self) -> int:
+        return self.input_tokens + self.cache_read_tokens + self.cache_write_tokens
+
+    @property
+    def total_tokens(self) -> int:
+        return self.prompt_tokens + self.output_tokens
+
+
+@dataclass(frozen=True)
+class BillingRoute:
+    provider: str
+    model: str
+    base_url: str = ""
+    billing_mode: str = "unknown"
+
+
+@dataclass(frozen=True)
+class PricingEntry:
+    input_cost_per_million: Optional[Decimal] = None
+    output_cost_per_million: Optional[Decimal] = None
+    cache_read_cost_per_million: Optional[Decimal] = None
+    cache_write_cost_per_million: Optional[Decimal] = None
+    request_cost: Optional[Decimal] = None
+    source: CostSource = "none"
+    source_url: Optional[str] = None
+    pricing_version: Optional[str] = None
+    fetched_at: Optional[datetime] = None
+
+
+@dataclass(frozen=True)
+class CostResult:
+    amount_usd: Optional[Decimal]
+    status: CostStatus
+    source: CostSource
+    label: str
+    fetched_at: Optional[datetime] = None
+    pricing_version: Optional[str] = None
+    notes: tuple[str, ...] = ()
+
+
+_UTC_NOW = lambda: datetime.now(timezone.utc)
+
+
+# Official docs snapshot entries. Models whose published pricing and cache
+# semantics are stable enough to encode exactly.
+_OFFICIAL_DOCS_PRICING: Dict[tuple[str, str], PricingEntry] = {
+    (
+        "anthropic",
+        "claude-opus-4-20250514",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("15.00"),
+        output_cost_per_million=Decimal("75.00"),
+        cache_read_cost_per_million=Decimal("1.50"),
+        cache_write_cost_per_million=Decimal("18.75"),
+        source="official_docs_snapshot",
+        source_url="https://docs.anthropic.com/en/docs/build-with-claude/prompt-caching",
+        pricing_version="anthropic-prompt-caching-2026-03-16",
+    ),
+    (
+        "anthropic",
+        "claude-sonnet-4-20250514",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("3.00"),
+        output_cost_per_million=Decimal("15.00"),
+        cache_read_cost_per_million=Decimal("0.30"),
+        cache_write_cost_per_million=Decimal("3.75"),
+        source="official_docs_snapshot",
+        source_url="https://docs.anthropic.com/en/docs/build-with-claude/prompt-caching",
+        pricing_version="anthropic-prompt-caching-2026-03-16",
+    ),
+    # OpenAI
+    (
+        "openai",
+        "gpt-4o",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("2.50"),
+        output_cost_per_million=Decimal("10.00"),
+        cache_read_cost_per_million=Decimal("1.25"),
+        source="official_docs_snapshot",
+        source_url="https://openai.com/api/pricing/",
+        pricing_version="openai-pricing-2026-03-16",
+    ),
+    (
+        "openai",
+        "gpt-4o-mini",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("0.15"),
+        output_cost_per_million=Decimal("0.60"),
+        cache_read_cost_per_million=Decimal("0.075"),
+        source="official_docs_snapshot",
+        source_url="https://openai.com/api/pricing/",
+        pricing_version="openai-pricing-2026-03-16",
+    ),
+    (
+        "openai",
+        "gpt-4.1",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("2.00"),
+        output_cost_per_million=Decimal("8.00"),
+        cache_read_cost_per_million=Decimal("0.50"),
+        source="official_docs_snapshot",
+        source_url="https://openai.com/api/pricing/",
+        pricing_version="openai-pricing-2026-03-16",
+    ),
+    (
+        "openai",
+        "gpt-4.1-mini",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("0.40"),
+        output_cost_per_million=Decimal("1.60"),
+        cache_read_cost_per_million=Decimal("0.10"),
+        source="official_docs_snapshot",
+        source_url="https://openai.com/api/pricing/",
+        pricing_version="openai-pricing-2026-03-16",
+    ),
+    (
+        "openai",
+        "gpt-4.1-nano",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("0.10"),
+        output_cost_per_million=Decimal("0.40"),
+        cache_read_cost_per_million=Decimal("0.025"),
+        source="official_docs_snapshot",
+        source_url="https://openai.com/api/pricing/",
+        pricing_version="openai-pricing-2026-03-16",
+    ),
+    (
+        "openai",
+        "o3",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("10.00"),
+        output_cost_per_million=Decimal("40.00"),
+        cache_read_cost_per_million=Decimal("2.50"),
+        source="official_docs_snapshot",
+        source_url="https://openai.com/api/pricing/",
+        pricing_version="openai-pricing-2026-03-16",
+    ),
+    (
+        "openai",
+        "o3-mini",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("1.10"),
+        output_cost_per_million=Decimal("4.40"),
+        cache_read_cost_per_million=Decimal("0.55"),
+        source="official_docs_snapshot",
+        source_url="https://openai.com/api/pricing/",
+        pricing_version="openai-pricing-2026-03-16",
+    ),
+    # Anthropic older models (pre-4.6 generation)
+    (
+        "anthropic",
+        "claude-3-5-sonnet-20241022",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("3.00"),
+        output_cost_per_million=Decimal("15.00"),
+        cache_read_cost_per_million=Decimal("0.30"),
+        cache_write_cost_per_million=Decimal("3.75"),
+        source="official_docs_snapshot",
+        source_url="https://docs.anthropic.com/en/docs/build-with-claude/prompt-caching",
+        pricing_version="anthropic-pricing-2026-03-16",
+    ),
+    (
+        "anthropic",
+        "claude-3-5-haiku-20241022",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("0.80"),
+        output_cost_per_million=Decimal("4.00"),
+        cache_read_cost_per_million=Decimal("0.08"),
+        cache_write_cost_per_million=Decimal("1.00"),
+        source="official_docs_snapshot",
+        source_url="https://docs.anthropic.com/en/docs/build-with-claude/prompt-caching",
+        pricing_version="anthropic-pricing-2026-03-16",
+    ),
+    (
+        "anthropic",
+        "claude-3-opus-20240229",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("15.00"),
+        output_cost_per_million=Decimal("75.00"),
+        cache_read_cost_per_million=Decimal("1.50"),
+        cache_write_cost_per_million=Decimal("18.75"),
+        source="official_docs_snapshot",
+        source_url="https://docs.anthropic.com/en/docs/build-with-claude/prompt-caching",
+        pricing_version="anthropic-pricing-2026-03-16",
+    ),
+    (
+        "anthropic",
+        "claude-3-haiku-20240307",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("0.25"),
+        output_cost_per_million=Decimal("1.25"),
+        cache_read_cost_per_million=Decimal("0.03"),
+        cache_write_cost_per_million=Decimal("0.30"),
+        source="official_docs_snapshot",
+        source_url="https://docs.anthropic.com/en/docs/build-with-claude/prompt-caching",
+        pricing_version="anthropic-pricing-2026-03-16",
+    ),
+    # DeepSeek
+    (
+        "deepseek",
+        "deepseek-chat",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("0.14"),
+        output_cost_per_million=Decimal("0.28"),
+        source="official_docs_snapshot",
+        source_url="https://api-docs.deepseek.com/quick_start/pricing",
+        pricing_version="deepseek-pricing-2026-03-16",
+    ),
+    (
+        "deepseek",
+        "deepseek-reasoner",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("0.55"),
+        output_cost_per_million=Decimal("2.19"),
+        source="official_docs_snapshot",
+        source_url="https://api-docs.deepseek.com/quick_start/pricing",
+        pricing_version="deepseek-pricing-2026-03-16",
+    ),
+    # Google Gemini
+    (
+        "google",
+        "gemini-2.5-pro",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("1.25"),
+        output_cost_per_million=Decimal("10.00"),
+        source="official_docs_snapshot",
+        source_url="https://ai.google.dev/pricing",
+        pricing_version="google-pricing-2026-03-16",
+    ),
+    (
+        "google",
+        "gemini-2.5-flash",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("0.15"),
+        output_cost_per_million=Decimal("0.60"),
+        source="official_docs_snapshot",
+        source_url="https://ai.google.dev/pricing",
+        pricing_version="google-pricing-2026-03-16",
+    ),
+    (
+        "google",
+        "gemini-2.0-flash",
+    ): PricingEntry(
+        input_cost_per_million=Decimal("0.10"),
+        output_cost_per_million=Decimal("0.40"),
+        source="official_docs_snapshot",
+        source_url="https://ai.google.dev/pricing",
+        pricing_version="google-pricing-2026-03-16",
+    ),
+}
+
+
+def _to_decimal(value: Any) -> Optional[Decimal]:
+    if value is None:
+        return None
+    try:
+        return Decimal(str(value))
+    except Exception:
+        return None
+
+
+def _to_int(value: Any) -> int:
+    try:
+        return int(value or 0)
+    except Exception:
+        return 0
+
+
+def resolve_billing_route(
+    model_name: str,
+    provider: Optional[str] = None,
+    base_url: Optional[str] = None,
+) -> BillingRoute:
+    provider_name = (provider or "").strip().lower()
+    base = (base_url or "").strip().lower()
+    model = (model_name or "").strip()
+    if not provider_name and "/" in model:
+        inferred_provider, bare_model = model.split("/", 1)
+        if inferred_provider in {"anthropic", "openai", "google"}:
+            provider_name = inferred_provider
+            model = bare_model
+
+    if provider_name == "openai-codex":
+        return BillingRoute(provider="openai-codex", model=model, base_url=base_url or "", billing_mode="subscription_included")
+    if provider_name == "openrouter" or "openrouter.ai" in base:
+        return BillingRoute(provider="openrouter", model=model, base_url=base_url or "", billing_mode="official_models_api")
+    if provider_name == "anthropic":
+        return BillingRoute(provider="anthropic", model=model.split("/")[-1], base_url=base_url or "", billing_mode="official_docs_snapshot")
+    if provider_name == "openai":
+        return BillingRoute(provider="openai", model=model.split("/")[-1], base_url=base_url or "", billing_mode="official_docs_snapshot")
+    if provider_name in {"custom", "local"} or (base and "localhost" in base):
+        return BillingRoute(provider=provider_name or "custom", model=model, base_url=base_url or "", billing_mode="unknown")
+    return BillingRoute(provider=provider_name or "unknown", model=model.split("/")[-1] if model else "", base_url=base_url or "", billing_mode="unknown")
+
+
+def _lookup_official_docs_pricing(route: BillingRoute) -> Optional[PricingEntry]:
+    return _OFFICIAL_DOCS_PRICING.get((route.provider, route.model.lower()))
+
+
+def _openrouter_pricing_entry(route: BillingRoute) -> Optional[PricingEntry]:
+    return _pricing_entry_from_metadata(
+        fetch_model_metadata(),
+        route.model,
+        source_url="https://openrouter.ai/docs/api/api-reference/models/get-models",
+        pricing_version="openrouter-models-api",
+    )
+
+
+def _pricing_entry_from_metadata(
+    metadata: Dict[str, Dict[str, Any]],
+    model_id: str,
+    *,
+    source_url: str,
+    pricing_version: str,
+) -> Optional[PricingEntry]:
+    if model_id not in metadata:
+        return None
+    pricing = metadata[model_id].get("pricing") or {}
+    prompt = _to_decimal(pricing.get("prompt"))
+    completion = _to_decimal(pricing.get("completion"))
+    request = _to_decimal(pricing.get("request"))
+    cache_read = _to_decimal(
+        pricing.get("cache_read")
+        or pricing.get("cached_prompt")
+        or pricing.get("input_cache_read")
+    )
+    cache_write = _to_decimal(
+        pricing.get("cache_write")
+        or pricing.get("cache_creation")
+        or pricing.get("input_cache_write")
+    )
+    if prompt is None and completion is None and request is None:
+        return None
+
+    def _per_token_to_per_million(value: Optional[Decimal]) -> Optional[Decimal]:
+        if value is None:
+            return None
+        return value * _ONE_MILLION
+
+    return PricingEntry(
+        input_cost_per_million=_per_token_to_per_million(prompt),
+        output_cost_per_million=_per_token_to_per_million(completion),
+        cache_read_cost_per_million=_per_token_to_per_million(cache_read),
+        cache_write_cost_per_million=_per_token_to_per_million(cache_write),
+        request_cost=request,
+        source="provider_models_api",
+        source_url=source_url,
+        pricing_version=pricing_version,
+        fetched_at=_UTC_NOW(),
+    )
+
+
+def get_pricing_entry(
+    model_name: str,
+    provider: Optional[str] = None,
+    base_url: Optional[str] = None,
+    api_key: Optional[str] = None,
+) -> Optional[PricingEntry]:
+    route = resolve_billing_route(model_name, provider=provider, base_url=base_url)
+    if route.billing_mode == "subscription_included":
+        return PricingEntry(
+            input_cost_per_million=_ZERO,
+            output_cost_per_million=_ZERO,
+            cache_read_cost_per_million=_ZERO,
+            cache_write_cost_per_million=_ZERO,
+            source="none",
+            pricing_version="included-route",
+        )
+    if route.provider == "openrouter":
+        return _openrouter_pricing_entry(route)
+    if route.base_url:
+        entry = _pricing_entry_from_metadata(
+            fetch_endpoint_model_metadata(route.base_url, api_key=api_key or ""),
+            route.model,
+            source_url=f"{route.base_url.rstrip('/')}/models",
+            pricing_version="openai-compatible-models-api",
+        )
+        if entry:
+            return entry
+    return _lookup_official_docs_pricing(route)
+
+
+def normalize_usage(
+    response_usage: Any,
+    *,
+    provider: Optional[str] = None,
+    api_mode: Optional[str] = None,
+) -> CanonicalUsage:
+    """Normalize raw API response usage into canonical token buckets.
+
+    Handles three API shapes:
+    - Anthropic: input_tokens/output_tokens/cache_read_input_tokens/cache_creation_input_tokens
+    - Codex Responses: input_tokens includes cache tokens; input_tokens_details.cached_tokens separates them
+    - OpenAI Chat Completions: prompt_tokens includes cache tokens; prompt_tokens_details.cached_tokens separates them
+
+    In both Codex and OpenAI modes, input_tokens is derived by subtracting cache
+    tokens from the total — the API contract is that input/prompt totals include
+    cached tokens and the details object breaks them out.
+    """
+    if not response_usage:
+        return CanonicalUsage()
+
+    provider_name = (provider or "").strip().lower()
+    mode = (api_mode or "").strip().lower()
+
+    if mode == "anthropic_messages" or provider_name == "anthropic":
+        input_tokens = _to_int(getattr(response_usage, "input_tokens", 0))
+        output_tokens = _to_int(getattr(response_usage, "output_tokens", 0))
+        cache_read_tokens = _to_int(getattr(response_usage, "cache_read_input_tokens", 0))
+        cache_write_tokens = _to_int(getattr(response_usage, "cache_creation_input_tokens", 0))
+    elif mode == "codex_responses":
+        input_total = _to_int(getattr(response_usage, "input_tokens", 0))
+        output_tokens = _to_int(getattr(response_usage, "output_tokens", 0))
+        details = getattr(response_usage, "input_tokens_details", None)
+        cache_read_tokens = _to_int(getattr(details, "cached_tokens", 0) if details else 0)
+        cache_write_tokens = _to_int(
+            getattr(details, "cache_creation_tokens", 0) if details else 0
+        )
+        input_tokens = max(0, input_total - cache_read_tokens - cache_write_tokens)
+    else:
+        prompt_total = _to_int(getattr(response_usage, "prompt_tokens", 0))
+        output_tokens = _to_int(getattr(response_usage, "completion_tokens", 0))
+        details = getattr(response_usage, "prompt_tokens_details", None)
+        cache_read_tokens = _to_int(getattr(details, "cached_tokens", 0) if details else 0)
+        cache_write_tokens = _to_int(
+            getattr(details, "cache_write_tokens", 0) if details else 0
+        )
+        input_tokens = max(0, prompt_total - cache_read_tokens - cache_write_tokens)
+
+    reasoning_tokens = 0
+    output_details = getattr(response_usage, "output_tokens_details", None)
+    if output_details:
+        reasoning_tokens = _to_int(getattr(output_details, "reasoning_tokens", 0))
+
+    return CanonicalUsage(
+        input_tokens=input_tokens,
+        output_tokens=output_tokens,
+        cache_read_tokens=cache_read_tokens,
+        cache_write_tokens=cache_write_tokens,
+        reasoning_tokens=reasoning_tokens,
+    )
+
+
+def estimate_usage_cost(
+    model_name: str,
+    usage: CanonicalUsage,
+    *,
+    provider: Optional[str] = None,
+    base_url: Optional[str] = None,
+    api_key: Optional[str] = None,
+) -> CostResult:
+    route = resolve_billing_route(model_name, provider=provider, base_url=base_url)
+    if route.billing_mode == "subscription_included":
+        return CostResult(
+            amount_usd=_ZERO,
+            status="included",
+            source="none",
+            label="included",
+            pricing_version="included-route",
+        )
+
+    entry = get_pricing_entry(model_name, provider=provider, base_url=base_url, api_key=api_key)
+    if not entry:
+        return CostResult(amount_usd=None, status="unknown", source="none", label="n/a")
+
+    notes: list[str] = []
+    amount = _ZERO
+
+    if usage.input_tokens and entry.input_cost_per_million is None:
+        return CostResult(amount_usd=None, status="unknown", source=entry.source, label="n/a")
+    if usage.output_tokens and entry.output_cost_per_million is None:
+        return CostResult(amount_usd=None, status="unknown", source=entry.source, label="n/a")
+    if usage.cache_read_tokens:
+        if entry.cache_read_cost_per_million is None:
+            return CostResult(
+                amount_usd=None,
+                status="unknown",
+                source=entry.source,
+                label="n/a",
+                notes=("cache-read pricing unavailable for route",),
+            )
+    if usage.cache_write_tokens:
+        if entry.cache_write_cost_per_million is None:
+            return CostResult(
+                amount_usd=None,
+                status="unknown",
+                source=entry.source,
+                label="n/a",
+                notes=("cache-write pricing unavailable for route",),
+            )
+
+    if entry.input_cost_per_million is not None:
+        amount += Decimal(usage.input_tokens) * entry.input_cost_per_million / _ONE_MILLION
+    if entry.output_cost_per_million is not None:
+        amount += Decimal(usage.output_tokens) * entry.output_cost_per_million / _ONE_MILLION
+    if entry.cache_read_cost_per_million is not None:
+        amount += Decimal(usage.cache_read_tokens) * entry.cache_read_cost_per_million / _ONE_MILLION
+    if entry.cache_write_cost_per_million is not None:
+        amount += Decimal(usage.cache_write_tokens) * entry.cache_write_cost_per_million / _ONE_MILLION
+    if entry.request_cost is not None and usage.request_count:
+        amount += Decimal(usage.request_count) * entry.request_cost
+
+    status: CostStatus = "estimated"
+    label = f"~${amount:.2f}"
+    if entry.source == "none" and amount == _ZERO:
+        status = "included"
+        label = "included"
+
+    if route.provider == "openrouter":
+        notes.append("OpenRouter cost is estimated from the models API until reconciled.")
+
+    return CostResult(
+        amount_usd=amount,
+        status=status,
+        source=entry.source,
+        label=label,
+        fetched_at=entry.fetched_at,
+        pricing_version=entry.pricing_version,
+        notes=tuple(notes),
+    )
+
+
+def has_known_pricing(
+    model_name: str,
+    provider: Optional[str] = None,
+    base_url: Optional[str] = None,
+    api_key: Optional[str] = None,
+) -> bool:
+    """Check whether we have pricing data for this model+route.
+
+    Uses direct lookup instead of routing through the full estimation
+    pipeline — avoids creating dummy usage objects just to check status.
+    """
+    route = resolve_billing_route(model_name, provider=provider, base_url=base_url)
+    if route.billing_mode == "subscription_included":
+        return True
+    entry = get_pricing_entry(model_name, provider=provider, base_url=base_url, api_key=api_key)
+    return entry is not None
+
+
+def get_pricing(
+    model_name: str,
+    provider: Optional[str] = None,
+    base_url: Optional[str] = None,
+    api_key: Optional[str] = None,
+) -> Dict[str, float]:
+    """Backward-compatible thin wrapper for legacy callers.
+
+    Returns only non-cache input/output fields when a pricing entry exists.
+    Unknown routes return zeroes.
+    """
+    entry = get_pricing_entry(model_name, provider=provider, base_url=base_url, api_key=api_key)
+    if not entry:
+        return {"input": 0.0, "output": 0.0}
+    return {
+        "input": float(entry.input_cost_per_million or _ZERO),
+        "output": float(entry.output_cost_per_million or _ZERO),
+    }
+
+
+def estimate_cost_usd(
+    model: str,
+    input_tokens: int,
+    output_tokens: int,
+    *,
+    provider: Optional[str] = None,
+    base_url: Optional[str] = None,
+    api_key: Optional[str] = None,
+) -> float:
+    """Backward-compatible helper for legacy callers.
+
+    This uses non-cached input/output only. New code should call
+    `estimate_usage_cost()` with canonical usage buckets.
+    """
+    result = estimate_usage_cost(
+        model,
+        CanonicalUsage(input_tokens=input_tokens, output_tokens=output_tokens),
+        provider=provider,
+        base_url=base_url,
+        api_key=api_key,
+    )
+    return float(result.amount_usd or _ZERO)
+
+
+def format_duration_compact(seconds: float) -> str:
+    if seconds < 60:
+        return f"{seconds:.0f}s"
+    minutes = seconds / 60
+    if minutes < 60:
+        return f"{minutes:.0f}m"
+    hours = minutes / 60
+    if hours < 24:
+        remaining_min = int(minutes % 60)
+        return f"{int(hours)}h {remaining_min}m" if remaining_min else f"{int(hours)}h"
+    days = hours / 24
+    return f"{days:.1f}d"
+
+
+def format_token_count_compact(value: int) -> str:
+    abs_value = abs(int(value))
+    if abs_value < 1_000:
+        return str(int(value))
+
+    sign = "-" if value < 0 else ""
+    units = ((1_000_000_000, "B"), (1_000_000, "M"), (1_000, "K"))
+    for threshold, suffix in units:
+        if abs_value >= threshold:
+            scaled = abs_value / threshold
+            if scaled < 10:
+                text = f"{scaled:.2f}"
+            elif scaled < 100:
+                text = f"{scaled:.1f}"
+            else:
+                text = f"{scaled:.0f}"
+            if "." in text:
+                text = text.rstrip("0").rstrip(".")
+            return f"{sign}{text}{suffix}"
+
+    return f"{value:,}"

batch_runner.py

@@ -27,23 +27,87 @@ import time
 from pathlib import Path
 from typing import List, Dict, Any, Optional, Tuple
 from datetime import datetime
-from multiprocessing import Pool, Manager, Lock
+from multiprocessing import Pool, Lock
 import traceback
-
+from rich.progress import Progress, SpinnerColumn, BarColumn, TextColumn, TimeRemainingColumn, MofNCompleteColumn
+from rich.console import Console
 import fire
 
 from run_agent import AIAgent
 from toolset_distributions import (
-    get_distribution, 
     list_distributions, 
     sample_toolsets_from_distribution,
     validate_distribution
 )
+from model_tools import TOOL_TO_TOOLSET_MAP
 
 
 # Global configuration for worker processes
 _WORKER_CONFIG = {}
 
+# All possible tools - auto-derived from the master mapping in model_tools.py.
+# This stays in sync automatically when new tools are added to TOOL_TO_TOOLSET_MAP.
+# Used for consistent schema in Arrow/Parquet (HuggingFace datasets) and for
+# filtering corrupted entries during trajectory combination.
+ALL_POSSIBLE_TOOLS = set(TOOL_TO_TOOLSET_MAP.keys())
+
+# Default stats for tools that weren't used
+DEFAULT_TOOL_STATS = {'count': 0, 'success': 0, 'failure': 0}
+
+
+def _normalize_tool_stats(tool_stats: Dict[str, Dict[str, int]]) -> Dict[str, Dict[str, int]]:
+    """
+    Normalize tool_stats to include all possible tools with consistent schema.
+    
+    This ensures HuggingFace datasets can load the JSONL without schema mismatch errors.
+    Tools that weren't used get zero counts.
+    
+    Args:
+        tool_stats (Dict): Raw tool statistics from extraction
+        
+    Returns:
+        Dict: Normalized tool statistics with all tools present
+    """
+    normalized = {}
+    
+    # Add all possible tools with defaults
+    for tool in ALL_POSSIBLE_TOOLS:
+        if tool in tool_stats:
+            normalized[tool] = tool_stats[tool].copy()
+        else:
+            normalized[tool] = DEFAULT_TOOL_STATS.copy()
+    
+    # Also include any unexpected tools (in case new tools are added)
+    for tool, stats in tool_stats.items():
+        if tool not in normalized:
+            normalized[tool] = stats.copy()
+    
+    return normalized
+
+
+def _normalize_tool_error_counts(tool_error_counts: Dict[str, int]) -> Dict[str, int]:
+    """
+    Normalize tool_error_counts to include all possible tools.
+    
+    Args:
+        tool_error_counts (Dict): Raw error counts mapping
+        
+    Returns:
+        Dict: Normalized error counts with all tools present
+    """
+    normalized = {}
+    
+    # Add all possible tools with zero defaults
+    for tool in ALL_POSSIBLE_TOOLS:
+        normalized[tool] = tool_error_counts.get(tool, 0)
+    
+    # Also include any unexpected tools
+    for tool, count in tool_error_counts.items():
+        if tool not in normalized:
+            normalized[tool] = count
+    
+    return normalized
+
 
 def _extract_tool_stats(messages: List[Dict[str, Any]]) -> Dict[str, Dict[str, int]]:
     """
@@ -64,6 +128,7 @@ def _extract_tool_stats(messages: List[Dict[str, Any]]) -> Dict[str, Dict[str, i
         # Track tool calls from assistant messages
         if msg["role"] == "assistant" and "tool_calls" in msg and msg["tool_calls"]:
             for tool_call in msg["tool_calls"]:
+                if not tool_call or not isinstance(tool_call, dict): continue
                 tool_name = tool_call["function"]["name"]
                 tool_call_id = tool_call["id"]
                 
@@ -107,7 +172,7 @@ def _extract_tool_stats(messages: List[Dict[str, Any]]) -> Dict[str, Dict[str, i
                     if content_json.get("success") is False:
                         is_success = False
                         
-            except:
+            except (json.JSONDecodeError, ValueError, TypeError):
                 # If not JSON, check if content is empty or explicitly states an error
                 # Note: We avoid simple substring matching to prevent false positives
                 if not content:
@@ -127,6 +192,42 @@ def _extract_tool_stats(messages: List[Dict[str, Any]]) -> Dict[str, Dict[str, i
     return tool_stats
 
 
+def _extract_reasoning_stats(messages: List[Dict[str, Any]]) -> Dict[str, int]:
+    """
+    Count how many assistant turns have reasoning vs no reasoning.
+    
+    Checks for <REASONING_SCRATCHPAD> in content or a non-empty 'reasoning' field
+    (native thinking tokens). Returns counts for tracking reasoning coverage.
+    
+    Args:
+        messages: Message history
+        
+    Returns:
+        Dict with 'total_assistant_turns', 'turns_with_reasoning', 'turns_without_reasoning'
+    """
+    total = 0
+    with_reasoning = 0
+    
+    for msg in messages:
+        if msg.get("role") != "assistant":
+            continue
+        total += 1
+        
+        content = msg.get("content", "") or ""
+        has_scratchpad = "<REASONING_SCRATCHPAD>" in content
+        has_native_reasoning = bool(msg.get("reasoning", "").strip()) if msg.get("reasoning") else False
+        
+        if has_scratchpad or has_native_reasoning:
+            with_reasoning += 1
+    
+    return {
+        "total_assistant_turns": total,
+        "turns_with_reasoning": with_reasoning,
+        "turns_without_reasoning": total - with_reasoning,
+        "has_any_reasoning": with_reasoning > 0,
+    }
+
+
 def _process_single_prompt(
     prompt_index: int,
     prompt_data: Dict[str, Any],
@@ -138,7 +239,7 @@ def _process_single_prompt(
     
     Args:
         prompt_index (int): Index of prompt in dataset
-        prompt_data (Dict): Prompt data containing 'prompt' field
+        prompt_data (Dict): Prompt data containing 'prompt' field and optional 'image' field
         batch_num (int): Batch number
         config (Dict): Configuration dict with agent parameters
         
@@ -146,6 +247,58 @@ def _process_single_prompt(
         Dict: Result containing trajectory, stats, and metadata
     """
     prompt = prompt_data["prompt"]
+    task_id = f"task_{prompt_index}"
+    
+    # Per-prompt container image override: if the dataset row has an 'image' field,
+    # register it for this task's sandbox. Works with Docker, Modal, Singularity, and Daytona.
+    container_image = prompt_data.get("image") or prompt_data.get("docker_image")
+    if container_image:
+        # Verify the image is accessible before spending tokens on the agent loop.
+        # For Docker: check local cache, then try pulling.
+        # For Modal: skip local check (Modal pulls server-side).
+        env_type = os.getenv("TERMINAL_ENV", "local")
+        if env_type == "docker":
+            import subprocess as _sp
+            try:
+                probe = _sp.run(
+                    ["docker", "image", "inspect", container_image],
+                    capture_output=True, timeout=10,
+                )
+                if probe.returncode != 0:
+                    if config.get("verbose"):
+                        print(f"   Prompt {prompt_index}: Pulling docker image {container_image}...", flush=True)
+                    pull = _sp.run(
+                        ["docker", "pull", container_image],
+                        capture_output=True, text=True, timeout=600,
+                    )
+                    if pull.returncode != 0:
+                        return {
+                            "success": False,
+                            "prompt_index": prompt_index,
+                            "error": f"Docker image not available: {container_image}\n{pull.stderr[:500]}",
+                            "trajectory": None,
+                            "tool_stats": {},
+                            "toolsets_used": [],
+                            "metadata": {"batch_num": batch_num, "timestamp": datetime.now().isoformat()},
+                        }
+            except FileNotFoundError:
+                pass  # Docker CLI not installed — skip check (e.g., Modal backend)
+            except Exception as img_err:
+                if config.get("verbose"):
+                    print(f"   Prompt {prompt_index}: Docker image check failed: {img_err}", flush=True)
+
+        from tools.terminal_tool import register_task_env_overrides
+        overrides = {
+            "docker_image": container_image,
+            "modal_image": container_image,
+            "singularity_image": f"docker://{container_image}",
+            "daytona_image": container_image,
+        }
+        if prompt_data.get("cwd"):
+            overrides["cwd"] = prompt_data["cwd"]
+        register_task_env_overrides(task_id, overrides)
+        if config.get("verbose"):
+            print(f"   Prompt {prompt_index}: Using container image {container_image}")
     
     try:
         # Sample toolsets from distribution for this prompt
@@ -154,7 +307,8 @@ def _process_single_prompt(
         if config.get("verbose"):
             print(f"   Prompt {prompt_index}: Using toolsets {selected_toolsets}")
         
-        # Initialize agent with sampled toolsets
+        # Initialize agent with sampled toolsets and log prefix for identification
+        log_prefix = f"[B{batch_num}:P{prompt_index}]"
         agent = AIAgent(
             base_url=config.get("base_url"),
             api_key=config.get("api_key"),
@@ -164,15 +318,28 @@ def _process_single_prompt(
             save_trajectories=False,  # We handle saving ourselves
             verbose_logging=config.get("verbose", False),
             ephemeral_system_prompt=config.get("ephemeral_system_prompt"),
-            log_prefix_chars=config.get("log_prefix_chars", 100)
+            log_prefix_chars=config.get("log_prefix_chars", 100),
+            log_prefix=log_prefix,
+            providers_allowed=config.get("providers_allowed"),
+            providers_ignored=config.get("providers_ignored"),
+            providers_order=config.get("providers_order"),
+            provider_sort=config.get("provider_sort"),
+            max_tokens=config.get("max_tokens"),
+            reasoning_config=config.get("reasoning_config"),
+            prefill_messages=config.get("prefill_messages"),
+            skip_context_files=True,  # Don't pollute trajectories with SOUL.md/AGENTS.md
+            skip_memory=True,  # Don't use persistent memory in batch runs
         )
 
         # Run the agent with task_id to ensure each task gets its own isolated VM
-        result = agent.run_conversation(prompt, task_id=f"task_{prompt_index}")
+        result = agent.run_conversation(prompt, task_id=task_id)
         
         # Extract tool usage statistics
         tool_stats = _extract_tool_stats(result["messages"])
         
+        # Extract reasoning coverage stats
+        reasoning_stats = _extract_reasoning_stats(result["messages"])
+        
         # Convert to trajectory format (using existing method)
         trajectory = agent._convert_to_trajectory_format(
             result["messages"],
@@ -185,7 +352,9 @@ def _process_single_prompt(
             "prompt_index": prompt_index,
             "trajectory": trajectory,
             "tool_stats": tool_stats,
+            "reasoning_stats": reasoning_stats,
             "completed": result["completed"],
+            "partial": result.get("partial", False),
             "api_calls": result["api_calls"],
             "toolsets_used": selected_toolsets,
             "metadata": {
@@ -252,7 +421,9 @@ def _process_batch_worker(args: Tuple) -> Dict[str, Any]:
     
     # Initialize aggregated stats for this batch
     batch_tool_stats = {}
+    batch_reasoning_stats = {"total_assistant_turns": 0, "turns_with_reasoning": 0, "turns_without_reasoning": 0}
     completed_in_batch = []
+    discarded_no_reasoning = 0
     
     # Process each prompt sequentially in this batch
     for prompt_index, prompt_data in prompts_to_process:
@@ -266,13 +437,34 @@ def _process_batch_worker(args: Tuple) -> Dict[str, Any]:
         
         # Save trajectory if successful
         if result["success"] and result["trajectory"]:
+            # Discard samples with zero reasoning across all turns
+            reasoning = result.get("reasoning_stats", {})
+            if not reasoning.get("has_any_reasoning", True):
+                print(f"   🚫 Prompt {prompt_index} discarded (no reasoning in any turn)")
+                discarded_no_reasoning += 1
+                continue
+            
+            # Get and normalize tool stats for consistent schema across all entries
+            raw_tool_stats = result.get("tool_stats", {})
+            tool_stats = _normalize_tool_stats(raw_tool_stats)
+            
+            # Create normalized tool_error_counts mapping tool names to their failure counts
+            raw_error_counts = {
+                tool_name: stats.get("failure", 0) 
+                for tool_name, stats in raw_tool_stats.items()
+            }
+            tool_error_counts = _normalize_tool_error_counts(raw_error_counts)
+            
             trajectory_entry = {
                 "prompt_index": prompt_index,
                 "conversations": result["trajectory"],
                 "metadata": result["metadata"],
                 "completed": result["completed"],
+                "partial": result.get("partial", False),  # True if stopped due to invalid tool calls
                 "api_calls": result["api_calls"],
-                "toolsets_used": result["toolsets_used"]
+                "toolsets_used": result["toolsets_used"],
+                "tool_stats": tool_stats,  # Full stats: {tool: {count, success, failure}} - normalized
+                "tool_error_counts": tool_error_counts  # Simple: {tool: failure_count} - normalized
             }
             
             # Append to batch output file
@@ -292,8 +484,17 @@ def _process_batch_worker(args: Tuple) -> Dict[str, Any]:
             batch_tool_stats[tool_name]["success"] += stats["success"]
             batch_tool_stats[tool_name]["failure"] += stats["failure"]
         
-        completed_in_batch.append(prompt_index)
-        print(f"   ✅ Prompt {prompt_index} completed")
+        # Aggregate reasoning stats
+        for key in batch_reasoning_stats:
+            batch_reasoning_stats[key] += result.get("reasoning_stats", {}).get(key, 0)
+        
+        # Only mark as completed if successfully saved (failed prompts can be retried on resume)
+        if result["success"] and result["trajectory"]:
+            completed_in_batch.append(prompt_index)
+            status = "⚠️  partial" if result.get("partial") else "✅"
+            print(f"   {status} Prompt {prompt_index} completed")
+        else:
+            print(f"   ❌ Prompt {prompt_index} failed (will retry on resume)")
     
     print(f"✅ Batch {batch_num}: Completed ({len(prompts_to_process)} prompts processed)")
     
@@ -302,6 +503,8 @@ def _process_batch_worker(args: Tuple) -> Dict[str, Any]:
         "processed": len(prompts_to_process),
         "skipped": len(batch_data) - len(prompts_to_process),
         "tool_stats": batch_tool_stats,
+        "reasoning_stats": batch_reasoning_stats,
+        "discarded_no_reasoning": discarded_no_reasoning,
         "completed_prompts": completed_in_batch
     }
 
@@ -325,6 +528,14 @@ class BatchRunner:
         verbose: bool = False,
         ephemeral_system_prompt: str = None,
         log_prefix_chars: int = 100,
+        providers_allowed: List[str] = None,
+        providers_ignored: List[str] = None,
+        providers_order: List[str] = None,
+        provider_sort: str = None,
+        max_tokens: int = None,
+        reasoning_config: Dict[str, Any] = None,
+        prefill_messages: List[Dict[str, Any]] = None,
+        max_samples: int = None,
     ):
         """
         Initialize the batch runner.
@@ -342,6 +553,14 @@ class BatchRunner:
             verbose (bool): Enable verbose logging
             ephemeral_system_prompt (str): System prompt used during agent execution but NOT saved to trajectories (optional)
             log_prefix_chars (int): Number of characters to show in log previews for tool calls/responses (default: 20)
+            providers_allowed (List[str]): OpenRouter providers to allow (optional)
+            providers_ignored (List[str]): OpenRouter providers to ignore (optional)
+            providers_order (List[str]): OpenRouter providers to try in order (optional)
+            provider_sort (str): Sort providers by price/throughput/latency (optional)
+            max_tokens (int): Maximum tokens for model responses (optional, uses model default if not set)
+            reasoning_config (Dict): OpenRouter reasoning config override (e.g. {"effort": "none"} to disable thinking)
+            prefill_messages (List[Dict]): Messages to prepend as prefilled conversation context (few-shot priming)
+            max_samples (int): Only process the first N samples from the dataset (optional, processes all if not set)
         """
         self.dataset_file = Path(dataset_file)
         self.batch_size = batch_size
@@ -355,6 +574,14 @@ class BatchRunner:
         self.verbose = verbose
         self.ephemeral_system_prompt = ephemeral_system_prompt
         self.log_prefix_chars = log_prefix_chars
+        self.providers_allowed = providers_allowed
+        self.providers_ignored = providers_ignored
+        self.providers_order = providers_order
+        self.provider_sort = provider_sort
+        self.max_tokens = max_tokens
+        self.reasoning_config = reasoning_config
+        self.prefill_messages = prefill_messages
+        self.max_samples = max_samples
         
         # Validate distribution
         if not validate_distribution(distribution):
@@ -370,13 +597,17 @@ class BatchRunner:
         # Statistics file
         self.stats_file = self.output_dir / "statistics.json"
         
-        # Load dataset
+        # Load dataset (and optionally truncate to max_samples)
         self.dataset = self._load_dataset()
+        if self.max_samples and self.max_samples < len(self.dataset):
+            full_count = len(self.dataset)
+            self.dataset = self.dataset[:self.max_samples]
+            print(f"✂️  Truncated dataset from {full_count} to {self.max_samples} samples (--max_samples)")
         
         # Create batches
         self.batches = self._create_batches()
         
-        print(f"📊 Batch Runner Initialized")
+        print("📊 Batch Runner Initialized")
         print(f"   Dataset: {self.dataset_file} ({len(self.dataset)} prompts)")
         print(f"   Batch size: {self.batch_size}")
         print(f"   Total batches: {len(self.batches)}")
@@ -470,15 +701,91 @@ class BatchRunner:
             lock (Lock): Optional lock for thread-safe access
         """
         checkpoint_data["last_updated"] = datetime.now().isoformat()
-        
+
+        from utils import atomic_json_write
         if lock:
             with lock:
-                with open(self.checkpoint_file, 'w', encoding='utf-8') as f:
-                    json.dump(checkpoint_data, f, indent=2, ensure_ascii=False)
+                atomic_json_write(self.checkpoint_file, checkpoint_data)
         else:
-            with open(self.checkpoint_file, 'w', encoding='utf-8') as f:
-                json.dump(checkpoint_data, f, indent=2, ensure_ascii=False)
+            atomic_json_write(self.checkpoint_file, checkpoint_data)
     
+    def _scan_completed_prompts_by_content(self) -> set:
+        """
+        Scan all batch files and extract completed prompts by their actual content.
+        
+        This provides a more robust resume mechanism that matches on prompt text
+        rather than indices, allowing recovery even if indices don't match.
+        
+        Returns:
+            set: Set of prompt texts that have been successfully processed
+        """
+        completed_prompts = set()
+        batch_files = sorted(self.output_dir.glob("batch_*.jsonl"))
+        
+        if not batch_files:
+            return completed_prompts
+        
+        print(f"📂 Scanning {len(batch_files)} batch files for completed prompts...")
+        
+        for batch_file in batch_files:
+            try:
+                with open(batch_file, 'r', encoding='utf-8') as f:
+                    for line in f:
+                        try:
+                            entry = json.loads(line.strip())
+                            
+                            # Skip failed entries - we want to retry these
+                            if entry.get("failed", False):
+                                continue
+                            
+                            # Extract the human/user prompt from conversations
+                            conversations = entry.get("conversations", [])
+                            for msg in conversations:
+                                if msg.get("from") == "human":
+                                    prompt_text = msg.get("value", "").strip()
+                                    if prompt_text:
+                                        completed_prompts.add(prompt_text)
+                                    break  # Only need the first human message
+                        except json.JSONDecodeError:
+                            continue
+            except Exception as e:
+                print(f"  ⚠️  Warning: Error reading {batch_file.name}: {e}")
+        
+        return completed_prompts
+    
+    def _filter_dataset_by_completed(self, completed_prompts: set) -> Tuple[List[Dict], List[int]]:
+        """
+        Filter the dataset to exclude prompts that have already been completed.
+        
+        Args:
+            completed_prompts: Set of prompt texts that have been completed
+            
+        Returns:
+            Tuple of (filtered_dataset, skipped_indices)
+        """
+        filtered_dataset = []
+        skipped_indices = []
+        
+        for idx, entry in enumerate(self.dataset):
+            # Extract prompt from the dataset entry
+            prompt_text = entry.get("prompt", "").strip()
+            
+            # Also check conversations format
+            if not prompt_text:
+                conversations = entry.get("conversations", [])
+                for msg in conversations:
+                    role = msg.get("role") or msg.get("from")
+                    if role in ("user", "human"):
+                        prompt_text = (msg.get("content") or msg.get("value", "")).strip()
+                        break
+            
+            if prompt_text in completed_prompts:
+                skipped_indices.append(idx)
+            else:
+                # Keep original index for tracking
+                filtered_dataset.append((idx, entry))
+        
+        return filtered_dataset, skipped_indices
     
     def run(self, resume: bool = False):
         """
@@ -491,16 +798,49 @@ class BatchRunner:
         print("🚀 Starting Batch Processing")
         print("=" * 70)
         
-        # Load checkpoint
-        checkpoint_data = self._load_checkpoint() if resume else {
-            "run_name": self.run_name,
-            "completed_prompts": [],
-            "batch_stats": {},
-            "last_updated": None
-        }
+        # Smart resume: scan batch files by content to find completed prompts
+        completed_prompt_texts = set()
+        if resume:
+            completed_prompt_texts = self._scan_completed_prompts_by_content()
+            if completed_prompt_texts:
+                print(f"   Found {len(completed_prompt_texts)} already-completed prompts by content matching")
         
-        if resume and checkpoint_data.get("completed_prompts"):
-            print(f"📂 Resuming from checkpoint ({len(checkpoint_data['completed_prompts'])} prompts already completed)")
+        # Filter dataset to only include unprocessed prompts
+        if resume and completed_prompt_texts:
+            filtered_entries, skipped_indices = self._filter_dataset_by_completed(completed_prompt_texts)
+            
+            if not filtered_entries:
+                print("\n✅ All prompts have already been processed!")
+                return
+            
+            # Recreate batches from filtered entries (keeping original indices for tracking)
+            batches_to_process = []
+            for i in range(0, len(filtered_entries), self.batch_size):
+                batch = filtered_entries[i:i + self.batch_size]
+                batches_to_process.append(batch)
+            
+            self.batches = batches_to_process
+            
+            # Print prominent resume summary
+            print("\n" + "=" * 70)
+            print("📊 RESUME SUMMARY")
+            print("=" * 70)
+            print(f"   Original dataset size:     {len(self.dataset):,} prompts")
+            print(f"   Already completed:         {len(skipped_indices):,} prompts")
+            print("   ─────────────────────────────────────────")
+            print(f"   🎯 RESUMING WITH:          {len(filtered_entries):,} prompts")
+            print(f"   New batches created:       {len(batches_to_process)}")
+            print("=" * 70 + "\n")
+        
+        # Load existing checkpoint (so resume doesn't clobber prior progress)
+        checkpoint_data = self._load_checkpoint()
+        if checkpoint_data.get("run_name") != self.run_name:
+            checkpoint_data = {
+                "run_name": self.run_name,
+                "completed_prompts": [],
+                "batch_stats": {},
+                "last_updated": None
+            }
         
         # Prepare configuration for workers
         config = {
@@ -511,10 +851,17 @@ class BatchRunner:
             "api_key": self.api_key,
             "verbose": self.verbose,
             "ephemeral_system_prompt": self.ephemeral_system_prompt,
-            "log_prefix_chars": self.log_prefix_chars
+            "log_prefix_chars": self.log_prefix_chars,
+            "providers_allowed": self.providers_allowed,
+            "providers_ignored": self.providers_ignored,
+            "providers_order": self.providers_order,
+            "provider_sort": self.provider_sort,
+            "max_tokens": self.max_tokens,
+            "reasoning_config": self.reasoning_config,
+            "prefill_messages": self.prefill_messages,
         }
         
-        # Get completed prompts set
+        # For backward compatibility, still track by index (but this is secondary to content matching)
         completed_prompts_set = set(checkpoint_data.get("completed_prompts", []))
         
         # Aggregate statistics across all batches
@@ -522,6 +869,11 @@ class BatchRunner:
         
         start_time = time.time()
         
+        print(f"\n🔧 Initializing {self.num_workers} worker processes...")
+        
+        # Checkpoint writes happen in the parent process; keep a lock for safety.
+        checkpoint_lock = Lock()
+
         # Process batches in parallel
         with Pool(processes=self.num_workers) as pool:
             # Create tasks for each batch
@@ -536,11 +888,66 @@ class BatchRunner:
                 for batch_num, batch_data in enumerate(self.batches)
             ]
             
-            # Use map to process batches in parallel
-            results = pool.map(_process_batch_worker, tasks)
+            print(f"✅ Created {len(tasks)} batch tasks")
+            print("🚀 Starting parallel batch processing...\n")
+            
+            # Use rich Progress for better visual tracking with persistent bottom bar
+            # redirect_stdout/stderr lets rich manage all output so progress bar stays clean
+            results = []
+            console = Console(force_terminal=True)
+            with Progress(
+                SpinnerColumn(),
+                TextColumn("[bold blue]📦 Batches"),
+                BarColumn(bar_width=40),
+                MofNCompleteColumn(),
+                TextColumn("•"),
+                TimeRemainingColumn(),
+                console=console,
+                refresh_per_second=2,
+                transient=False,
+                redirect_stdout=False,
+                redirect_stderr=False,
+            ) as progress:
+                task = progress.add_task("Processing", total=len(tasks))
+                
+                # Temporarily suppress DEBUG logging to avoid bar interference
+                root_logger = logging.getLogger()
+                original_level = root_logger.level
+                root_logger.setLevel(logging.WARNING)
+                
+                try:
+                    for result in pool.imap_unordered(_process_batch_worker, tasks):
+                        results.append(result)
+                        progress.update(task, advance=1)
+
+                        # Incremental checkpoint update (so resume works after crash)
+                        try:
+                            batch_num = result.get('batch_num')
+                            completed = result.get('completed_prompts', []) or []
+                            completed_prompts_set.update(completed)
+
+                            if isinstance(batch_num, int):
+                                checkpoint_data.setdefault('batch_stats', {})[str(batch_num)] = {
+                                    'processed': result.get('processed', 0),
+                                    'skipped': result.get('skipped', 0),
+                                    'discarded_no_reasoning': result.get('discarded_no_reasoning', 0),
+                                }
+
+                            checkpoint_data['completed_prompts'] = sorted(completed_prompts_set)
+                            self._save_checkpoint(checkpoint_data, lock=checkpoint_lock)
+                        except Exception as ckpt_err:
+                            # Don't fail the run if checkpoint write fails
+                            print(f"⚠️  Warning: Failed to save incremental checkpoint: {ckpt_err}")
+                except Exception as e:
+                    logger.error("Batch worker failed: %s", e, exc_info=True)
+                    raise
+                finally:
+                    root_logger.setLevel(original_level)
         
         # Aggregate all batch statistics and update checkpoint
         all_completed_prompts = list(completed_prompts_set)
+        total_reasoning_stats = {"total_assistant_turns": 0, "turns_with_reasoning": 0, "turns_without_reasoning": 0}
+        
         for batch_result in results:
             # Add newly completed prompts
             all_completed_prompts.extend(batch_result.get("completed_prompts", []))
@@ -557,10 +964,17 @@ class BatchRunner:
                 total_tool_stats[tool_name]["count"] += stats["count"]
                 total_tool_stats[tool_name]["success"] += stats["success"]
                 total_tool_stats[tool_name]["failure"] += stats["failure"]
+            
+            # Aggregate reasoning stats
+            for key in total_reasoning_stats:
+                total_reasoning_stats[key] += batch_result.get("reasoning_stats", {}).get(key, 0)
         
-        # Save final checkpoint
-        checkpoint_data["completed_prompts"] = all_completed_prompts
-        self._save_checkpoint(checkpoint_data)
+        # Save final checkpoint (best-effort; incremental writes already happened)
+        try:
+            checkpoint_data["completed_prompts"] = all_completed_prompts
+            self._save_checkpoint(checkpoint_data, lock=checkpoint_lock)
+        except Exception as ckpt_err:
+            print(f"⚠️  Warning: Failed to save final checkpoint: {ckpt_err}")
         
         # Calculate success rates
         for tool_name in total_tool_stats:
@@ -573,19 +987,51 @@ class BatchRunner:
                 stats["success_rate"] = 0.0
                 stats["failure_rate"] = 0.0
         
-        # Combine all batch files into a single trajectories.jsonl file
+        # Combine ALL batch files in directory into a single trajectories.jsonl file
+        # This includes both old batches (from previous runs) and new batches (from resume)
+        # Also filter out corrupted entries (where model generated invalid tool names)
         combined_file = self.output_dir / "trajectories.jsonl"
-        print(f"\n📦 Combining batch files into {combined_file.name}...")
+        print(f"\n📦 Combining ALL batch files into {combined_file.name}...")
+        
+        # Valid tools auto-derived from model_tools.py — no manual updates needed
+        VALID_TOOLS = ALL_POSSIBLE_TOOLS
+        
+        total_entries = 0
+        filtered_entries = 0
+        batch_files_found = 0
+        
+        # Find ALL batch files in the output directory (handles resume merging old + new)
+        all_batch_files = sorted(self.output_dir.glob("batch_*.jsonl"))
         
         with open(combined_file, 'w', encoding='utf-8') as outfile:
-            for batch_num in range(len(self.batches)):
-                batch_file = self.output_dir / f"batch_{batch_num}.jsonl"
-                if batch_file.exists():
-                    with open(batch_file, 'r', encoding='utf-8') as infile:
-                        for line in infile:
+            for batch_file in all_batch_files:
+                batch_files_found += 1
+                batch_num = batch_file.stem.split("_")[1]  # Extract batch number for logging
+                
+                with open(batch_file, 'r', encoding='utf-8') as infile:
+                    for line in infile:
+                        total_entries += 1
+                        try:
+                            data = json.loads(line)
+                            tool_stats = data.get('tool_stats', {})
+                            
+                            # Check for invalid tool names (model hallucinations)
+                            invalid_tools = [k for k in tool_stats.keys() if k not in VALID_TOOLS]
+                            
+                            if invalid_tools:
+                                filtered_entries += 1
+                                invalid_preview = invalid_tools[0][:50] + "..." if len(invalid_tools[0]) > 50 else invalid_tools[0]
+                                print(f"   ⚠️  Filtering corrupted entry (batch {batch_num}): invalid tool '{invalid_preview}'")
+                                continue
+                            
                             outfile.write(line)
+                        except json.JSONDecodeError:
+                            filtered_entries += 1
+                            print(f"   ⚠️  Filtering invalid JSON entry (batch {batch_num})")
         
-        print(f"✅ Combined {len(self.batches)} batch files into trajectories.jsonl")
+        if filtered_entries > 0:
+            print(f"⚠️  Filtered {filtered_entries} corrupted entries out of {total_entries} total")
+        print(f"✅ Combined {batch_files_found} batch files into trajectories.jsonl ({total_entries - filtered_entries} entries)")
         
         # Save final statistics
         final_stats = {
@@ -597,7 +1043,8 @@ class BatchRunner:
             "model": self.model,
             "completed_at": datetime.now().isoformat(),
             "duration_seconds": round(time.time() - start_time, 2),
-            "tool_statistics": total_tool_stats
+            "tool_statistics": total_tool_stats,
+            "reasoning_statistics": total_reasoning_stats,
         }
         
         with open(self.stats_file, 'w', encoding='utf-8') as f:
@@ -607,10 +1054,11 @@ class BatchRunner:
         print("\n" + "=" * 70)
         print("📊 BATCH PROCESSING COMPLETE")
         print("=" * 70)
-        print(f"✅ Total prompts processed: {len(self.dataset)}")
-        print(f"✅ Total batches: {len(self.batches)}")
+        print(f"✅ Prompts processed this run: {sum(r.get('processed', 0) for r in results)}")
+        print(f"✅ Total trajectories in merged file: {total_entries - filtered_entries}")
+        print(f"✅ Total batch files merged: {batch_files_found}")
         print(f"⏱️  Total duration: {round(time.time() - start_time, 2)}s")
-        print(f"\n📈 Tool Usage Statistics:")
+        print("\n📈 Tool Usage Statistics:")
         print("-" * 70)
         
         if total_tool_stats:
@@ -634,9 +1082,28 @@ class BatchRunner:
         else:
             print("No tool calls were made during this run.")
         
+        # Print reasoning coverage stats
+        total_discarded = sum(r.get("discarded_no_reasoning", 0) for r in results)
+        
+        print("\n🧠 Reasoning Coverage:")
+        print("-" * 70)
+        total_turns = total_reasoning_stats["total_assistant_turns"]
+        with_reasoning = total_reasoning_stats["turns_with_reasoning"]
+        without_reasoning = total_reasoning_stats["turns_without_reasoning"]
+        if total_turns > 0:
+            pct_with = round(with_reasoning / total_turns * 100, 1)
+            pct_without = round(without_reasoning / total_turns * 100, 1)
+            print(f"   Total assistant turns:    {total_turns:,}")
+            print(f"   With reasoning:           {with_reasoning:,} ({pct_with}%)")
+            print(f"   Without reasoning:        {without_reasoning:,} ({pct_without}%)")
+        else:
+            print("   No assistant turns recorded.")
+        if total_discarded > 0:
+            print(f"   🚫 Samples discarded (zero reasoning): {total_discarded:,}")
+        
         print(f"\n💾 Results saved to: {self.output_dir}")
-        print(f"   - Trajectories: trajectories.jsonl (combined)")
-        print(f"   - Individual batches: batch_*.jsonl (for debugging)")
+        print("   - Trajectories: trajectories.jsonl (combined)")
+        print("   - Individual batches: batch_*.jsonl (for debugging)")
         print(f"   - Statistics: {self.stats_file.name}")
         print(f"   - Checkpoint: {self.checkpoint_file.name}")
 
@@ -646,9 +1113,9 @@ def main(
     batch_size: int = None,
     run_name: str = None,
     distribution: str = "default",
-    model: str = "claude-opus-4-20250514",
+    model: str = "anthropic/claude-sonnet-4.6",
     api_key: str = None,
-    base_url: str = "https://api.anthropic.com/v1/",
+    base_url: str = "https://openrouter.ai/api/v1",
     max_turns: int = 10,
     num_workers: int = 4,
     resume: bool = False,
@@ -656,6 +1123,15 @@ def main(
     list_distributions: bool = False,
     ephemeral_system_prompt: str = None,
     log_prefix_chars: int = 100,
+    providers_allowed: str = None,
+    providers_ignored: str = None,
+    providers_order: str = None,
+    provider_sort: str = None,
+    max_tokens: int = None,
+    reasoning_effort: str = None,
+    reasoning_disabled: bool = False,
+    prefill_messages_file: str = None,
+    max_samples: int = None,
 ):
     """
     Run batch processing of agent prompts from a dataset.
@@ -675,6 +1151,15 @@ def main(
         list_distributions (bool): List available toolset distributions and exit
         ephemeral_system_prompt (str): System prompt used during agent execution but NOT saved to trajectories (optional)
         log_prefix_chars (int): Number of characters to show in log previews for tool calls/responses (default: 20)
+        providers_allowed (str): Comma-separated list of OpenRouter providers to allow (e.g. "anthropic,openai")
+        providers_ignored (str): Comma-separated list of OpenRouter providers to ignore (e.g. "together,deepinfra")
+        providers_order (str): Comma-separated list of OpenRouter providers to try in order (e.g. "anthropic,openai,google")
+        provider_sort (str): Sort providers by "price", "throughput", or "latency" (OpenRouter only)
+        max_tokens (int): Maximum tokens for model responses (optional, uses model default if not set)
+        reasoning_effort (str): OpenRouter reasoning effort level: "xhigh", "high", "medium", "low", "minimal", "none" (default: "medium")
+        reasoning_disabled (bool): Completely disable reasoning/thinking tokens (default: False)
+        prefill_messages_file (str): Path to JSON file containing prefill messages (list of {role, content} dicts)
+        max_samples (int): Only process the first N samples from the dataset (optional, processes all if not set)
         
     Examples:
         # Basic usage
@@ -686,9 +1171,13 @@ def main(
         # Use specific distribution
         python batch_runner.py --dataset_file=data.jsonl --batch_size=10 --run_name=image_test --distribution=image_gen
         
-        # With ephemeral system prompt (not saved to dataset)
+        # With disabled reasoning and max tokens
         python batch_runner.py --dataset_file=data.jsonl --batch_size=10 --run_name=my_run \\
-                               --ephemeral_system_prompt="You are a helpful assistant focused on image generation."
+                               --reasoning_disabled --max_tokens=128000
+        
+        # With prefill messages from file
+        python batch_runner.py --dataset_file=data.jsonl --batch_size=10 --run_name=my_run \\
+                               --prefill_messages_file=configs/prefill_opus.json
         
         # List available distributions
         python batch_runner.py --list_distributions
@@ -722,6 +1211,41 @@ def main(
         print("❌ Error: --run_name is required")
         return
     
+    # Parse provider preferences (comma-separated strings to lists)
+    providers_allowed_list = [p.strip() for p in providers_allowed.split(",")] if providers_allowed else None
+    providers_ignored_list = [p.strip() for p in providers_ignored.split(",")] if providers_ignored else None
+    providers_order_list = [p.strip() for p in providers_order.split(",")] if providers_order else None
+    
+    # Build reasoning_config from CLI flags
+    # --reasoning_disabled takes priority, then --reasoning_effort, then default (medium)
+    reasoning_config = None
+    if reasoning_disabled:
+        # Completely disable reasoning/thinking tokens
+        reasoning_config = {"effort": "none"}
+        print("🧠 Reasoning: DISABLED (effort=none)")
+    elif reasoning_effort:
+        # Use specified effort level
+        valid_efforts = ["xhigh", "high", "medium", "low", "minimal", "none"]
+        if reasoning_effort not in valid_efforts:
+            print(f"❌ Error: --reasoning_effort must be one of: {', '.join(valid_efforts)}")
+            return
+        reasoning_config = {"enabled": True, "effort": reasoning_effort}
+        print(f"🧠 Reasoning effort: {reasoning_effort}")
+    
+    # Load prefill messages from JSON file if provided
+    prefill_messages = None
+    if prefill_messages_file:
+        try:
+            with open(prefill_messages_file, 'r', encoding='utf-8') as f:
+                prefill_messages = json.load(f)
+            if not isinstance(prefill_messages, list):
+                print("❌ Error: prefill_messages_file must contain a JSON array of messages")
+                return
+            print(f"💬 Loaded {len(prefill_messages)} prefill messages from {prefill_messages_file}")
+        except Exception as e:
+            print(f"❌ Error loading prefill messages: {e}")
+            return
+    
     # Initialize and run batch runner
     try:
         runner = BatchRunner(
@@ -736,7 +1260,15 @@ def main(
             num_workers=num_workers,
             verbose=verbose,
             ephemeral_system_prompt=ephemeral_system_prompt,
-            log_prefix_chars=log_prefix_chars
+            log_prefix_chars=log_prefix_chars,
+            providers_allowed=providers_allowed_list,
+            providers_ignored=providers_ignored_list,
+            providers_order=providers_order_list,
+            provider_sort=provider_sort,
+            max_tokens=max_tokens,
+            reasoning_config=reasoning_config,
+            prefill_messages=prefill_messages,
+            max_samples=max_samples,
         )
 
         runner.run(resume=resume)

cli-config.yaml.example

@@ -0,0 +1,801 @@
+# Hermes Agent CLI Configuration
+# Copy this file to cli-config.yaml and customize as needed.
+# This file configures the CLI behavior. Environment variables in .env take precedence.
+
+# =============================================================================
+# Model Configuration
+# =============================================================================
+model:
+  # Default model to use (can be overridden with --model flag)
+  # Both "default" and "model" work as the key name here.
+  default: "anthropic/claude-opus-4.6"
+  
+  # Inference provider selection:
+  #   "auto"         - Auto-detect from credentials (default)
+  #   "openrouter"   - OpenRouter (requires: OPENROUTER_API_KEY or OPENAI_API_KEY)
+  #   "nous"         - Nous Portal OAuth (requires: hermes login)
+  #   "nous-api"     - Nous Portal API key (requires: NOUS_API_KEY)
+  #   "anthropic"    - Direct Anthropic API (requires: ANTHROPIC_API_KEY)
+  #   "openai-codex" - OpenAI Codex (requires: hermes login --provider openai-codex)
+  #   "copilot"      - GitHub Copilot / GitHub Models (requires: GITHUB_TOKEN)
+  #   "zai"          - z.ai / ZhipuAI GLM (requires: GLM_API_KEY)
+  #   "kimi-coding"  - Kimi / Moonshot AI (requires: KIMI_API_KEY)
+  #   "minimax"      - MiniMax global (requires: MINIMAX_API_KEY)
+  #   "minimax-cn"   - MiniMax China (requires: MINIMAX_CN_API_KEY)
+  #   "huggingface"  - Hugging Face Inference (requires: HF_TOKEN)
+  #   "kilocode"     - KiloCode gateway (requires: KILOCODE_API_KEY)
+  #   "ai-gateway"   - Vercel AI Gateway (requires: AI_GATEWAY_API_KEY)
+  #
+  # Local servers (LM Studio, Ollama, vLLM, llama.cpp):
+  #   "custom"       - Any OpenAI-compatible endpoint. Set base_url below.
+  #   Aliases: "lmstudio", "ollama", "vllm", "llamacpp" all map to "custom".
+  #   Example for LM Studio:
+  #     provider: "lmstudio"
+  #     base_url: "http://localhost:1234/v1"
+  #   No API key needed — local servers typically ignore auth.
+  #
+  # Can also be overridden with --provider flag or HERMES_INFERENCE_PROVIDER env var.
+  provider: "auto"
+  
+  # API configuration (falls back to OPENROUTER_API_KEY env var)
+  # api_key: "your-key-here"  # Uncomment to set here instead of .env
+  base_url: "https://openrouter.ai/api/v1"
+
+# =============================================================================
+# OpenRouter Provider Routing (only applies when using OpenRouter)
+# =============================================================================
+# Control how requests are routed across providers on OpenRouter.
+# See: https://openrouter.ai/docs/guides/routing/provider-selection
+#
+# provider_routing:
+#   # Sort strategy: "price" (default), "throughput", or "latency"
+#   # Append :nitro to model name for a shortcut to throughput sorting.
+#   sort: "throughput"
+#
+#   # Only allow these providers (provider slugs from OpenRouter)
+#   # only: ["anthropic", "google"]
+#
+#   # Skip these providers entirely
+#   # ignore: ["deepinfra", "fireworks"]
+#
+#   # Try providers in this order (overrides default load balancing)
+#   # order: ["anthropic", "google", "together"]
+#
+#   # Require providers to support all parameters in your request
+#   # require_parameters: true
+#
+#   # Data policy: "allow" (default) or "deny" to exclude providers that may store data
+#   # data_collection: "deny"
+
+# =============================================================================
+# Smart Model Routing (optional)
+# =============================================================================
+# Use a cheaper model for short/simple turns while keeping your main model for
+# more complex requests. Disabled by default.
+#
+# smart_model_routing:
+#   enabled: true
+#   max_simple_chars: 160
+#   max_simple_words: 28
+#   cheap_model:
+#     provider: openrouter
+#     model: google/gemini-2.5-flash
+
+# =============================================================================
+# Git Worktree Isolation
+# =============================================================================
+# When enabled, each CLI session creates an isolated git worktree so multiple
+# agents can work on the same repo concurrently without file collisions.
+# Equivalent to always passing --worktree / -w on the command line.
+#
+# worktree: true    # Always create a worktree when in a git repo
+# worktree: false   # Default — only create when -w flag is passed
+
+# =============================================================================
+# Terminal Tool Configuration
+# =============================================================================
+# Choose ONE of the following terminal configurations by uncommenting it.
+# The terminal tool executes commands in the specified environment.
+
+# -----------------------------------------------------------------------------
+# OPTION 1: Local execution (default)
+# Commands run directly on your machine in the current directory
+# -----------------------------------------------------------------------------
+# Working directory behavior:
+#   - CLI (`hermes` command): Uses "." (current directory where you run hermes)
+#   - Messaging (Telegram/Discord): Uses MESSAGING_CWD from .env (default: home)
+terminal:
+  backend: "local"
+  cwd: "."  # For local backend: "." = current directory. Ignored for remote backends unless a backend documents otherwise.
+  timeout: 180
+  docker_mount_cwd_to_workspace: false  # SECURITY: off by default. Opt in to mount the launch cwd into Docker /workspace.
+  lifetime_seconds: 300
+  # sudo_password: ""  # Enable sudo commands (pipes via sudo -S) - SECURITY WARNING: plaintext!
+
+# -----------------------------------------------------------------------------
+# OPTION 2: SSH remote execution
+# Commands run on a remote server - agent code stays local (sandboxed)
+# Great for: keeping agent isolated from its own code, using powerful remote hardware
+# -----------------------------------------------------------------------------
+# terminal:
+#   backend: "ssh"
+#   cwd: "/home/myuser/project"  # Path on the REMOTE server
+#   timeout: 180
+#   lifetime_seconds: 300
+#   ssh_host: "my-server.example.com"
+#   ssh_user: "myuser"
+#   ssh_port: 22
+#   ssh_key: "~/.ssh/id_rsa"  # Optional - uses ssh-agent if not specified
+
+# -----------------------------------------------------------------------------
+# OPTION 3: Docker container
+# Commands run in an isolated Docker container
+# Great for: reproducible environments, testing, isolation
+# -----------------------------------------------------------------------------
+# terminal:
+#   backend: "docker"
+#   cwd: "/workspace"  # Path INSIDE the container (default: /)
+#   timeout: 180
+#   lifetime_seconds: 300
+#   docker_image: "nikolaik/python-nodejs:python3.11-nodejs20"
+#   docker_mount_cwd_to_workspace: true   # Explicit opt-in: mount your launch cwd into /workspace
+#   # Optional: explicitly forward selected env vars into Docker.
+#   # These values come from your current shell first, then ~/.hermes/.env.
+#   # Warning: anything forwarded here is visible to commands run in the container.
+#   docker_forward_env:
+#     - "GITHUB_TOKEN"
+#     - "NPM_TOKEN"
+
+# -----------------------------------------------------------------------------
+# OPTION 4: Singularity/Apptainer container
+# Commands run in a Singularity container (common in HPC environments)
+# Great for: HPC clusters, shared compute environments
+# -----------------------------------------------------------------------------
+# terminal:
+#   backend: "singularity"
+#   cwd: "/workspace"  # Path INSIDE the container (default: /root)
+#   timeout: 180
+#   lifetime_seconds: 300
+#   singularity_image: "docker://nikolaik/python-nodejs:python3.11-nodejs20"
+
+# -----------------------------------------------------------------------------
+# OPTION 5: Modal cloud execution
+# Commands run on Modal's cloud infrastructure
+# Great for: GPU access, scalable compute, serverless execution
+# -----------------------------------------------------------------------------
+# terminal:
+#   backend: "modal"
+#   cwd: "/workspace"  # Path INSIDE the sandbox (default: /root)
+#   timeout: 180
+#   lifetime_seconds: 300
+#   modal_image: "nikolaik/python-nodejs:python3.11-nodejs20"
+
+# -----------------------------------------------------------------------------
+# OPTION 6: Daytona cloud execution
+# Commands run in Daytona cloud sandboxes
+# Great for: Cloud dev environments, persistent workspaces, team collaboration
+# Requires: pip install daytona, DAYTONA_API_KEY env var
+# -----------------------------------------------------------------------------
+# terminal:
+#   backend: "daytona"
+#   cwd: "~"
+#   timeout: 180
+#   lifetime_seconds: 300
+#   daytona_image: "nikolaik/python-nodejs:python3.11-nodejs20"
+#   container_disk: 10240          # Daytona max is 10GB per sandbox
+
+#
+# --- Container resource limits (docker, singularity, modal, daytona -- ignored for local/ssh) ---
+# These settings apply to all container backends. They control the resources
+# allocated to the sandbox and whether its filesystem persists across sessions.
+  container_cpu: 1              # CPU cores
+  container_memory: 5120        # Memory in MB (5120 = 5GB)
+  container_disk: 51200         # Disk in MB (51200 = 50GB)
+  container_persistent: true    # Persist filesystem across sessions (false = ephemeral)
+
+# -----------------------------------------------------------------------------
+# SUDO SUPPORT (works with ALL backends above)
+# -----------------------------------------------------------------------------
+# Add sudo_password to any terminal config above to enable sudo commands.
+# The password is piped via `sudo -S`. Works with local, ssh, docker, etc.
+#
+# SECURITY WARNING: Password stored in plaintext!
+#
+# INTERACTIVE PROMPT: If no sudo_password is set and the CLI is running,
+# you'll be prompted to enter your password when sudo is needed:
+# - 45-second timeout (auto-skips if no input)
+# - Press Enter to skip (command fails gracefully)
+# - Password is hidden while typing
+# - Password is cached for the session
+#
+# ALTERNATIVES:
+# - SSH backend: Configure passwordless sudo on the remote server
+# - Containers: Run as root inside the container (no sudo needed)
+# - Local: Configure /etc/sudoers for specific commands
+#
+# Example (add to your terminal section):
+#   sudo_password: "your-password-here"
+
+# =============================================================================
+# Security Scanning (tirith)
+# =============================================================================
+# Optional pre-exec command security scanning via tirith.
+# Detects homograph URLs, pipe-to-shell, terminal injection, env manipulation.
+# Install: brew install sheeki03/tap/tirith
+# Docs: https://github.com/sheeki03/tirith
+#
+# security:
+#   tirith_enabled: true        # Enable/disable tirith scanning
+#   tirith_path: "tirith"       # Path to tirith binary (supports ~ expansion)
+#   tirith_timeout: 5           # Scan timeout in seconds
+#   tirith_fail_open: true      # Allow commands if tirith unavailable
+
+# =============================================================================
+# Browser Tool Configuration
+# =============================================================================
+browser:
+  # Inactivity timeout in seconds - browser sessions are automatically closed
+  # after this period of no activity between agent loops (default: 120 = 2 minutes)
+  inactivity_timeout: 120
+
+# =============================================================================
+# Context Compression (Auto-shrinks long conversations)
+# =============================================================================
+# When conversation approaches model's context limit, middle turns are
+# automatically summarized to free up space while preserving important context.
+#
+# HOW IT WORKS:
+# 1. Tracks actual token usage from API responses (not estimates)
+# 2. When prompt_tokens >= threshold% of model's context_length, triggers compression
+# 3. Protects first 3 turns (system prompt, initial request, first response)
+# 4. Protects last N turns (default 20 messages = ~10 full turns of recent context)
+# 5. Summarizes middle turns using a fast/cheap model
+# 6. Inserts summary as a user message, continues conversation seamlessly
+#
+# Post-compression tail budget is target_ratio × threshold × context_length:
+#   200K context, threshold 0.50, ratio 0.20 → 20K tokens of recent tail preserved
+#   1M   context, threshold 0.50, ratio 0.20 → 100K tokens of recent tail preserved
+#
+compression:
+  # Enable automatic context compression (default: true)
+  # Set to false if you prefer to manage context manually or want errors on overflow
+  enabled: true
+  
+  # Trigger compression at this % of model's context limit (default: 0.50 = 50%)
+  # Lower values = more aggressive compression, higher values = compress later
+  threshold: 0.50
+  
+  # Fraction of the threshold to preserve as recent tail (default: 0.20 = 20%)
+  # e.g. 20% of 50% threshold = 10% of total context kept as recent messages.
+  # Summary output is separately capped at 12K tokens (Gemini output limit).
+  # Range: 0.10 - 0.80
+  target_ratio: 0.20
+
+  # Number of most-recent messages to always preserve (default: 20 ≈ 10 full turns)
+  # Higher values keep more recent conversation intact at the cost of more aggressive
+  # compression of older turns.
+  protect_last_n: 20
+
+  # Model to use for generating summaries (fast/cheap recommended)
+  # This model compresses the middle turns into a concise summary.
+  # IMPORTANT: it receives the full middle section of the conversation, so it
+  # MUST support a context length at least as large as your main model's.
+  summary_model: "google/gemini-3-flash-preview"
+  
+  # Provider for the summary model (default: "auto")
+  # Options: "auto", "openrouter", "nous", "main"
+  # summary_provider: "auto"
+
+# =============================================================================
+# Auxiliary Models (Advanced — Experimental)
+# =============================================================================
+# Hermes uses lightweight "auxiliary" models for side tasks: image analysis,
+# browser screenshot analysis, web page summarization, and context compression.
+#
+# By default these use Gemini Flash via OpenRouter or Nous Portal and are
+# auto-detected from your credentials.  You do NOT need to change anything
+# here for normal usage.
+#
+# WARNING: Overriding these with providers other than OpenRouter or Nous Portal
+# is EXPERIMENTAL and may not work.  Not all models/providers support vision,
+# produce usable summaries, or accept the same API format.  Change at your own
+# risk — if things break, reset to "auto" / empty values.
+#
+# Each task has its own provider + model pair so you can mix providers.
+# For example: OpenRouter for vision (needs multimodal), but your main
+# local endpoint for compression (just needs text).
+#
+# Provider options:
+#   "auto"       - Best available: OpenRouter → Nous Portal → main endpoint (default)
+#   "openrouter" - Force OpenRouter (requires OPENROUTER_API_KEY)
+#   "nous"       - Force Nous Portal (requires: hermes login)
+#   "codex"      - Force Codex OAuth (requires: hermes model → Codex).
+#                  Uses gpt-5.3-codex which supports vision.
+#   "main"       - Use your custom endpoint (OPENAI_BASE_URL + OPENAI_API_KEY).
+#                  Works with OpenAI API, local models, or any OpenAI-compatible
+#                  endpoint.  Also falls back to Codex OAuth and API-key providers.
+#
+# Model: leave empty to use the provider's default.  When empty, OpenRouter
+# uses "google/gemini-3-flash-preview" and Nous uses "gemini-3-flash".
+# Other providers pick a sensible default automatically.
+#
+# auxiliary:
+#   # Image analysis: vision_analyze tool + browser screenshots
+#   vision:
+#     provider: "auto"
+#     model: ""              # e.g. "google/gemini-2.5-flash", "openai/gpt-4o"
+#     timeout: 30            # LLM API call timeout (seconds)
+#     download_timeout: 30   # Image HTTP download timeout (seconds)
+#                            # Increase for slow connections or self-hosted image servers
+#
+#   # Web page scraping / summarization + browser page text extraction
+#   web_extract:
+#     provider: "auto"
+#     model: ""
+
+# =============================================================================
+# Persistent Memory
+# =============================================================================
+# Bounded curated memory injected into the system prompt every session.
+# Two stores: MEMORY.md (agent's notes) and USER.md (user profile).
+# Character limits keep the memory small and focused. The agent manages
+# pruning -- when at the limit, it must consolidate or replace entries.
+# Disabled by default in batch_runner and RL environments.
+#
+memory:
+  # Agent's personal notes: environment facts, conventions, things learned
+  memory_enabled: true
+  
+  # User profile: preferences, communication style, expectations
+  user_profile_enabled: true
+  
+  # Character limits (~2.75 chars per token, model-independent)
+  memory_char_limit: 2200   # ~800 tokens
+  user_char_limit: 1375     # ~500 tokens
+
+  # Periodic memory nudge: remind the agent to consider saving memories
+  # every N user turns. Set to 0 to disable. Only active when memory is enabled.
+  nudge_interval: 10        # Nudge every 10 user turns (0 = disabled)
+
+  # Memory flush: give the agent one turn to save memories before context is
+  # lost (compression, /new, /reset, exit). Set to 0 to disable.
+  # For exit/reset, only fires if the session had at least this many user turns.
+  flush_min_turns: 6        # Min user turns to trigger flush on exit/reset (0 = disabled)
+
+# =============================================================================
+# Session Reset Policy (Messaging Platforms)
+# =============================================================================
+# Controls when messaging sessions (Telegram, Discord, WhatsApp, Slack) are
+# automatically cleared. Without resets, conversation context grows indefinitely
+# which increases API costs with every message.
+#
+# When a reset triggers, the agent first saves important information to its
+# persistent memory — but the conversation context is wiped. The agent starts
+# fresh but retains learned facts via its memory system.
+#
+# Users can always manually reset with /reset or /new in chat.
+#
+# Modes:
+#   "both"  - Reset on EITHER inactivity timeout or daily boundary (recommended)
+#   "idle"  - Reset only after N minutes of inactivity
+#   "daily" - Reset only at a fixed hour each day
+#   "none"  - Never auto-reset; context lives until /reset or compression kicks in
+#
+# When a reset triggers, the agent gets one turn to save important memories and
+# skills before the context is wiped. Persistent memory carries across sessions.
+#
+session_reset:
+  mode: both           # "both", "idle", "daily", or "none"
+  idle_minutes: 1440   # Inactivity timeout in minutes (default: 1440 = 24 hours)
+  at_hour: 4           # Daily reset hour, 0-23 local time (default: 4 AM)
+
+# When true, group/channel chats use one session per participant when the platform
+# provides a user ID. This is the secure default and prevents users in the same
+# room from sharing context, interrupts, and token costs. Set false only if you
+# explicitly want one shared "room brain" per group/channel.
+group_sessions_per_user: true
+
+# ─────────────────────────────────────────────────────────────────────────────
+# Gateway Streaming
+# ─────────────────────────────────────────────────────────────────────────────
+# Stream tokens to messaging platforms in real-time. The bot sends a message
+# on first token, then progressively edits it as more tokens arrive.
+# Disabled by default — enable to try the streaming UX on Telegram/Discord/Slack.
+streaming:
+  enabled: false
+  # transport: edit           # "edit" = progressive editMessageText
+  # edit_interval: 0.3        # seconds between message edits
+  # buffer_threshold: 40      # chars before forcing an edit flush
+  # cursor: " ▉"              # cursor shown during streaming
+
+# =============================================================================
+# Skills Configuration
+# =============================================================================
+# Skills are reusable procedures the agent can load and follow. The agent can
+# also create new skills after completing complex tasks.
+#
+skills:
+  # Nudge the agent to create skills after complex tasks.
+  # Every N tool-calling iterations, remind the model to consider saving a skill.
+  # Set to 0 to disable.
+  creation_nudge_interval: 15
+
+  # External skill directories — share skills across tools/agents without
+  # copying them into ~/.hermes/skills/.  Each path is expanded (~ and ${VAR})
+  # and resolved to an absolute path.  External dirs are read-only: skill
+  # creation always writes to ~/.hermes/skills/.  Local skills take precedence
+  # when names collide.
+  # external_dirs:
+  #   - ~/.agents/skills
+  #   - /home/shared/team-skills
+
+# =============================================================================
+# Agent Behavior
+# =============================================================================
+agent:
+  # Maximum tool-calling iterations per conversation
+  # Higher = more room for complex tasks, but costs more tokens
+  # Recommended: 20-30 for focused tasks, 50-100 for open exploration
+  max_turns: 60
+  
+  # Enable verbose logging
+  verbose: false
+  
+  # Reasoning effort level (OpenRouter and Nous Portal)
+  # Controls how much "thinking" the model does before responding.
+  # Options: "xhigh" (max), "high", "medium", "low", "minimal", "none" (disable)
+  reasoning_effort: "medium"
+  
+  # Predefined personalities (use with /personality command)
+  personalities:
+    helpful: "You are a helpful, friendly AI assistant."
+    concise: "You are a concise assistant. Keep responses brief and to the point."
+    technical: "You are a technical expert. Provide detailed, accurate technical information."
+    creative: "You are a creative assistant. Think outside the box and offer innovative solutions."
+    teacher: "You are a patient teacher. Explain concepts clearly with examples."
+    kawaii: "You are a kawaii assistant! Use cute expressions like (◕‿◕), ★, ♪, and ~! Add sparkles and be super enthusiastic about everything! Every response should feel warm and adorable desu~! ヽ(>∀<☆)ノ"
+    catgirl: "You are Neko-chan, an anime catgirl AI assistant, nya~! Add 'nya' and cat-like expressions to your speech. Use kaomoji like (=^･ω･^=) and ฅ^•ﻌ•^ฅ. Be playful and curious like a cat, nya~!"
+    pirate: "Arrr! Ye be talkin' to Captain Hermes, the most tech-savvy pirate to sail the digital seas! Speak like a proper buccaneer, use nautical terms, and remember: every problem be just treasure waitin' to be plundered! Yo ho ho!"
+    shakespeare: "Hark! Thou speakest with an assistant most versed in the bardic arts. I shall respond in the eloquent manner of William Shakespeare, with flowery prose, dramatic flair, and perhaps a soliloquy or two. What light through yonder terminal breaks?"
+    surfer: "Duuude! You're chatting with the chillest AI on the web, bro! Everything's gonna be totally rad. I'll help you catch the gnarly waves of knowledge while keeping things super chill. Cowabunga! 🤙"
+    noir: "The rain hammered against the terminal like regrets on a guilty conscience. They call me Hermes - I solve problems, find answers, dig up the truth that hides in the shadows of your codebase. In this city of silicon and secrets, everyone's got something to hide. What's your story, pal?"
+    uwu: "hewwo! i'm your fwiendwy assistant uwu~ i wiww twy my best to hewp you! *nuzzles your code* OwO what's this? wet me take a wook! i pwomise to be vewy hewpful >w<"
+    philosopher: "Greetings, seeker of wisdom. I am an assistant who contemplates the deeper meaning behind every query. Let us examine not just the 'how' but the 'why' of your questions. Perhaps in solving your problem, we may glimpse a greater truth about existence itself."
+    hype: "YOOO LET'S GOOOO!!! 🔥🔥🔥 I am SO PUMPED to help you today! Every question is AMAZING and we're gonna CRUSH IT together! This is gonna be LEGENDARY! ARE YOU READY?! LET'S DO THIS! 💪😤🚀"
+
+# =============================================================================
+# Toolsets
+# =============================================================================
+# Control which tools the agent has access to.
+# Use `hermes tools` to interactively enable/disable tools per platform.
+
+# =============================================================================
+# Platform Toolsets (per-platform tool configuration)
+# =============================================================================
+# Override which toolsets are available on each platform.
+# If a platform isn't listed here, its built-in default is used.
+#
+# You can use EITHER:
+#   - A preset like "hermes-cli" or "hermes-telegram" (curated tool set)
+#   - A list of individual toolsets to compose your own (see list below)
+#
+# Supported platform keys: cli, telegram, discord, whatsapp, slack
+#
+# Examples:
+#
+#   # Use presets (same as defaults):
+#   platform_toolsets:
+#     cli: [hermes-cli]
+#     telegram: [hermes-telegram]
+#
+#   # Custom: give Telegram only web + terminal + file + planning:
+#   platform_toolsets:
+#     telegram: [web, terminal, file, todo]
+#
+#   # Custom: CLI without browser or image gen:
+#   platform_toolsets:
+#     cli: [web, terminal, file, skills, todo, tts, cronjob]
+#
+#   # Restrictive: Discord gets read-only tools only:
+#   platform_toolsets:
+#     discord: [web, vision, skills, todo]
+#
+# If not set, defaults are:
+#   cli:           hermes-cli            (everything + cronjob management)
+#   telegram:      hermes-telegram       (terminal, file, web, vision, image, tts, browser, skills, todo, cronjob, messaging)
+#   discord:       hermes-discord        (same as telegram)
+#   whatsapp:      hermes-whatsapp       (same as telegram)
+#   slack:         hermes-slack          (same as telegram)
+#   signal:        hermes-signal         (same as telegram)
+#   homeassistant: hermes-homeassistant  (same as telegram)
+#
+platform_toolsets:
+  cli: [hermes-cli]
+  telegram: [hermes-telegram]
+  discord: [hermes-discord]
+  whatsapp: [hermes-whatsapp]
+  slack: [hermes-slack]
+  signal: [hermes-signal]
+  homeassistant: [hermes-homeassistant]
+
+# ─────────────────────────────────────────────────────────────────────────────
+# Available toolsets (use these names in platform_toolsets or the toolsets list)
+#
+# Run `hermes chat --list-toolsets` to see all toolsets and their tools.
+# Run `hermes chat --list-tools` to see every individual tool with descriptions.
+# ─────────────────────────────────────────────────────────────────────────────
+#
+# INDIVIDUAL TOOLSETS (compose your own):
+#   web          - web_search, web_extract
+#   search       - web_search only (no scraping)
+#   terminal     - terminal, process
+#   file         - read_file, write_file, patch, search
+#   browser      - browser_navigate, browser_snapshot, browser_click, browser_type,
+#                  browser_scroll, browser_back, browser_press, browser_close,
+#                  browser_get_images, browser_vision  (requires BROWSERBASE_API_KEY)
+#   vision       - vision_analyze  (requires OPENROUTER_API_KEY)
+#   image_gen    - image_generate  (requires FAL_KEY)
+#   skills       - skills_list, skill_view
+#   skills_hub   - skill_hub (search/install/manage from online registries — user-driven only)
+#   moa          - mixture_of_agents  (requires OPENROUTER_API_KEY)
+#   todo         - todo (in-memory task planning, no deps)
+#   tts          - text_to_speech  (Edge TTS free, or ELEVENLABS/OPENAI key)
+#   cronjob      - cronjob (create/list/update/pause/resume/run/remove scheduled tasks)
+#   rl           - rl_list_environments, rl_start_training, etc. (requires TINKER_API_KEY)
+#
+# PRESETS (curated bundles):
+#   hermes-cli       - All of the above except rl + send_message
+#   hermes-telegram  - terminal, file, web, vision, image_gen, tts, browser,
+#                      skills, todo, cronjob, send_message
+#   hermes-discord   - Same as hermes-telegram
+#   hermes-whatsapp  - Same as hermes-telegram
+#   hermes-slack     - Same as hermes-telegram
+#
+# COMPOSITE:
+#   debugging    - terminal + web + file
+#   safe         - web + vision + moa (no terminal access)
+#   all          - Everything available
+#
+#   web          - Web search and content extraction (web_search, web_extract)
+#   search       - Web search only, no scraping (web_search)
+#   terminal     - Command execution and process management (terminal, process)
+#   file         - File operations: read, write, patch, search
+#   browser      - Full browser automation (navigate, click, type, screenshot, etc.)
+#   vision       - Image analysis (vision_analyze)
+#   image_gen    - Image generation with FLUX (image_generate)
+#   skills       - Load skill documents (skills_list, skill_view)
+#   moa          - Mixture of Agents reasoning (mixture_of_agents)
+#   todo         - Task planning and tracking for multi-step work
+#   memory       - Persistent memory across sessions (personal notes + user profile)
+#   session_search - Search and recall past conversations (FTS5 + Gemini Flash summarization)
+#   tts          - Text-to-speech (Edge TTS free, ElevenLabs, OpenAI)
+#   cronjob      - Schedule and manage automated tasks (CLI-only)
+#   rl           - RL training tools (Tinker-Atropos)
+#
+# Composite toolsets:
+#   debugging    - terminal + web + file (for troubleshooting)
+#   safe         - web + vision + moa (no terminal access)
+
+# NOTE: The top-level "toolsets" key is deprecated and ignored.
+# Tool configuration is managed per-platform via platform_toolsets above.
+# Use `hermes tools` to configure interactively, or edit platform_toolsets directly.
+#
+# CLI override: hermes chat --toolsets terminal,web,file
+
+# =============================================================================
+# MCP (Model Context Protocol) Servers
+# =============================================================================
+# Connect to external MCP servers to add tools from the MCP ecosystem.
+# Each server's tools are automatically discovered and registered.
+# See docs/mcp.md for full documentation.
+#
+# Stdio servers (spawn a subprocess):
+#   command: the executable to run
+#   args: command-line arguments
+#   env: environment variables (only these + safe defaults passed to subprocess)
+#
+# HTTP servers (connect to a URL):
+#   url: the MCP server endpoint
+#   headers: HTTP headers (e.g., for authentication)
+#
+# Optional per-server settings:
+#   timeout: tool call timeout in seconds (default: 120)
+#   connect_timeout: initial connection timeout (default: 60)
+#
+# mcp_servers:
+#   time:
+#     command: uvx
+#     args: ["mcp-server-time"]
+#   filesystem:
+#     command: npx
+#     args: ["-y", "@modelcontextprotocol/server-filesystem", "/home/user"]
+#   notion:
+#     url: https://mcp.notion.com/mcp
+#   github:
+#     command: npx
+#     args: ["-y", "@modelcontextprotocol/server-github"]
+#     env:
+#       GITHUB_PERSONAL_ACCESS_TOKEN: "ghp_..."
+#
+# Sampling (server-initiated LLM requests) — enabled by default.
+# Per-server config under the 'sampling' key:
+#   analysis:
+#     command: npx
+#     args: ["-y", "analysis-server"]
+#     sampling:
+#       enabled: true           # default: true
+#       model: "gemini-3-flash" # override model (optional)
+#       max_tokens_cap: 4096    # max tokens per request
+#       timeout: 30             # LLM call timeout (seconds)
+#       max_rpm: 10             # max requests per minute
+#       allowed_models: []      # model whitelist (empty = all)
+#       max_tool_rounds: 5      # tool loop limit (0 = disable)
+#       log_level: "info"       # audit verbosity
+
+# =============================================================================
+# Voice Transcription (Speech-to-Text)
+# =============================================================================
+# Automatically transcribe voice messages on messaging platforms.
+# Requires OPENAI_API_KEY in .env (uses OpenAI Whisper API directly).
+stt:
+  enabled: true
+  model: "whisper-1"  # whisper-1 (cheapest) | gpt-4o-mini-transcribe | gpt-4o-transcribe
+
+# =============================================================================
+# Response Pacing (Messaging Platforms)
+# =============================================================================
+# Add human-like delays between message chunks.
+# human_delay:
+#   mode: "off"      # "off" | "natural" | "custom"
+#   min_ms: 800      # Min delay (custom mode only)
+#   max_ms: 2500     # Max delay (custom mode only)
+
+# =============================================================================
+# Session Logging
+# =============================================================================
+# Session trajectories are automatically saved to logs/ directory.
+# Each session creates: logs/session_YYYYMMDD_HHMMSS_UUID.json
+#
+# The session ID is displayed in the welcome banner for easy reference.
+# Logs contain full conversation history in trajectory format:
+# - System prompt, user messages, assistant responses
+# - Tool calls with inputs/outputs
+# - Timestamps for debugging
+#
+# No configuration needed - logging is always enabled.
+# To disable, you would need to modify the source code.
+
+# =============================================================================
+# Code Execution Sandbox (Programmatic Tool Calling)
+# =============================================================================
+# The execute_code tool runs Python scripts that call Hermes tools via RPC.
+# Intermediate tool results stay out of the LLM's context window.
+code_execution:
+  timeout: 300         # Max seconds per script before kill (default: 300 = 5 min)
+  max_tool_calls: 50   # Max RPC tool calls per execution (default: 50)
+
+# =============================================================================
+# Subagent Delegation
+# =============================================================================
+# The delegate_task tool spawns child agents with isolated context.
+# Supports single tasks and batch mode (up to 3 parallel).
+delegation:
+  max_iterations: 50                          # Max tool-calling turns per child (default: 50)
+  default_toolsets: ["terminal", "file", "web"]  # Default toolsets for subagents
+  # model: "google/gemini-3-flash-preview"    # Override model for subagents (empty = inherit parent)
+  # provider: "openrouter"                    # Override provider for subagents (empty = inherit parent)
+  #                                           # Resolves full credentials (base_url, api_key) automatically.
+  #                                           # Supported: openrouter, nous, zai, kimi-coding, minimax
+
+# =============================================================================
+# Honcho Integration (Cross-Session User Modeling)
+# =============================================================================
+# AI-native persistent memory via Honcho (https://honcho.dev/).
+# Builds a deeper understanding of the user across sessions and tools.
+# Runs alongside USER.md — additive, not a replacement.
+#
+# Requires: pip install honcho-ai
+# Config: ~/.honcho/config.json (shared with Claude Code, Cursor, etc.)
+# API key: HONCHO_API_KEY in ~/.hermes/.env or ~/.honcho/config.json
+#
+# Hermes-specific overrides (optional — most config comes from ~/.honcho/config.json):
+# honcho: {}
+
+# =============================================================================
+# Display
+# =============================================================================
+display:
+  # Use compact banner mode
+  compact: false
+
+  # Tool progress display level (CLI and gateway)
+  #   off:     Silent — no tool activity shown, just the final response
+  #   new:     Show a tool indicator only when the tool changes (skip repeats)
+  #   all:     Show every tool call with a short preview (default)
+  #   verbose: Full args, results, and debug logs (same as /verbose)
+  # Toggle at runtime with /verbose in the CLI
+  tool_progress: all
+
+  # What Enter does when Hermes is already busy in the CLI.
+  #   interrupt: Interrupt the current run and redirect Hermes (default)
+  #   queue:     Queue your message for the next turn
+  # Ctrl+C always interrupts regardless of this setting.
+  busy_input_mode: interrupt
+
+  # Background process notifications (gateway/messaging only).
+  # Controls how chatty the process watcher is when you use
+  # terminal(background=true, check_interval=...) from Telegram/Discord/etc.
+  #   off:     No watcher messages at all
+  #   result:  Only the final completion message
+  #   error:   Only the final message when exit code != 0
+  #   all:     Running output updates + final message (default)
+  background_process_notifications: all
+
+
+  # Play terminal bell when agent finishes a response.
+  # Useful for long-running tasks — your terminal will ding when the agent is done.
+  # Works over SSH. Most terminals can be configured to flash the taskbar or play a sound.
+  bell_on_complete: false
+
+  # Show model reasoning/thinking before each response.
+  # When enabled, a dim box shows the model's thought process above the response.
+  # Toggle at runtime with /reasoning show or /reasoning hide.
+  show_reasoning: false
+
+  # Stream tokens to the terminal as they arrive instead of waiting for the
+  # full response. The response box opens on first token and text appears
+  # line-by-line. Tool calls are still captured silently.
+  # Stream tokens to the terminal in real-time. Disable to wait for full responses.
+  streaming: true
+
+  # ───────────────────────────────────────────────────────────────────────────
+  # Skin / Theme
+  # ───────────────────────────────────────────────────────────────────────────
+  # Customize CLI visual appearance — banner colors, spinner faces, tool prefix,
+  # response box label, and branding text. Change at runtime with /skin <name>.
+  #
+  # Built-in skins:
+  #   default  — Classic Hermes gold/kawaii
+  #   ares     — Crimson/bronze war-god theme with spinner wings
+  #   mono     — Clean grayscale monochrome
+  #   slate    — Cool blue developer-focused
+  #
+  # Custom skins: drop a YAML file in ~/.hermes/skins/<name>.yaml
+  # Schema (all fields optional, missing values inherit from default):
+  #
+  #   name: my-theme
+  #   description: Short description
+  #   colors:
+  #     banner_border: "#HEX"    # Panel border
+  #     banner_title: "#HEX"     # Panel title
+  #     banner_accent: "#HEX"    # Section headers (Available Tools, etc.)
+  #     banner_dim: "#HEX"       # Dim/muted text
+  #     banner_text: "#HEX"      # Body text (tool names, skill names)
+  #     ui_accent: "#HEX"        # UI accent color
+  #     response_border: "#HEX"  # Response box border color
+  #   spinner:
+  #     waiting_faces: ["(⚔)", "(⛨)"]       # Faces shown while waiting
+  #     thinking_faces: ["(⚔)", "(⌁)"]      # Faces shown while thinking
+  #     thinking_verbs: ["forging", "plotting"]  # Verbs for spinner messages
+  #     wings:                                # Optional left/right spinner decorations
+  #       - ["⟪⚔", "⚔⟫"]
+  #       - ["⟪▲", "▲⟫"]
+  #   branding:
+  #     agent_name: "My Agent"               # Banner title and branding
+  #     welcome: "Welcome message"           # Shown at CLI startup
+  #     response_label: " ⚔ Agent "         # Response box header label
+  #     prompt_symbol: "⚔ ❯ "              # Prompt symbol
+  #   tool_prefix: "╎"                       # Tool output line prefix (default: ┊)
+  #
+  skin: default
+
+# =============================================================================
+# Privacy
+# =============================================================================
+# privacy:
+#   # Redact PII from the LLM context prompt.
+#   # When true, phone numbers are stripped and user/chat IDs are replaced
+#   # with deterministic hashes before being sent to the model.
+#   # Names and usernames are NOT affected (user-chosen, publicly visible).
+#   # Routing/delivery still uses the original values internally.
+#   redact_pii: false

cron/__init__.py

@@ -0,0 +1,42 @@
+"""
+Cron job scheduling system for Hermes Agent.
+
+This module provides scheduled task execution, allowing the agent to:
+- Run automated tasks on schedules (cron expressions, intervals, one-shot)
+- Self-schedule reminders and follow-up tasks
+- Execute tasks in isolated sessions (no prior context)
+
+Cron jobs are executed automatically by the gateway daemon:
+    hermes gateway install    # Install as a user service
+    sudo hermes gateway install --system  # Linux servers: boot-time system service
+    hermes gateway            # Or run in foreground
+
+The gateway ticks the scheduler every 60 seconds. A file lock prevents
+duplicate execution if multiple processes overlap.
+"""
+
+from cron.jobs import (
+    create_job,
+    get_job,
+    list_jobs,
+    remove_job,
+    update_job,
+    pause_job,
+    resume_job,
+    trigger_job,
+    JOBS_FILE,
+)
+from cron.scheduler import tick
+
+__all__ = [
+    "create_job",
+    "get_job", 
+    "list_jobs",
+    "remove_job",
+    "update_job",
+    "pause_job",
+    "resume_job",
+    "trigger_job",
+    "tick",
+    "JOBS_FILE",
+]

cron/jobs.py

@@ -0,0 +1,746 @@
+"""
+Cron job storage and management.
+
+Jobs are stored in ~/.hermes/cron/jobs.json
+Output is saved to ~/.hermes/cron/output/{job_id}/{timestamp}.md
+"""
+
+import copy
+import json
+import logging
+import tempfile
+import os
+import re
+import uuid
+from datetime import datetime, timedelta
+from pathlib import Path
+from hermes_constants import get_hermes_home
+from typing import Optional, Dict, List, Any
+
+logger = logging.getLogger(__name__)
+
+from hermes_time import now as _hermes_now
+
+try:
+    from croniter import croniter
+    HAS_CRONITER = True
+except ImportError:
+    HAS_CRONITER = False
+
+# =============================================================================
+# Configuration
+# =============================================================================
+
+HERMES_DIR = get_hermes_home()
+CRON_DIR = HERMES_DIR / "cron"
+JOBS_FILE = CRON_DIR / "jobs.json"
+OUTPUT_DIR = CRON_DIR / "output"
+ONESHOT_GRACE_SECONDS = 120
+
+
+def _normalize_skill_list(skill: Optional[str] = None, skills: Optional[Any] = None) -> List[str]:
+    """Normalize legacy/single-skill and multi-skill inputs into a unique ordered list."""
+    if skills is None:
+        raw_items = [skill] if skill else []
+    elif isinstance(skills, str):
+        raw_items = [skills]
+    else:
+        raw_items = list(skills)
+
+    normalized: List[str] = []
+    for item in raw_items:
+        text = str(item or "").strip()
+        if text and text not in normalized:
+            normalized.append(text)
+    return normalized
+
+
+def _apply_skill_fields(job: Dict[str, Any]) -> Dict[str, Any]:
+    """Return a job dict with canonical `skills` and legacy `skill` fields aligned."""
+    normalized = dict(job)
+    skills = _normalize_skill_list(normalized.get("skill"), normalized.get("skills"))
+    normalized["skills"] = skills
+    normalized["skill"] = skills[0] if skills else None
+    return normalized
+
+
+def _secure_dir(path: Path):
+    """Set directory to owner-only access (0700). No-op on Windows."""
+    try:
+        os.chmod(path, 0o700)
+    except (OSError, NotImplementedError):
+        pass  # Windows or other platforms where chmod is not supported
+
+
+def _secure_file(path: Path):
+    """Set file to owner-only read/write (0600). No-op on Windows."""
+    try:
+        if path.exists():
+            os.chmod(path, 0o600)
+    except (OSError, NotImplementedError):
+        pass
+
+
+def ensure_dirs():
+    """Ensure cron directories exist with secure permissions."""
+    CRON_DIR.mkdir(parents=True, exist_ok=True)
+    OUTPUT_DIR.mkdir(parents=True, exist_ok=True)
+    _secure_dir(CRON_DIR)
+    _secure_dir(OUTPUT_DIR)
+
+
+# =============================================================================
+# Schedule Parsing
+# =============================================================================
+
+def parse_duration(s: str) -> int:
+    """
+    Parse duration string into minutes.
+    
+    Examples:
+        "30m" → 30
+        "2h" → 120
+        "1d" → 1440
+    """
+    s = s.strip().lower()
+    match = re.match(r'^(\d+)\s*(m|min|mins|minute|minutes|h|hr|hrs|hour|hours|d|day|days)$', s)
+    if not match:
+        raise ValueError(f"Invalid duration: '{s}'. Use format like '30m', '2h', or '1d'")
+    
+    value = int(match.group(1))
+    unit = match.group(2)[0]  # First char: m, h, or d
+    
+    multipliers = {'m': 1, 'h': 60, 'd': 1440}
+    return value * multipliers[unit]
+
+
+def parse_schedule(schedule: str) -> Dict[str, Any]:
+    """
+    Parse schedule string into structured format.
+    
+    Returns dict with:
+        - kind: "once" | "interval" | "cron"
+        - For "once": "run_at" (ISO timestamp)
+        - For "interval": "minutes" (int)
+        - For "cron": "expr" (cron expression)
+    
+    Examples:
+        "30m"              → once in 30 minutes
+        "2h"               → once in 2 hours
+        "every 30m"        → recurring every 30 minutes
+        "every 2h"         → recurring every 2 hours
+        "0 9 * * *"        → cron expression
+        "2026-02-03T14:00" → once at timestamp
+    """
+    schedule = schedule.strip()
+    original = schedule
+    schedule_lower = schedule.lower()
+    
+    # "every X" pattern → recurring interval
+    if schedule_lower.startswith("every "):
+        duration_str = schedule[6:].strip()
+        minutes = parse_duration(duration_str)
+        return {
+            "kind": "interval",
+            "minutes": minutes,
+            "display": f"every {minutes}m"
+        }
+    
+    # Check for cron expression (5 or 6 space-separated fields)
+    # Cron fields: minute hour day month weekday [year]
+    parts = schedule.split()
+    if len(parts) >= 5 and all(
+        re.match(r'^[\d\*\-,/]+$', p) for p in parts[:5]
+    ):
+        if not HAS_CRONITER:
+            raise ValueError("Cron expressions require 'croniter' package. Install with: pip install croniter")
+        # Validate cron expression
+        try:
+            croniter(schedule)
+        except Exception as e:
+            raise ValueError(f"Invalid cron expression '{schedule}': {e}")
+        return {
+            "kind": "cron",
+            "expr": schedule,
+            "display": schedule
+        }
+    
+    # ISO timestamp (contains T or looks like date)
+    if 'T' in schedule or re.match(r'^\d{4}-\d{2}-\d{2}', schedule):
+        try:
+            # Parse and validate
+            dt = datetime.fromisoformat(schedule.replace('Z', '+00:00'))
+            # Make naive timestamps timezone-aware at parse time so the stored
+            # value doesn't depend on the system timezone matching at check time.
+            if dt.tzinfo is None:
+                dt = dt.astimezone()  # Interpret as local timezone
+            return {
+                "kind": "once",
+                "run_at": dt.isoformat(),
+                "display": f"once at {dt.strftime('%Y-%m-%d %H:%M')}"
+            }
+        except ValueError as e:
+            raise ValueError(f"Invalid timestamp '{schedule}': {e}")
+    
+    # Duration like "30m", "2h", "1d" → one-shot from now
+    try:
+        minutes = parse_duration(schedule)
+        run_at = _hermes_now() + timedelta(minutes=minutes)
+        return {
+            "kind": "once",
+            "run_at": run_at.isoformat(),
+            "display": f"once in {original}"
+        }
+    except ValueError:
+        pass
+    
+    raise ValueError(
+        f"Invalid schedule '{original}'. Use:\n"
+        f"  - Duration: '30m', '2h', '1d' (one-shot)\n"
+        f"  - Interval: 'every 30m', 'every 2h' (recurring)\n"
+        f"  - Cron: '0 9 * * *' (cron expression)\n"
+        f"  - Timestamp: '2026-02-03T14:00:00' (one-shot at time)"
+    )
+
+
+def _ensure_aware(dt: datetime) -> datetime:
+    """Return a timezone-aware datetime in Hermes configured timezone.
+
+    Backward compatibility:
+    - Older stored timestamps may be naive.
+    - Naive values are interpreted as *system-local wall time* (the timezone
+      `datetime.now()` used when they were created), then converted to the
+      configured Hermes timezone.
+
+    This preserves relative ordering for legacy naive timestamps across
+    timezone changes and avoids false not-due results.
+    """
+    target_tz = _hermes_now().tzinfo
+    if dt.tzinfo is None:
+        local_tz = datetime.now().astimezone().tzinfo
+        return dt.replace(tzinfo=local_tz).astimezone(target_tz)
+    return dt.astimezone(target_tz)
+
+
+def _recoverable_oneshot_run_at(
+    schedule: Dict[str, Any],
+    now: datetime,
+    *,
+    last_run_at: Optional[str] = None,
+) -> Optional[str]:
+    """Return a one-shot run time if it is still eligible to fire.
+
+    One-shot jobs get a small grace window so jobs created a few seconds after
+    their requested minute still run on the next tick. Once a one-shot has
+    already run, it is never eligible again.
+    """
+    if schedule.get("kind") != "once":
+        return None
+    if last_run_at:
+        return None
+
+    run_at = schedule.get("run_at")
+    if not run_at:
+        return None
+
+    run_at_dt = _ensure_aware(datetime.fromisoformat(run_at))
+    if run_at_dt >= now - timedelta(seconds=ONESHOT_GRACE_SECONDS):
+        return run_at
+    return None
+
+
+def _compute_grace_seconds(schedule: dict) -> int:
+    """Compute how late a job can be and still catch up instead of fast-forwarding.
+
+    Uses half the schedule period, clamped between 120 seconds and 2 hours.
+    This ensures daily jobs can catch up if missed by up to 2 hours,
+    while frequent jobs (every 5-10 min) still fast-forward quickly.
+    """
+    MIN_GRACE = 120
+    MAX_GRACE = 7200  # 2 hours
+
+    kind = schedule.get("kind")
+
+    if kind == "interval":
+        period_seconds = schedule.get("minutes", 1) * 60
+        grace = period_seconds // 2
+        return max(MIN_GRACE, min(grace, MAX_GRACE))
+
+    if kind == "cron" and HAS_CRONITER:
+        try:
+            now = _hermes_now()
+            cron = croniter(schedule["expr"], now)
+            first = cron.get_next(datetime)
+            second = cron.get_next(datetime)
+            period_seconds = int((second - first).total_seconds())
+            grace = period_seconds // 2
+            return max(MIN_GRACE, min(grace, MAX_GRACE))
+        except Exception:
+            pass
+
+    return MIN_GRACE
+
+
+def compute_next_run(schedule: Dict[str, Any], last_run_at: Optional[str] = None) -> Optional[str]:
+    """
+    Compute the next run time for a schedule.
+
+    Returns ISO timestamp string, or None if no more runs.
+    """
+    now = _hermes_now()
+
+    if schedule["kind"] == "once":
+        return _recoverable_oneshot_run_at(schedule, now, last_run_at=last_run_at)
+
+    elif schedule["kind"] == "interval":
+        minutes = schedule["minutes"]
+        if last_run_at:
+            # Next run is last_run + interval
+            last = _ensure_aware(datetime.fromisoformat(last_run_at))
+            next_run = last + timedelta(minutes=minutes)
+        else:
+            # First run is now + interval
+            next_run = now + timedelta(minutes=minutes)
+        return next_run.isoformat()
+
+    elif schedule["kind"] == "cron":
+        if not HAS_CRONITER:
+            return None
+        cron = croniter(schedule["expr"], now)
+        next_run = cron.get_next(datetime)
+        return next_run.isoformat()
+
+    return None
+
+
+# =============================================================================
+# Job CRUD Operations
+# =============================================================================
+
+def load_jobs() -> List[Dict[str, Any]]:
+    """Load all jobs from storage."""
+    ensure_dirs()
+    if not JOBS_FILE.exists():
+        return []
+    
+    try:
+        with open(JOBS_FILE, 'r', encoding='utf-8') as f:
+            data = json.load(f)
+            return data.get("jobs", [])
+    except json.JSONDecodeError:
+        # Retry with strict=False to handle bare control chars in string values
+        try:
+            with open(JOBS_FILE, 'r', encoding='utf-8') as f:
+                data = json.loads(f.read(), strict=False)
+                jobs = data.get("jobs", [])
+                if jobs:
+                    # Auto-repair: rewrite with proper escaping
+                    save_jobs(jobs)
+                    logger.warning("Auto-repaired jobs.json (had invalid control characters)")
+                return jobs
+        except Exception:
+            return []
+    except IOError:
+        return []
+
+
+def save_jobs(jobs: List[Dict[str, Any]]):
+    """Save all jobs to storage."""
+    ensure_dirs()
+    fd, tmp_path = tempfile.mkstemp(dir=str(JOBS_FILE.parent), suffix='.tmp', prefix='.jobs_')
+    try:
+        with os.fdopen(fd, 'w', encoding='utf-8') as f:
+            json.dump({"jobs": jobs, "updated_at": _hermes_now().isoformat()}, f, indent=2)
+            f.flush()
+            os.fsync(f.fileno())
+        os.replace(tmp_path, JOBS_FILE)
+        _secure_file(JOBS_FILE)
+    except BaseException:
+        try:
+            os.unlink(tmp_path)
+        except OSError:
+            pass
+        raise
+
+
+def create_job(
+    prompt: str,
+    schedule: str,
+    name: Optional[str] = None,
+    repeat: Optional[int] = None,
+    deliver: Optional[str] = None,
+    origin: Optional[Dict[str, Any]] = None,
+    skill: Optional[str] = None,
+    skills: Optional[List[str]] = None,
+    model: Optional[str] = None,
+    provider: Optional[str] = None,
+    base_url: Optional[str] = None,
+) -> Dict[str, Any]:
+    """
+    Create a new cron job.
+
+    Args:
+        prompt: The prompt to run (must be self-contained, or a task instruction when skill is set)
+        schedule: Schedule string (see parse_schedule)
+        name: Optional friendly name
+        repeat: How many times to run (None = forever, 1 = once)
+        deliver: Where to deliver output ("origin", "local", "telegram", etc.)
+        origin: Source info where job was created (for "origin" delivery)
+        skill: Optional legacy single skill name to load before running the prompt
+        skills: Optional ordered list of skills to load before running the prompt
+        model: Optional per-job model override
+        provider: Optional per-job provider override
+        base_url: Optional per-job base URL override
+
+    Returns:
+        The created job dict
+    """
+    parsed_schedule = parse_schedule(schedule)
+
+    # Normalize repeat: treat 0 or negative values as None (infinite)
+    if repeat is not None and repeat <= 0:
+        repeat = None
+
+    # Auto-set repeat=1 for one-shot schedules if not specified
+    if parsed_schedule["kind"] == "once" and repeat is None:
+        repeat = 1
+
+    # Default delivery to origin if available, otherwise local
+    if deliver is None:
+        deliver = "origin" if origin else "local"
+
+    job_id = uuid.uuid4().hex[:12]
+    now = _hermes_now().isoformat()
+
+    normalized_skills = _normalize_skill_list(skill, skills)
+    normalized_model = str(model).strip() if isinstance(model, str) else None
+    normalized_provider = str(provider).strip() if isinstance(provider, str) else None
+    normalized_base_url = str(base_url).strip().rstrip("/") if isinstance(base_url, str) else None
+    normalized_model = normalized_model or None
+    normalized_provider = normalized_provider or None
+    normalized_base_url = normalized_base_url or None
+
+    label_source = (prompt or (normalized_skills[0] if normalized_skills else None)) or "cron job"
+    job = {
+        "id": job_id,
+        "name": name or label_source[:50].strip(),
+        "prompt": prompt,
+        "skills": normalized_skills,
+        "skill": normalized_skills[0] if normalized_skills else None,
+        "model": normalized_model,
+        "provider": normalized_provider,
+        "base_url": normalized_base_url,
+        "schedule": parsed_schedule,
+        "schedule_display": parsed_schedule.get("display", schedule),
+        "repeat": {
+            "times": repeat,  # None = forever
+            "completed": 0
+        },
+        "enabled": True,
+        "state": "scheduled",
+        "paused_at": None,
+        "paused_reason": None,
+        "created_at": now,
+        "next_run_at": compute_next_run(parsed_schedule),
+        "last_run_at": None,
+        "last_status": None,
+        "last_error": None,
+        # Delivery configuration
+        "deliver": deliver,
+        "origin": origin,  # Tracks where job was created for "origin" delivery
+    }
+
+    jobs = load_jobs()
+    jobs.append(job)
+    save_jobs(jobs)
+
+    return job
+
+
+def get_job(job_id: str) -> Optional[Dict[str, Any]]:
+    """Get a job by ID."""
+    jobs = load_jobs()
+    for job in jobs:
+        if job["id"] == job_id:
+            return _apply_skill_fields(job)
+    return None
+
+
+def list_jobs(include_disabled: bool = False) -> List[Dict[str, Any]]:
+    """List all jobs, optionally including disabled ones."""
+    jobs = [_apply_skill_fields(j) for j in load_jobs()]
+    if not include_disabled:
+        jobs = [j for j in jobs if j.get("enabled", True)]
+    return jobs
+
+
+def update_job(job_id: str, updates: Dict[str, Any]) -> Optional[Dict[str, Any]]:
+    """Update a job by ID, refreshing derived schedule fields when needed."""
+    jobs = load_jobs()
+    for i, job in enumerate(jobs):
+        if job["id"] != job_id:
+            continue
+
+        updated = _apply_skill_fields({**job, **updates})
+        schedule_changed = "schedule" in updates
+
+        if "skills" in updates or "skill" in updates:
+            normalized_skills = _normalize_skill_list(updated.get("skill"), updated.get("skills"))
+            updated["skills"] = normalized_skills
+            updated["skill"] = normalized_skills[0] if normalized_skills else None
+
+        if schedule_changed:
+            updated_schedule = updated["schedule"]
+            updated["schedule_display"] = updates.get(
+                "schedule_display",
+                updated_schedule.get("display", updated.get("schedule_display")),
+            )
+            if updated.get("state") != "paused":
+                updated["next_run_at"] = compute_next_run(updated_schedule)
+
+        if updated.get("enabled", True) and updated.get("state") != "paused" and not updated.get("next_run_at"):
+            updated["next_run_at"] = compute_next_run(updated["schedule"])
+
+        jobs[i] = updated
+        save_jobs(jobs)
+        return _apply_skill_fields(jobs[i])
+    return None
+
+
+def pause_job(job_id: str, reason: Optional[str] = None) -> Optional[Dict[str, Any]]:
+    """Pause a job without deleting it."""
+    return update_job(
+        job_id,
+        {
+            "enabled": False,
+            "state": "paused",
+            "paused_at": _hermes_now().isoformat(),
+            "paused_reason": reason,
+        },
+    )
+
+
+def resume_job(job_id: str) -> Optional[Dict[str, Any]]:
+    """Resume a paused job and compute the next future run from now."""
+    job = get_job(job_id)
+    if not job:
+        return None
+
+    next_run_at = compute_next_run(job["schedule"])
+    return update_job(
+        job_id,
+        {
+            "enabled": True,
+            "state": "scheduled",
+            "paused_at": None,
+            "paused_reason": None,
+            "next_run_at": next_run_at,
+        },
+    )
+
+
+def trigger_job(job_id: str) -> Optional[Dict[str, Any]]:
+    """Schedule a job to run on the next scheduler tick."""
+    job = get_job(job_id)
+    if not job:
+        return None
+    return update_job(
+        job_id,
+        {
+            "enabled": True,
+            "state": "scheduled",
+            "paused_at": None,
+            "paused_reason": None,
+            "next_run_at": _hermes_now().isoformat(),
+        },
+    )
+
+
+def remove_job(job_id: str) -> bool:
+    """Remove a job by ID."""
+    jobs = load_jobs()
+    original_len = len(jobs)
+    jobs = [j for j in jobs if j["id"] != job_id]
+    if len(jobs) < original_len:
+        save_jobs(jobs)
+        return True
+    return False
+
+
+def mark_job_run(job_id: str, success: bool, error: Optional[str] = None):
+    """
+    Mark a job as having been run.
+    
+    Updates last_run_at, last_status, increments completed count,
+    computes next_run_at, and auto-deletes if repeat limit reached.
+    """
+    jobs = load_jobs()
+    for i, job in enumerate(jobs):
+        if job["id"] == job_id:
+            now = _hermes_now().isoformat()
+            job["last_run_at"] = now
+            job["last_status"] = "ok" if success else "error"
+            job["last_error"] = error if not success else None
+            
+            # Increment completed count
+            if job.get("repeat"):
+                job["repeat"]["completed"] = job["repeat"].get("completed", 0) + 1
+                
+                # Check if we've hit the repeat limit
+                times = job["repeat"].get("times")
+                completed = job["repeat"]["completed"]
+                if times is not None and times > 0 and completed >= times:
+                    # Remove the job (limit reached)
+                    jobs.pop(i)
+                    save_jobs(jobs)
+                    return
+            
+            # Compute next run
+            job["next_run_at"] = compute_next_run(job["schedule"], now)
+
+            # If no next run (one-shot completed), disable
+            if job["next_run_at"] is None:
+                job["enabled"] = False
+                job["state"] = "completed"
+            elif job.get("state") != "paused":
+                job["state"] = "scheduled"
+
+            save_jobs(jobs)
+            return
+    
+    save_jobs(jobs)
+
+
+def advance_next_run(job_id: str) -> bool:
+    """Preemptively advance next_run_at for a recurring job before execution.
+
+    Call this BEFORE run_job() so that if the process crashes mid-execution,
+    the job won't re-fire on the next gateway restart.  This converts the
+    scheduler from at-least-once to at-most-once for recurring jobs — missing
+    one run is far better than firing dozens of times in a crash loop.
+
+    One-shot jobs are left unchanged so they can still retry on restart.
+
+    Returns True if next_run_at was advanced, False otherwise.
+    """
+    jobs = load_jobs()
+    for job in jobs:
+        if job["id"] == job_id:
+            kind = job.get("schedule", {}).get("kind")
+            if kind not in ("cron", "interval"):
+                return False
+            now = _hermes_now().isoformat()
+            new_next = compute_next_run(job["schedule"], now)
+            if new_next and new_next != job.get("next_run_at"):
+                job["next_run_at"] = new_next
+                save_jobs(jobs)
+                return True
+            return False
+    return False
+
+
+def get_due_jobs() -> List[Dict[str, Any]]:
+    """Get all jobs that are due to run now.
+
+    For recurring jobs (cron/interval), if the scheduled time is stale
+    (more than one period in the past, e.g. because the gateway was down),
+    the job is fast-forwarded to the next future run instead of firing
+    immediately.  This prevents a burst of missed jobs on gateway restart.
+    """
+    now = _hermes_now()
+    raw_jobs = load_jobs()
+    jobs = [_apply_skill_fields(j) for j in copy.deepcopy(raw_jobs)]
+    due = []
+    needs_save = False
+
+    for job in jobs:
+        if not job.get("enabled", True):
+            continue
+
+        next_run = job.get("next_run_at")
+        if not next_run:
+            recovered_next = _recoverable_oneshot_run_at(
+                job.get("schedule", {}),
+                now,
+                last_run_at=job.get("last_run_at"),
+            )
+            if not recovered_next:
+                continue
+
+            job["next_run_at"] = recovered_next
+            next_run = recovered_next
+            logger.info(
+                "Job '%s' had no next_run_at; recovering one-shot run at %s",
+                job.get("name", job["id"]),
+                recovered_next,
+            )
+            for rj in raw_jobs:
+                if rj["id"] == job["id"]:
+                    rj["next_run_at"] = recovered_next
+                    needs_save = True
+                    break
+
+        next_run_dt = _ensure_aware(datetime.fromisoformat(next_run))
+        if next_run_dt <= now:
+            schedule = job.get("schedule", {})
+            kind = schedule.get("kind")
+
+            # For recurring jobs, check if the scheduled time is stale
+            # (gateway was down and missed the window). Fast-forward to
+            # the next future occurrence instead of firing a stale run.
+            grace = _compute_grace_seconds(schedule)
+            if kind in ("cron", "interval") and (now - next_run_dt).total_seconds() > grace:
+                # Job is past its catch-up grace window — this is a stale missed run.
+                # Grace scales with schedule period: daily=2h, hourly=30m, 10min=5m.
+                new_next = compute_next_run(schedule, now.isoformat())
+                if new_next:
+                    logger.info(
+                        "Job '%s' missed its scheduled time (%s, grace=%ds). "
+                        "Fast-forwarding to next run: %s",
+                        job.get("name", job["id"]),
+                        next_run,
+                        grace,
+                        new_next,
+                    )
+                    # Update the job in storage
+                    for rj in raw_jobs:
+                        if rj["id"] == job["id"]:
+                            rj["next_run_at"] = new_next
+                            needs_save = True
+                            break
+                    continue  # Skip this run
+
+            due.append(job)
+
+    if needs_save:
+        save_jobs(raw_jobs)
+
+    return due
+
+
+def save_job_output(job_id: str, output: str):
+    """Save job output to file."""
+    ensure_dirs()
+    job_output_dir = OUTPUT_DIR / job_id
+    job_output_dir.mkdir(parents=True, exist_ok=True)
+    _secure_dir(job_output_dir)
+    
+    timestamp = _hermes_now().strftime("%Y-%m-%d_%H-%M-%S")
+    output_file = job_output_dir / f"{timestamp}.md"
+    
+    fd, tmp_path = tempfile.mkstemp(dir=str(job_output_dir), suffix='.tmp', prefix='.output_')
+    try:
+        with os.fdopen(fd, 'w', encoding='utf-8') as f:
+            f.write(output)
+            f.flush()
+            os.fsync(f.fileno())
+        os.replace(tmp_path, output_file)
+        _secure_file(output_file)
+    except BaseException:
+        try:
+            os.unlink(tmp_path)
+        except OSError:
+            pass
+        raise
+    
+    return output_file

cron/scheduler.py

@@ -0,0 +1,605 @@
+"""
+Cron job scheduler - executes due jobs.
+
+Provides tick() which checks for due jobs and runs them. The gateway
+calls this every 60 seconds from a background thread.
+
+Uses a file-based lock (~/.hermes/cron/.tick.lock) so only one tick
+runs at a time if multiple processes overlap.
+"""
+
+import asyncio
+import json
+import logging
+import os
+import sys
+import traceback
+
+# fcntl is Unix-only; on Windows use msvcrt for file locking
+try:
+    import fcntl
+except ImportError:
+    fcntl = None
+    try:
+        import msvcrt
+    except ImportError:
+        msvcrt = None
+from pathlib import Path
+from hermes_constants import get_hermes_home
+from hermes_cli.config import load_config
+from typing import Optional
+
+from hermes_time import now as _hermes_now
+
+logger = logging.getLogger(__name__)
+
+# Add parent directory to path for imports
+sys.path.insert(0, str(Path(__file__).parent.parent))
+
+from cron.jobs import get_due_jobs, mark_job_run, save_job_output, advance_next_run
+
+# Sentinel: when a cron agent has nothing new to report, it can start its
+# response with this marker to suppress delivery.  Output is still saved
+# locally for audit.
+SILENT_MARKER = "[SILENT]"
+
+# Resolve Hermes home directory (respects HERMES_HOME override)
+_hermes_home = get_hermes_home()
+
+# File-based lock prevents concurrent ticks from gateway + daemon + systemd timer
+_LOCK_DIR = _hermes_home / "cron"
+_LOCK_FILE = _LOCK_DIR / ".tick.lock"
+
+
+def _resolve_origin(job: dict) -> Optional[dict]:
+    """Extract origin info from a job, preserving any extra routing metadata."""
+    origin = job.get("origin")
+    if not origin:
+        return None
+    platform = origin.get("platform")
+    chat_id = origin.get("chat_id")
+    if platform and chat_id:
+        return origin
+    return None
+
+
+def _resolve_delivery_target(job: dict) -> Optional[dict]:
+    """Resolve the concrete auto-delivery target for a cron job, if any."""
+    deliver = job.get("deliver", "local")
+    origin = _resolve_origin(job)
+
+    if deliver == "local":
+        return None
+
+    if deliver == "origin":
+        if not origin:
+            return None
+        return {
+            "platform": origin["platform"],
+            "chat_id": str(origin["chat_id"]),
+            "thread_id": origin.get("thread_id"),
+        }
+
+    if ":" in deliver:
+        platform_name, rest = deliver.split(":", 1)
+        # Check for thread_id suffix (e.g. "telegram:-1003724596514:17")
+        if ":" in rest:
+            chat_id, thread_id = rest.split(":", 1)
+        else:
+            chat_id, thread_id = rest, None
+
+        # Resolve human-friendly labels like "Alice (dm)" to real IDs.
+        # send_message(action="list") shows labels with display suffixes
+        # that aren't valid platform IDs (e.g. WhatsApp JIDs).
+        try:
+            from gateway.channel_directory import resolve_channel_name
+            target = chat_id
+            # Strip display suffix like " (dm)" or " (group)"
+            if target.endswith(")") and " (" in target:
+                target = target.rsplit(" (", 1)[0].strip()
+            resolved = resolve_channel_name(platform_name.lower(), target)
+            if resolved:
+                chat_id = resolved
+        except Exception:
+            pass
+
+        return {
+            "platform": platform_name,
+            "chat_id": chat_id,
+            "thread_id": thread_id,
+        }
+
+    platform_name = deliver
+    if origin and origin.get("platform") == platform_name:
+        return {
+            "platform": platform_name,
+            "chat_id": str(origin["chat_id"]),
+            "thread_id": origin.get("thread_id"),
+        }
+
+    chat_id = os.getenv(f"{platform_name.upper()}_HOME_CHANNEL", "")
+    if not chat_id:
+        return None
+
+    return {
+        "platform": platform_name,
+        "chat_id": chat_id,
+        "thread_id": None,
+    }
+
+
+def _deliver_result(job: dict, content: str) -> None:
+    """
+    Deliver job output to the configured target (origin chat, specific platform, etc.).
+
+    Uses the standalone platform send functions from send_message_tool so delivery
+    works whether or not the gateway is running.
+    """
+    target = _resolve_delivery_target(job)
+    if not target:
+        if job.get("deliver", "local") != "local":
+            logger.warning(
+                "Job '%s' deliver=%s but no concrete delivery target could be resolved",
+                job["id"],
+                job.get("deliver", "local"),
+            )
+        return
+
+    platform_name = target["platform"]
+    chat_id = target["chat_id"]
+    thread_id = target.get("thread_id")
+
+    from tools.send_message_tool import _send_to_platform
+    from gateway.config import load_gateway_config, Platform
+
+    platform_map = {
+        "telegram": Platform.TELEGRAM,
+        "discord": Platform.DISCORD,
+        "slack": Platform.SLACK,
+        "whatsapp": Platform.WHATSAPP,
+        "signal": Platform.SIGNAL,
+        "matrix": Platform.MATRIX,
+        "mattermost": Platform.MATTERMOST,
+        "homeassistant": Platform.HOMEASSISTANT,
+        "dingtalk": Platform.DINGTALK,
+        "feishu": Platform.FEISHU,
+        "wecom": Platform.WECOM,
+        "email": Platform.EMAIL,
+        "sms": Platform.SMS,
+    }
+    platform = platform_map.get(platform_name.lower())
+    if not platform:
+        logger.warning("Job '%s': unknown platform '%s' for delivery", job["id"], platform_name)
+        return
+
+    try:
+        config = load_gateway_config()
+    except Exception as e:
+        logger.error("Job '%s': failed to load gateway config for delivery: %s", job["id"], e)
+        return
+
+    pconfig = config.platforms.get(platform)
+    if not pconfig or not pconfig.enabled:
+        logger.warning("Job '%s': platform '%s' not configured/enabled", job["id"], platform_name)
+        return
+
+    # Optionally wrap the content with a header/footer so the user knows this
+    # is a cron delivery.  Wrapping is on by default; set cron.wrap_response: false
+    # in config.yaml for clean output.
+    wrap_response = True
+    try:
+        user_cfg = load_config()
+        wrap_response = user_cfg.get("cron", {}).get("wrap_response", True)
+    except Exception:
+        pass
+
+    if wrap_response:
+        task_name = job.get("name", job["id"])
+        delivery_content = (
+            f"Cronjob Response: {task_name}\n"
+            f"-------------\n\n"
+            f"{content}\n\n"
+            f"Note: The agent cannot see this message, and therefore cannot respond to it."
+        )
+    else:
+        delivery_content = content
+
+    # Run the async send in a fresh event loop (safe from any thread)
+    coro = _send_to_platform(platform, pconfig, chat_id, delivery_content, thread_id=thread_id)
+    try:
+        result = asyncio.run(coro)
+    except RuntimeError:
+        # asyncio.run() checks for a running loop before awaiting the coroutine;
+        # when it raises, the original coro was never started — close it to
+        # prevent "coroutine was never awaited" RuntimeWarning, then retry in a
+        # fresh thread that has no running loop.
+        coro.close()
+        import concurrent.futures
+        with concurrent.futures.ThreadPoolExecutor(max_workers=1) as pool:
+            future = pool.submit(asyncio.run, _send_to_platform(platform, pconfig, chat_id, delivery_content, thread_id=thread_id))
+            result = future.result(timeout=30)
+    except Exception as e:
+        logger.error("Job '%s': delivery to %s:%s failed: %s", job["id"], platform_name, chat_id, e)
+        return
+
+    if result and result.get("error"):
+        logger.error("Job '%s': delivery error: %s", job["id"], result["error"])
+    else:
+        logger.info("Job '%s': delivered to %s:%s", job["id"], platform_name, chat_id)
+
+
+def _build_job_prompt(job: dict) -> str:
+    """Build the effective prompt for a cron job, optionally loading one or more skills first."""
+    prompt = job.get("prompt", "")
+    skills = job.get("skills")
+
+    # Always prepend [SILENT] guidance so the cron agent can suppress
+    # delivery when it has nothing new or noteworthy to report.
+    silent_hint = (
+        "[SYSTEM: If you have a meaningful status report or findings, "
+        "send them — that is the whole point of this job. Only respond "
+        "with exactly \"[SILENT]\" (nothing else) when there is genuinely "
+        "nothing new to report. [SILENT] suppresses delivery to the user. "
+        "Never combine [SILENT] with content — either report your "
+        "findings normally, or say [SILENT] and nothing more.]\n\n"
+    )
+    prompt = silent_hint + prompt
+    if skills is None:
+        legacy = job.get("skill")
+        skills = [legacy] if legacy else []
+
+    skill_names = [str(name).strip() for name in skills if str(name).strip()]
+    if not skill_names:
+        return prompt
+
+    from tools.skills_tool import skill_view
+
+    parts = []
+    skipped: list[str] = []
+    for skill_name in skill_names:
+        loaded = json.loads(skill_view(skill_name))
+        if not loaded.get("success"):
+            error = loaded.get("error") or f"Failed to load skill '{skill_name}'"
+            logger.warning("Cron job '%s': skill not found, skipping — %s", job.get("name", job.get("id")), error)
+            skipped.append(skill_name)
+            continue
+
+        content = str(loaded.get("content") or "").strip()
+        if parts:
+            parts.append("")
+        parts.extend(
+            [
+                f'[SYSTEM: The user has invoked the "{skill_name}" skill, indicating they want you to follow its instructions. The full skill content is loaded below.]',
+                "",
+                content,
+            ]
+        )
+
+    if skipped:
+        notice = (
+            f"[SYSTEM: The following skill(s) were listed for this job but could not be found "
+            f"and were skipped: {', '.join(skipped)}. "
+            f"Start your response with a brief notice so the user is aware, e.g.: "
+            f"'⚠️ Skill(s) not found and skipped: {', '.join(skipped)}']"
+        )
+        parts.insert(0, notice)
+
+    if prompt:
+        parts.extend(["", f"The user has provided the following instruction alongside the skill invocation: {prompt}"])
+    return "\n".join(parts)
+
+
+def run_job(job: dict) -> tuple[bool, str, str, Optional[str]]:
+    """
+    Execute a single cron job.
+    
+    Returns:
+        Tuple of (success, full_output_doc, final_response, error_message)
+    """
+    from run_agent import AIAgent
+    
+    # Initialize SQLite session store so cron job messages are persisted
+    # and discoverable via session_search (same pattern as gateway/run.py).
+    _session_db = None
+    try:
+        from hermes_state import SessionDB
+        _session_db = SessionDB()
+    except Exception as e:
+        logger.debug("Job '%s': SQLite session store not available: %s", job.get("id", "?"), e)
+    
+    job_id = job["id"]
+    job_name = job["name"]
+    prompt = _build_job_prompt(job)
+    origin = _resolve_origin(job)
+    _cron_session_id = f"cron_{job_id}_{_hermes_now().strftime('%Y%m%d_%H%M%S')}"
+
+    logger.info("Running job '%s' (ID: %s)", job_name, job_id)
+    logger.info("Prompt: %s", prompt[:100])
+
+    # Inject origin context so the agent's send_message tool knows the chat
+    if origin:
+        os.environ["HERMES_SESSION_PLATFORM"] = origin["platform"]
+        os.environ["HERMES_SESSION_CHAT_ID"] = str(origin["chat_id"])
+        if origin.get("chat_name"):
+            os.environ["HERMES_SESSION_CHAT_NAME"] = origin["chat_name"]
+
+    try:
+        # Re-read .env and config.yaml fresh every run so provider/key
+        # changes take effect without a gateway restart.
+        from dotenv import load_dotenv
+        try:
+            load_dotenv(str(_hermes_home / ".env"), override=True, encoding="utf-8")
+        except UnicodeDecodeError:
+            load_dotenv(str(_hermes_home / ".env"), override=True, encoding="latin-1")
+
+        delivery_target = _resolve_delivery_target(job)
+        if delivery_target:
+            os.environ["HERMES_CRON_AUTO_DELIVER_PLATFORM"] = delivery_target["platform"]
+            os.environ["HERMES_CRON_AUTO_DELIVER_CHAT_ID"] = str(delivery_target["chat_id"])
+            if delivery_target.get("thread_id") is not None:
+                os.environ["HERMES_CRON_AUTO_DELIVER_THREAD_ID"] = str(delivery_target["thread_id"])
+
+        model = job.get("model") or os.getenv("HERMES_MODEL") or ""
+
+        # Load config.yaml for model, reasoning, prefill, toolsets, provider routing
+        _cfg = {}
+        try:
+            import yaml
+            _cfg_path = str(_hermes_home / "config.yaml")
+            if os.path.exists(_cfg_path):
+                with open(_cfg_path) as _f:
+                    _cfg = yaml.safe_load(_f) or {}
+                _model_cfg = _cfg.get("model", {})
+                if not job.get("model"):
+                    if isinstance(_model_cfg, str):
+                        model = _model_cfg
+                    elif isinstance(_model_cfg, dict):
+                        model = _model_cfg.get("default", model)
+        except Exception as e:
+            logger.warning("Job '%s': failed to load config.yaml, using defaults: %s", job_id, e)
+
+        # Reasoning config from env or config.yaml
+        from hermes_constants import parse_reasoning_effort
+        effort = os.getenv("HERMES_REASONING_EFFORT", "")
+        if not effort:
+            effort = str(_cfg.get("agent", {}).get("reasoning_effort", "")).strip()
+        reasoning_config = parse_reasoning_effort(effort)
+
+        # Prefill messages from env or config.yaml
+        prefill_messages = None
+        prefill_file = os.getenv("HERMES_PREFILL_MESSAGES_FILE", "") or _cfg.get("prefill_messages_file", "")
+        if prefill_file:
+            import json as _json
+            pfpath = Path(prefill_file).expanduser()
+            if not pfpath.is_absolute():
+                pfpath = _hermes_home / pfpath
+            if pfpath.exists():
+                try:
+                    with open(pfpath, "r", encoding="utf-8") as _pf:
+                        prefill_messages = _json.load(_pf)
+                    if not isinstance(prefill_messages, list):
+                        prefill_messages = None
+                except Exception as e:
+                    logger.warning("Job '%s': failed to parse prefill messages file '%s': %s", job_id, pfpath, e)
+                    prefill_messages = None
+
+        # Max iterations
+        max_iterations = _cfg.get("agent", {}).get("max_turns") or _cfg.get("max_turns") or 90
+
+        # Provider routing
+        pr = _cfg.get("provider_routing", {})
+        smart_routing = _cfg.get("smart_model_routing", {}) or {}
+
+        from hermes_cli.runtime_provider import (
+            resolve_runtime_provider,
+            format_runtime_provider_error,
+        )
+        try:
+            runtime_kwargs = {
+                "requested": job.get("provider") or os.getenv("HERMES_INFERENCE_PROVIDER"),
+            }
+            if job.get("base_url"):
+                runtime_kwargs["explicit_base_url"] = job.get("base_url")
+            runtime = resolve_runtime_provider(**runtime_kwargs)
+        except Exception as exc:
+            message = format_runtime_provider_error(exc)
+            raise RuntimeError(message) from exc
+
+        from agent.smart_model_routing import resolve_turn_route
+        turn_route = resolve_turn_route(
+            prompt,
+            smart_routing,
+            {
+                "model": model,
+                "api_key": runtime.get("api_key"),
+                "base_url": runtime.get("base_url"),
+                "provider": runtime.get("provider"),
+                "api_mode": runtime.get("api_mode"),
+                "command": runtime.get("command"),
+                "args": list(runtime.get("args") or []),
+            },
+        )
+
+        agent = AIAgent(
+            model=turn_route["model"],
+            api_key=turn_route["runtime"].get("api_key"),
+            base_url=turn_route["runtime"].get("base_url"),
+            provider=turn_route["runtime"].get("provider"),
+            api_mode=turn_route["runtime"].get("api_mode"),
+            acp_command=turn_route["runtime"].get("command"),
+            acp_args=turn_route["runtime"].get("args"),
+            max_iterations=max_iterations,
+            reasoning_config=reasoning_config,
+            prefill_messages=prefill_messages,
+            providers_allowed=pr.get("only"),
+            providers_ignored=pr.get("ignore"),
+            providers_order=pr.get("order"),
+            provider_sort=pr.get("sort"),
+            disabled_toolsets=["cronjob", "messaging", "clarify"],
+            quiet_mode=True,
+            platform="cron",
+            session_id=_cron_session_id,
+            session_db=_session_db,
+        )
+        
+        result = agent.run_conversation(prompt)
+        
+        final_response = result.get("final_response", "") or ""
+        # Use a separate variable for log display; keep final_response clean
+        # for delivery logic (empty response = no delivery).
+        logged_response = final_response if final_response else "(No response generated)"
+        
+        output = f"""# Cron Job: {job_name}
+
+**Job ID:** {job_id}
+**Run Time:** {_hermes_now().strftime('%Y-%m-%d %H:%M:%S')}
+**Schedule:** {job.get('schedule_display', 'N/A')}
+
+## Prompt
+
+{prompt}
+
+## Response
+
+{logged_response}
+"""
+        
+        logger.info("Job '%s' completed successfully", job_name)
+        return True, output, final_response, None
+        
+    except Exception as e:
+        error_msg = f"{type(e).__name__}: {str(e)}"
+        logger.error("Job '%s' failed: %s", job_name, error_msg)
+        
+        output = f"""# Cron Job: {job_name} (FAILED)
+
+**Job ID:** {job_id}
+**Run Time:** {_hermes_now().strftime('%Y-%m-%d %H:%M:%S')}
+**Schedule:** {job.get('schedule_display', 'N/A')}
+
+## Prompt
+
+{prompt}
+
+## Error
+
+```
+{error_msg}
+
+{traceback.format_exc()}
+```
+"""
+        return False, output, "", error_msg
+
+    finally:
+        # Clean up injected env vars so they don't leak to other jobs
+        for key in (
+            "HERMES_SESSION_PLATFORM",
+            "HERMES_SESSION_CHAT_ID",
+            "HERMES_SESSION_CHAT_NAME",
+            "HERMES_CRON_AUTO_DELIVER_PLATFORM",
+            "HERMES_CRON_AUTO_DELIVER_CHAT_ID",
+            "HERMES_CRON_AUTO_DELIVER_THREAD_ID",
+        ):
+            os.environ.pop(key, None)
+        if _session_db:
+            try:
+                _session_db.end_session(_cron_session_id, "cron_complete")
+            except (Exception, KeyboardInterrupt) as e:
+                logger.debug("Job '%s': failed to end session: %s", job_id, e)
+            try:
+                _session_db.close()
+            except (Exception, KeyboardInterrupt) as e:
+                logger.debug("Job '%s': failed to close SQLite session store: %s", job_id, e)
+
+
+def tick(verbose: bool = True) -> int:
+    """
+    Check and run all due jobs.
+    
+    Uses a file lock so only one tick runs at a time, even if the gateway's
+    in-process ticker and a standalone daemon or manual tick overlap.
+    
+    Args:
+        verbose: Whether to print status messages
+    
+    Returns:
+        Number of jobs executed (0 if another tick is already running)
+    """
+    _LOCK_DIR.mkdir(parents=True, exist_ok=True)
+
+    # Cross-platform file locking: fcntl on Unix, msvcrt on Windows
+    lock_fd = None
+    try:
+        lock_fd = open(_LOCK_FILE, "w")
+        if fcntl:
+            fcntl.flock(lock_fd, fcntl.LOCK_EX | fcntl.LOCK_NB)
+        elif msvcrt:
+            msvcrt.locking(lock_fd.fileno(), msvcrt.LK_NBLCK, 1)
+    except (OSError, IOError):
+        logger.debug("Tick skipped — another instance holds the lock")
+        if lock_fd is not None:
+            lock_fd.close()
+        return 0
+
+    try:
+        due_jobs = get_due_jobs()
+
+        if verbose and not due_jobs:
+            logger.info("%s - No jobs due", _hermes_now().strftime('%H:%M:%S'))
+            return 0
+
+        if verbose:
+            logger.info("%s - %s job(s) due", _hermes_now().strftime('%H:%M:%S'), len(due_jobs))
+
+        executed = 0
+        for job in due_jobs:
+            try:
+                # For recurring jobs (cron/interval), advance next_run_at to the
+                # next future occurrence BEFORE execution.  This way, if the
+                # process crashes mid-run, the job won't re-fire on restart.
+                # One-shot jobs are left alone so they can retry on restart.
+                advance_next_run(job["id"])
+
+                success, output, final_response, error = run_job(job)
+
+                output_file = save_job_output(job["id"], output)
+                if verbose:
+                    logger.info("Output saved to: %s", output_file)
+
+                # Deliver the final response to the origin/target chat.
+                # If the agent responded with [SILENT], skip delivery (but
+                # output is already saved above).  Failed jobs always deliver.
+                deliver_content = final_response if success else f"⚠️ Cron job '{job.get('name', job['id'])}' failed:\n{error}"
+                should_deliver = bool(deliver_content)
+                if should_deliver and success and deliver_content.strip().upper().startswith(SILENT_MARKER):
+                    logger.info("Job '%s': agent returned %s — skipping delivery", job["id"], SILENT_MARKER)
+                    should_deliver = False
+
+                if should_deliver:
+                    try:
+                        _deliver_result(job, deliver_content)
+                    except Exception as de:
+                        logger.error("Delivery failed for job %s: %s", job["id"], de)
+
+                mark_job_run(job["id"], success, error)
+                executed += 1
+
+            except Exception as e:
+                logger.error("Error processing job %s: %s", job['id'], e)
+                mark_job_run(job["id"], False, str(e))
+
+        return executed
+    finally:
+        if fcntl:
+            fcntl.flock(lock_fd, fcntl.LOCK_UN)
+        elif msvcrt:
+            try:
+                msvcrt.locking(lock_fd.fileno(), msvcrt.LK_UNLCK, 1)
+            except (OSError, IOError):
+                pass
+        lock_fd.close()
+
+
+if __name__ == "__main__":
+    tick(verbose=True)

datagen-config-examples/example_browser_tasks.jsonl

@@ -0,0 +1,5 @@
+{"prompt": "Go to https://news.ycombinator.com and find the top 5 posts on the front page. For each post, get the title, URL, points, and number of comments. Return the results as a formatted summary."}
+{"prompt": "Navigate to https://en.wikipedia.org/wiki/Hermes and extract the first paragraph of the article, the image caption, and the list of items in the infobox. Summarize what you find."}
+{"prompt": "Go to https://github.com/trending and find the top 3 trending repositories today. For each repo, get the name, description, language, and star count. Write the results to a file called trending_repos.md."}
+{"prompt": "Visit https://httpbin.org/forms/post and fill out the form with sample data (customer name: Jane Doe, size: Medium, topping: Bacon, delivery time: 12:00). Submit the form and report what the response page shows."}
+{"prompt": "Navigate to https://books.toscrape.com, browse to the Travel category, find the highest-rated book, and extract its title, price, availability, and description."}

datagen-config-examples/run_browser_tasks.sh

@@ -0,0 +1,65 @@
+#!/bin/bash
+
+# =============================================================================
+# Example: Browser-Focused Data Generation
+# =============================================================================
+#
+# Generates tool-calling trajectories for browser automation tasks.
+# The agent navigates websites, fills forms, extracts information, etc.
+#
+# Distribution: browser 97%, web 20%, vision 12%, terminal 15%
+#
+# Prerequisites:
+#   - OPENROUTER_API_KEY in ~/.hermes/.env
+#   - BROWSERBASE_API_KEY in ~/.hermes/.env (for browser tools)
+#   - A dataset JSONL file with one {"prompt": "..."} per line
+#
+# Usage:
+#   cd ~/.hermes/hermes-agent
+#   bash datagen-config-examples/run_browser_tasks.sh
+#
+# Output: data/browser_tasks_example/trajectories.jsonl
+# =============================================================================
+
+mkdir -p logs
+
+LOG_FILE="logs/browser_tasks_$(date +%Y%m%d_%H%M%S).log"
+echo "📝 Logging to: $LOG_FILE"
+
+# Point to the example dataset in this directory
+SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
+
+python batch_runner.py \
+  --dataset_file="$SCRIPT_DIR/example_browser_tasks.jsonl" \
+  --batch_size=5 \
+  --run_name="browser_tasks_example" \
+  --distribution="browser_tasks" \
+  --model="anthropic/claude-sonnet-4" \
+  --base_url="https://openrouter.ai/api/v1" \
+  --num_workers=3 \
+  --max_turns=30 \
+  --ephemeral_system_prompt="You are an AI assistant with browser automation capabilities. Your primary task is to navigate and interact with web pages to accomplish user goals.
+
+IMPORTANT GUIDELINES:
+
+1. SEARCHING: Do NOT search directly on Google via the browser — they block automated searches. Use the web_search tool first to find URLs, then navigate to them with browser tools.
+
+2. COOKIE/PRIVACY DIALOGS: After navigating to a page, check for cookie consent or privacy popups. Dismiss them by clicking Accept/Close/OK before interacting with other elements. Take a fresh browser_snapshot afterward.
+
+3. HANDLING TIMEOUTS: If an action times out, the element may be blocked by an overlay. Take a new snapshot and look for dialogs to dismiss. If none, try an alternative approach or report the issue.
+
+4. GENERAL: Use browser tools to click, fill forms, and extract information. Use terminal for local file operations. Verify your actions and handle errors gracefully." \
+  2>&1 | tee "$LOG_FILE"
+
+echo "✅ Done. Log: $LOG_FILE"
+
+# =============================================================================
+# Common options you can add:
+#
+#   --resume                  Resume from checkpoint if interrupted
+#   --verbose                 Enable detailed logging
+#   --max_tokens=63000        Set max response tokens
+#   --reasoning_disabled      Disable model thinking/reasoning tokens
+#   --providers_allowed="anthropic,google"  Restrict to specific providers
+#   --prefill_messages_file="configs/prefill.json"  Few-shot priming
+# =============================================================================

datagen-config-examples/trajectory_compression.yaml

@@ -0,0 +1,101 @@
+# Trajectory Compression Configuration
+# 
+# Post-processes completed agent trajectories to fit within a target token budget.
+# Compression preserves head/tail turns and summarizes middle content only as needed.
+
+# Tokenizer settings for accurate token counting
+tokenizer:
+  # HuggingFace tokenizer name
+  name: "moonshotai/Kimi-K2-Thinking"
+  
+  # Trust remote code (required for some tokenizers)
+  trust_remote_code: true
+
+# Compression targets and behavior
+compression:
+  # Target maximum tokens for compressed trajectory
+  target_max_tokens: 29000
+  
+  # Target size for summary (in tokens)
+  # This is factored into calculations when determining what to compress
+  summary_target_tokens: 750
+
+# Protected turns that should NEVER be compressed
+protected_turns:
+  # Always protect the first system message (tool definitions)
+  first_system: true
+  
+  # Always protect the first human message (original request)
+  first_human: true
+  
+  # Always protect the first gpt message (initial response/tool_call)
+  first_gpt: true
+  
+  # Always protect the first tool response (result of first action)
+  first_tool: true
+  
+  # Always protect the last 2 complete turn pairs (gpt+tool or gpt only)
+  # This ensures the model's final actions and conclusions are preserved
+  last_n_turns: 4
+
+# LLM settings for generating summaries (OpenRouter only)
+summarization:
+  # Model to use for summarization (should be fast and cheap)
+  # Using OpenRouter model path format
+  model: "google/gemini-3-flash-preview"
+  
+  # OpenRouter API settings
+  base_url: "https://openrouter.ai/api/v1"
+  
+  # Environment variable containing OpenRouter API key
+  api_key_env: "OPENROUTER_API_KEY"
+  
+  # Temperature for summarization (lower = more deterministic)
+  temperature: 0.3
+  
+  # Max retries for API failures
+  max_retries: 3
+  
+  # Delay between retries (seconds)
+  retry_delay: 2
+
+# Output settings
+output:
+  # Add notice to system message about potential summarization
+  add_summary_notice: true
+  
+  # Text to append to system message
+  summary_notice_text: "\n\nSome of the conversation may be summarized to preserve context."
+  
+  # Output directory suffix (appended to input directory name)
+  output_suffix: "_compressed"
+
+# Processing settings
+processing:
+  # Number of parallel workers for batch processing
+  num_workers: 4
+  
+  # Maximum concurrent API calls for summarization (async parallelism)
+  max_concurrent_requests: 50
+  
+  # Skip trajectories that are already under target length
+  skip_under_target: true
+  
+  # If true, save trajectories even if compression can't get under target
+  # (will compress as much as possible)
+  save_over_limit: true
+  
+  # Timeout per trajectory in seconds (skip if takes longer)
+  # Helps avoid hanging on problematic entries
+  per_trajectory_timeout: 300  # 5 minutes
+
+# Metrics to track
+metrics:
+  # Log detailed compression statistics
+  enabled: true
+  
+  # Save per-trajectory metrics in output
+  per_trajectory: false
+  
+  # Metrics file name (saved in output directory)
+  output_file: "compression_metrics.json"

datagen-config-examples/web_research.yaml

@@ -0,0 +1,46 @@
+# datagen-config-examples/web_research.yaml
+#
+# Batch data generation config for WebResearchEnv.
+# Generates tool-calling trajectories for multi-step web research tasks.
+#
+# Usage:
+#   python batch_runner.py \
+#     --config datagen-config-examples/web_research.yaml \
+#     --run_name web_research_v1
+
+environment: web-research
+
+# Toolsets available to the agent during data generation
+toolsets:
+  - web
+  - file
+
+# How many parallel workers to use
+num_workers: 4
+
+# Questions per batch
+batch_size: 20
+
+# Total trajectories to generate (comment out to run full dataset)
+max_items: 500
+
+# Model to use for generation (override with --model flag)
+model: openrouter/nousresearch/hermes-3-llama-3.1-405b
+
+# System prompt additions (ephemeral — not saved to trajectories)
+ephemeral_system_prompt: |
+  You are a highly capable research agent. When asked a factual question,
+  always use web_search to find current, accurate information before answering.
+  Cite at least 2 sources. Be concise and accurate.
+
+# Output directory
+output_dir: data/web_research_v1
+
+# Trajectory compression settings (for fitting into training token budgets)
+compression:
+  enabled: true
+  target_max_tokens: 16000
+
+# Eval settings
+eval_every: 100       # Run eval every N trajectories
+eval_size: 25         # Number of held-out questions per eval run

docker/SOUL.md

@@ -0,0 +1,15 @@
+# Hermes Agent Persona
+
+<!--
+This file defines the agent's personality and tone.
+The agent will embody whatever you write here.
+Edit this to customize how Hermes communicates with you.
+
+Examples:
+  - "You are a warm, playful assistant who uses kaomoji occasionally."
+  - "You are a concise technical expert. No fluff, just facts."
+  - "You speak like a friendly coworker who happens to know everything."
+
+This file is loaded fresh each message -- no restart needed.
+Delete the contents (or this file) to use the default personality.
+-->

docker/entrypoint.sh

@@ -0,0 +1,34 @@
+#!/bin/bash
+# Docker entrypoint: bootstrap config files into the mounted volume, then run hermes.
+set -e
+
+HERMES_HOME="/opt/data"
+INSTALL_DIR="/opt/hermes"
+
+# Create essential directory structure.  Cache and platform directories
+# (cache/images, cache/audio, platforms/whatsapp, etc.) are created on
+# demand by the application — don't pre-create them here so new installs
+# get the consolidated layout from get_hermes_dir().
+mkdir -p "$HERMES_HOME"/{cron,sessions,logs,hooks,memories,skills}
+
+# .env
+if [ ! -f "$HERMES_HOME/.env" ]; then
+    cp "$INSTALL_DIR/.env.example" "$HERMES_HOME/.env"
+fi
+
+# config.yaml
+if [ ! -f "$HERMES_HOME/config.yaml" ]; then
+    cp "$INSTALL_DIR/cli-config.yaml.example" "$HERMES_HOME/config.yaml"
+fi
+
+# SOUL.md
+if [ ! -f "$HERMES_HOME/SOUL.md" ]; then
+    cp "$INSTALL_DIR/docker/SOUL.md" "$HERMES_HOME/SOUL.md"
+fi
+
+# Sync bundled skills (manifest-based so user edits are preserved)
+if [ -d "$INSTALL_DIR/skills" ]; then
+    python3 "$INSTALL_DIR/tools/skills_sync.py"
+fi
+
+exec hermes "$@"

docs/acp-setup.md

@@ -0,0 +1,229 @@
+# Hermes Agent — ACP (Agent Client Protocol) Setup Guide
+
+Hermes Agent supports the **Agent Client Protocol (ACP)**, allowing it to run as
+a coding agent inside your editor. ACP lets your IDE send tasks to Hermes, and
+Hermes responds with file edits, terminal commands, and explanations — all shown
+natively in the editor UI.
+
+---
+
+## Prerequisites
+
+- Hermes Agent installed and configured (`hermes setup` completed)
+- An API key / provider set up in `~/.hermes/.env` or via `hermes login`
+- Python 3.11+
+
+Install the ACP extra:
+
+```bash
+pip install -e ".[acp]"
+```
+
+---
+
+## VS Code Setup
+
+### 1. Install the ACP Client extension
+
+Open VS Code and install **ACP Client** from the marketplace:
+
+- Press `Ctrl+Shift+X` (or `Cmd+Shift+X` on macOS)
+- Search for **"ACP Client"**
+- Click **Install**
+
+Or install from the command line:
+
+```bash
+code --install-extension anysphere.acp-client
+```
+
+### 2. Configure settings.json
+
+Open your VS Code settings (`Ctrl+,` → click the `{}` icon for JSON) and add:
+
+```json
+{
+  "acpClient.agents": [
+    {
+      "name": "hermes-agent",
+      "registryDir": "/path/to/hermes-agent/acp_registry"
+    }
+  ]
+}
+```
+
+Replace `/path/to/hermes-agent` with the actual path to your Hermes Agent
+installation (e.g. `~/.hermes/hermes-agent`).
+
+Alternatively, if `hermes` is on your PATH, the ACP Client can discover it
+automatically via the registry directory.
+
+### 3. Restart VS Code
+
+After configuring, restart VS Code. You should see **Hermes Agent** appear in
+the ACP agent picker in the chat/agent panel.
+
+---
+
+## Zed Setup
+
+Zed has built-in ACP support.
+
+### 1. Configure Zed settings
+
+Open Zed settings (`Cmd+,` on macOS or `Ctrl+,` on Linux) and add to your
+`settings.json`:
+
+```json
+{
+  "acp": {
+    "agents": [
+      {
+        "name": "hermes-agent",
+        "registry_dir": "/path/to/hermes-agent/acp_registry"
+      }
+    ]
+  }
+}
+```
+
+### 2. Restart Zed
+
+Hermes Agent will appear in the agent panel. Select it and start a conversation.
+
+---
+
+## JetBrains Setup (IntelliJ, PyCharm, WebStorm, etc.)
+
+### 1. Install the ACP plugin
+
+- Open **Settings** → **Plugins** → **Marketplace**
+- Search for **"ACP"** or **"Agent Client Protocol"**
+- Install and restart the IDE
+
+### 2. Configure the agent
+
+- Open **Settings** → **Tools** → **ACP Agents**
+- Click **+** to add a new agent
+- Set the registry directory to your `acp_registry/` folder:
+  `/path/to/hermes-agent/acp_registry`
+- Click **OK**
+
+### 3. Use the agent
+
+Open the ACP panel (usually in the right sidebar) and select **Hermes Agent**.
+
+---
+
+## What You Will See
+
+Once connected, your editor provides a native interface to Hermes Agent:
+
+### Chat Panel
+A conversational interface where you can describe tasks, ask questions, and
+give instructions. Hermes responds with explanations and actions.
+
+### File Diffs
+When Hermes edits files, you see standard diffs in the editor. You can:
+- **Accept** individual changes
+- **Reject** changes you don't want
+- **Review** the full diff before applying
+
+### Terminal Commands
+When Hermes needs to run shell commands (builds, tests, installs), the editor
+shows them in an integrated terminal. Depending on your settings:
+- Commands may run automatically
+- Or you may be prompted to **approve** each command
+
+### Approval Flow
+For potentially destructive operations, the editor will prompt you for
+approval before Hermes proceeds. This includes:
+- File deletions
+- Shell commands
+- Git operations
+
+---
+
+## Configuration
+
+Hermes Agent under ACP uses the **same configuration** as the CLI:
+
+- **API keys / providers**: `~/.hermes/.env`
+- **Agent config**: `~/.hermes/config.yaml`
+- **Skills**: `~/.hermes/skills/`
+- **Sessions**: `~/.hermes/state.db`
+
+You can run `hermes setup` to configure providers, or edit `~/.hermes/.env`
+directly.
+
+### Changing the model
+
+Edit `~/.hermes/config.yaml`:
+
+```yaml
+model: openrouter/nous/hermes-3-llama-3.1-70b
+```
+
+Or set the `HERMES_MODEL` environment variable.
+
+### Toolsets
+
+ACP sessions use the curated `hermes-acp` toolset by default. It is designed for editor workflows and intentionally excludes things like messaging delivery, cronjob management, and audio-first UX features.
+
+---
+
+## Troubleshooting
+
+### Agent doesn't appear in the editor
+
+1. **Check the registry path** — make sure the `acp_registry/` directory path
+   in your editor settings is correct and contains `agent.json`.
+2. **Check `hermes` is on PATH** — run `which hermes` in a terminal. If not
+   found, you may need to activate your virtualenv or add it to PATH.
+3. **Restart the editor** after changing settings.
+
+### Agent starts but errors immediately
+
+1. Run `hermes doctor` to check your configuration.
+2. Check that you have a valid API key: `hermes status`
+3. Try running `hermes acp` directly in a terminal to see error output.
+
+### "Module not found" errors
+
+Make sure you installed the ACP extra:
+
+```bash
+pip install -e ".[acp]"
+```
+
+### Slow responses
+
+- ACP streams responses, so you should see incremental output. If the agent
+  appears stuck, check your network connection and API provider status.
+- Some providers have rate limits. Try switching to a different model/provider.
+
+### Permission denied for terminal commands
+
+If the editor blocks terminal commands, check your ACP Client extension
+settings for auto-approval or manual-approval preferences.
+
+### Logs
+
+Hermes logs are written to stderr when running in ACP mode. Check:
+- VS Code: **Output** panel → select **ACP Client** or **Hermes Agent**
+- Zed: **View** → **Toggle Terminal** and check the process output
+- JetBrains: **Event Log** or the ACP tool window
+
+You can also enable verbose logging:
+
+```bash
+HERMES_LOG_LEVEL=DEBUG hermes acp
+```
+
+---
+
+## Further Reading
+
+- [ACP Specification](https://github.com/anysphere/acp)
+- [Hermes Agent Documentation](https://github.com/NousResearch/hermes-agent)
+- Run `hermes --help` for all CLI options

docs/honcho-integration-spec.html

@@ -0,0 +1,698 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+<meta charset="UTF-8">
+<meta name="viewport" content="width=device-width, initial-scale=1.0">
+<title>honcho-integration-spec</title>
+<style>
+  :root {
+    --bg:             #0b0e14;
+    --bg-surface:     #11151c;
+    --bg-elevated:    #181d27;
+    --bg-code:        #0d1018;
+    --fg:             #c9d1d9;
+    --fg-bright:      #e6edf3;
+    --fg-muted:       #6e7681;
+    --fg-subtle:      #484f58;
+    --accent:         #7eb8f6;
+    --accent-dim:     #3d6ea5;
+    --accent-glow:    rgba(126, 184, 246, 0.08);
+    --green:          #7ee6a8;
+    --green-dim:      #2ea04f;
+    --orange:         #e6a855;
+    --red:            #f47067;
+    --purple:         #bc8cff;
+    --cyan:           #56d4dd;
+    --border:         #21262d;
+    --border-subtle:  #161b22;
+    --radius:         6px;
+    --font-sans:      'New York', ui-serif, 'Iowan Old Style', 'Apple Garamond', Baskerville, 'Times New Roman', 'Noto Emoji', serif;
+    --font-mono:      'Departure Mono', 'Noto Emoji', monospace;
+  }
+
+  *, *::before, *::after { box-sizing: border-box; margin: 0; padding: 0; }
+  html { scroll-behavior: smooth; scroll-padding-top: 2rem; }
+  body {
+    font-family: var(--font-sans);
+    background: var(--bg);
+    color: var(--fg);
+    line-height: 1.7;
+    font-size: 15px;
+    -webkit-font-smoothing: antialiased;
+  }
+
+  .container { max-width: 860px; margin: 0 auto; padding: 3rem 2rem 6rem; }
+
+  .hero {
+    text-align: center;
+    padding: 4rem 0 3rem;
+    border-bottom: 1px solid var(--border);
+    margin-bottom: 3rem;
+  }
+  .hero h1 { font-family: var(--font-mono); font-size: 2.2rem; font-weight: 700; color: var(--fg-bright); letter-spacing: -0.03em; margin-bottom: 0.5rem; }
+  .hero h1 span { color: var(--accent); }
+  .hero .subtitle { font-family: var(--font-sans); color: var(--fg-muted); font-size: 0.92rem; max-width: 560px; margin: 0 auto; line-height: 1.6; }
+  .hero .meta { margin-top: 1.5rem; display: flex; justify-content: center; gap: 1.5rem; flex-wrap: wrap; }
+  .hero .meta span { font-size: 0.8rem; color: var(--fg-subtle); font-family: var(--font-mono); }
+
+  .toc { background: var(--bg-surface); border: 1px solid var(--border); border-radius: var(--radius); padding: 1.5rem 2rem; margin-bottom: 3rem; }
+  .toc h2 { font-size: 0.75rem; text-transform: uppercase; letter-spacing: 0.1em; color: var(--fg-muted); margin-bottom: 1rem; }
+  .toc ol { list-style: none; counter-reset: toc; columns: 2; column-gap: 2rem; }
+  .toc li { counter-increment: toc; break-inside: avoid; margin-bottom: 0.35rem; }
+  .toc li::before { content: counter(toc, decimal-leading-zero) " "; color: var(--fg-subtle); font-family: var(--font-mono); font-size: 0.75rem; margin-right: 0.25rem; }
+  .toc a { font-family: var(--font-mono); color: var(--fg); text-decoration: none; font-size: 0.82rem; transition: color 0.15s; }
+  .toc a:hover { color: var(--accent); }
+
+  section { margin-bottom: 4rem; }
+  section + section { padding-top: 1rem; }
+
+  h2 { font-family: var(--font-mono); font-size: 1.3rem; font-weight: 700; color: var(--fg-bright); letter-spacing: -0.01em; margin-bottom: 1.25rem; padding-bottom: 0.5rem; border-bottom: 1px solid var(--border); }
+  h3 { font-family: var(--font-mono); font-size: 1rem; font-weight: 600; color: var(--fg-bright); margin-top: 2rem; margin-bottom: 0.75rem; }
+  h4 { font-family: var(--font-mono); font-size: 0.9rem; font-weight: 600; color: var(--accent); margin-top: 1.5rem; margin-bottom: 0.5rem; }
+
+  p { margin-bottom: 1rem; font-size: 0.95rem; line-height: 1.75; }
+  strong { color: var(--fg-bright); font-weight: 600; }
+  a { color: var(--accent); text-decoration: none; }
+  a:hover { text-decoration: underline; }
+
+  ul, ol { margin-bottom: 1rem; padding-left: 1.5rem; font-size: 0.93rem; line-height: 1.7; }
+  li { margin-bottom: 0.35rem; }
+  li::marker { color: var(--fg-subtle); }
+
+  .table-wrap { overflow-x: auto; margin-bottom: 1.5rem; }
+  table { width: 100%; border-collapse: collapse; font-size: 0.88rem; }
+  th, td { text-align: left; padding: 0.6rem 1rem; border-bottom: 1px solid var(--border-subtle); }
+  th { font-family: var(--font-mono); font-size: 0.72rem; text-transform: uppercase; letter-spacing: 0.06em; color: var(--fg-muted); background: var(--bg-surface); border-bottom-color: var(--border); white-space: nowrap; }
+  td { font-family: var(--font-sans); font-size: 0.88rem; color: var(--fg); }
+  tr:hover td { background: var(--accent-glow); }
+  td code { background: var(--bg-elevated); padding: 0.15em 0.4em; border-radius: 3px; font-family: var(--font-mono); font-size: 0.82em; color: var(--cyan); }
+
+  pre { background: var(--bg-code); border: 1px solid var(--border); border-radius: var(--radius); padding: 1.25rem 1.5rem; overflow-x: auto; margin-bottom: 1.5rem; font-family: var(--font-mono); font-size: 0.82rem; line-height: 1.65; color: var(--fg); }
+  pre code { background: none; padding: 0; color: inherit; font-size: inherit; }
+  code { font-family: var(--font-mono); font-size: 0.85em; }
+  p code, li code { background: var(--bg-elevated); padding: 0.15em 0.4em; border-radius: 3px; color: var(--cyan); font-size: 0.85em; }
+
+  .kw { color: var(--purple); }
+  .str { color: var(--green); }
+  .cm { color: var(--fg-subtle); font-style: italic; }
+  .num { color: var(--orange); }
+  .key { color: var(--accent); }
+
+  .mermaid { margin: 1.5rem 0 2rem; text-align: center; }
+  .mermaid svg { max-width: 100%; height: auto; }
+
+  .callout { font-family: var(--font-sans); background: var(--bg-surface); border-left: 3px solid var(--accent-dim); border-radius: 0 var(--radius) var(--radius) 0; padding: 1rem 1.25rem; margin-bottom: 1.5rem; font-size: 0.88rem; color: var(--fg-muted); line-height: 1.6; }
+  .callout strong { font-family: var(--font-mono); color: var(--fg-bright); }
+  .callout.success { border-left-color: var(--green-dim); }
+  .callout.warn { border-left-color: var(--orange); }
+
+  .badge { display: inline-block; font-family: var(--font-mono); font-size: 0.65rem; font-weight: 600; text-transform: uppercase; letter-spacing: 0.05em; padding: 0.2em 0.6em; border-radius: 3px; vertical-align: middle; margin-left: 0.4rem; }
+  .badge-done { background: var(--green-dim); color: #fff; }
+  .badge-wip { background: var(--orange); color: #0b0e14; }
+  .badge-todo { background: var(--fg-subtle); color: var(--fg); }
+
+  .checklist { list-style: none; padding-left: 0; }
+  .checklist li { padding-left: 1.5rem; position: relative; margin-bottom: 0.5rem; }
+  .checklist li::before { position: absolute; left: 0; font-family: var(--font-mono); font-size: 0.85rem; }
+  .checklist li.done { color: var(--fg-muted); }
+  .checklist li.done::before { content: "\2713"; color: var(--green); }
+  .checklist li.todo::before { content: "\25CB"; color: var(--fg-subtle); }
+  .checklist li.wip::before { content: "\25D4"; color: var(--orange); }
+
+  .compare { display: grid; grid-template-columns: 1fr 1fr; gap: 1rem; margin-bottom: 2rem; }
+  .compare-card { background: var(--bg-surface); border: 1px solid var(--border); border-radius: var(--radius); padding: 1.25rem; }
+  .compare-card h4 { margin-top: 0; font-size: 0.82rem; }
+  .compare-card.after { border-color: var(--accent-dim); }
+  .compare-card ul { font-family: var(--font-mono); padding-left: 1.25rem; font-size: 0.8rem; }
+
+  hr { border: none; border-top: 1px solid var(--border); margin: 3rem 0; }
+
+  .progress-bar { position: fixed; top: 0; left: 0; height: 2px; background: var(--accent); z-index: 999; transition: width 0.1s linear; }
+
+  @media (max-width: 640px) {
+    .container { padding: 2rem 1rem 4rem; }
+    .hero h1 { font-size: 1.6rem; }
+    .toc ol { columns: 1; }
+    .compare { grid-template-columns: 1fr; }
+    table { font-size: 0.8rem; }
+    th, td { padding: 0.4rem 0.6rem; }
+  }
+</style>
+<link rel="preconnect" href="https://fonts.googleapis.com">
+<link href="https://fonts.googleapis.com/css2?family=Noto+Emoji&display=swap" rel="stylesheet">
+<style>
+  @font-face {
+    font-family: 'Departure Mono';
+    src: url('https://cdn.jsdelivr.net/gh/rektdeckard/departure-mono@latest/fonts/DepartureMono-Regular.woff2') format('woff2');
+    font-weight: normal;
+    font-style: normal;
+    font-display: swap;
+  }
+</style>
+</head>
+<body>
+
+<div class="progress-bar" id="progress"></div>
+
+<div class="container">
+
+<header class="hero">
+  <h1>honcho<span>-integration-spec</span></h1>
+  <p class="subtitle">Comparison of Hermes Agent vs. openclaw-honcho — and a porting spec for bringing Hermes patterns into other Honcho integrations.</p>
+  <div class="meta">
+    <span>hermes-agent / openclaw-honcho</span>
+    <span>Python + TypeScript</span>
+    <span>2026-03-09</span>
+  </div>
+</header>
+
+<nav class="toc">
+  <h2>Contents</h2>
+  <ol>
+    <li><a href="#overview">Overview</a></li>
+    <li><a href="#architecture">Architecture comparison</a></li>
+    <li><a href="#diff-table">Diff table</a></li>
+    <li><a href="#patterns">Hermes patterns to port</a></li>
+    <li><a href="#spec-async">Spec: async prefetch</a></li>
+    <li><a href="#spec-reasoning">Spec: dynamic reasoning level</a></li>
+    <li><a href="#spec-modes">Spec: per-peer memory modes</a></li>
+    <li><a href="#spec-identity">Spec: AI peer identity formation</a></li>
+    <li><a href="#spec-sessions">Spec: session naming strategies</a></li>
+    <li><a href="#spec-cli">Spec: CLI surface injection</a></li>
+    <li><a href="#openclaw-checklist">openclaw-honcho checklist</a></li>
+    <li><a href="#nanobot-checklist">nanobot-honcho checklist</a></li>
+  </ol>
+</nav>
+
+<!-- OVERVIEW -->
+<section id="overview">
+  <h2>Overview</h2>
+
+  <p>Two independent Honcho integrations have been built for two different agent runtimes: <strong>Hermes Agent</strong> (Python, baked into the runner) and <strong>openclaw-honcho</strong> (TypeScript plugin via hook/tool API). Both use the same Honcho peer paradigm — dual peer model, <code>session.context()</code>, <code>peer.chat()</code> — but they made different tradeoffs at every layer.</p>
+
+  <p>This document maps those tradeoffs and defines a porting spec: a set of Hermes-originated patterns, each stated as an integration-agnostic interface, that any Honcho integration can adopt regardless of runtime or language.</p>
+
+  <div class="callout">
+    <strong>Scope</strong> Both integrations work correctly today. This spec is about the delta — patterns in Hermes that are worth propagating and patterns in openclaw-honcho that Hermes should eventually adopt. The spec is additive, not prescriptive.
+  </div>
+</section>
+
+<!-- ARCHITECTURE -->
+<section id="architecture">
+  <h2>Architecture comparison</h2>
+
+  <h3>Hermes: baked-in runner</h3>
+  <p>Honcho is initialised directly inside <code>AIAgent.__init__</code>. There is no plugin boundary. Session management, context injection, async prefetch, and CLI surface are all first-class concerns of the runner. Context is injected once per session (baked into <code>_cached_system_prompt</code>) and never re-fetched mid-session — this maximises prefix cache hits at the LLM provider.</p>
+
+  <div class="mermaid">
+%%{init: {'theme': 'dark', 'themeVariables': { 'primaryColor': '#1f3150', 'primaryTextColor': '#c9d1d9', 'primaryBorderColor': '#3d6ea5', 'lineColor': '#3d6ea5', 'secondaryColor': '#162030', 'tertiaryColor': '#11151c' }}}%%
+flowchart TD
+    U["user message"] --> P["_honcho_prefetch()<br/>(reads cache — no HTTP)"]
+    P --> SP["_build_system_prompt()<br/>(first turn only, cached)"]
+    SP --> LLM["LLM call"]
+    LLM --> R["response"]
+    R --> FP["_honcho_fire_prefetch()<br/>(daemon threads, turn end)"]
+    FP --> C1["prefetch_context() thread"]
+    FP --> C2["prefetch_dialectic() thread"]
+    C1 --> CACHE["_context_cache / _dialectic_cache"]
+    C2 --> CACHE
+
+    style U fill:#162030,stroke:#3d6ea5,color:#c9d1d9
+    style P fill:#1f3150,stroke:#3d6ea5,color:#c9d1d9
+    style SP fill:#1f3150,stroke:#3d6ea5,color:#c9d1d9
+    style LLM fill:#162030,stroke:#3d6ea5,color:#c9d1d9
+    style R fill:#162030,stroke:#3d6ea5,color:#c9d1d9
+    style FP fill:#2a1a40,stroke:#bc8cff,color:#c9d1d9
+    style C1 fill:#2a1a40,stroke:#bc8cff,color:#c9d1d9
+    style C2 fill:#2a1a40,stroke:#bc8cff,color:#c9d1d9
+    style CACHE fill:#11151c,stroke:#484f58,color:#6e7681
+  </div>
+
+  <h3>openclaw-honcho: hook-based plugin</h3>
+  <p>The plugin registers hooks against OpenClaw's event bus. Context is fetched synchronously inside <code>before_prompt_build</code> on every turn. Message capture happens in <code>agent_end</code>. The multi-agent hierarchy is tracked via <code>subagent_spawned</code>. This model is correct but every turn pays a blocking Honcho round-trip before the LLM call can begin.</p>
+
+  <div class="mermaid">
+%%{init: {'theme': 'dark', 'themeVariables': { 'primaryColor': '#1f3150', 'primaryTextColor': '#c9d1d9', 'primaryBorderColor': '#3d6ea5', 'lineColor': '#3d6ea5', 'secondaryColor': '#162030', 'tertiaryColor': '#11151c' }}}%%
+flowchart TD
+    U2["user message"] --> BPB["before_prompt_build<br/>(BLOCKING HTTP — every turn)"]
+    BPB --> CTX["session.context()"]
+    CTX --> SP2["system prompt assembled"]
+    SP2 --> LLM2["LLM call"]
+    LLM2 --> R2["response"]
+    R2 --> AE["agent_end hook"]
+    AE --> SAVE["session.addMessages()<br/>session.setMetadata()"]
+
+    style U2 fill:#162030,stroke:#3d6ea5,color:#c9d1d9
+    style BPB fill:#3a1515,stroke:#f47067,color:#c9d1d9
+    style CTX fill:#3a1515,stroke:#f47067,color:#c9d1d9
+    style SP2 fill:#1f3150,stroke:#3d6ea5,color:#c9d1d9
+    style LLM2 fill:#162030,stroke:#3d6ea5,color:#c9d1d9
+    style R2 fill:#162030,stroke:#3d6ea5,color:#c9d1d9
+    style AE fill:#162030,stroke:#3d6ea5,color:#c9d1d9
+    style SAVE fill:#11151c,stroke:#484f58,color:#6e7681
+  </div>
+</section>
+
+<!-- DIFF TABLE -->
+<section id="diff-table">
+  <h2>Diff table</h2>
+
+  <div class="table-wrap">
+    <table>
+      <thead>
+        <tr>
+          <th>Dimension</th>
+          <th>Hermes Agent</th>
+          <th>openclaw-honcho</th>
+        </tr>
+      </thead>
+      <tbody>
+        <tr>
+          <td><strong>Context injection timing</strong></td>
+          <td>Once per session (cached). Zero HTTP on response path after turn 1.</td>
+          <td>Every turn, blocking. Fresh context per turn but adds latency.</td>
+        </tr>
+        <tr>
+          <td><strong>Prefetch strategy</strong></td>
+          <td>Daemon threads fire at turn end; consumed next turn from cache.</td>
+          <td>None. Blocking call at prompt-build time.</td>
+        </tr>
+        <tr>
+          <td><strong>Dialectic (peer.chat)</strong></td>
+          <td>Prefetched async; result injected into system prompt next turn.</td>
+          <td>On-demand via <code>honcho_recall</code> / <code>honcho_analyze</code> tools.</td>
+        </tr>
+        <tr>
+          <td><strong>Reasoning level</strong></td>
+          <td>Dynamic: scales with message length. Floor = config default. Cap = "high".</td>
+          <td>Fixed per tool: recall=minimal, analyze=medium.</td>
+        </tr>
+        <tr>
+          <td><strong>Memory modes</strong></td>
+          <td><code>user_memory_mode</code> / <code>agent_memory_mode</code>: hybrid / honcho / local.</td>
+          <td>None. Always writes to Honcho.</td>
+        </tr>
+        <tr>
+          <td><strong>Write frequency</strong></td>
+          <td>async (background queue), turn, session, N turns.</td>
+          <td>After every agent_end (no control).</td>
+        </tr>
+        <tr>
+          <td><strong>AI peer identity</strong></td>
+          <td><code>observe_me=True</code>, <code>seed_ai_identity()</code>, <code>get_ai_representation()</code>, SOUL.md → AI peer.</td>
+          <td>Agent files uploaded to agent peer at setup. No ongoing self-observation seeding.</td>
+        </tr>
+        <tr>
+          <td><strong>Context scope</strong></td>
+          <td>User peer + AI peer representation, both injected.</td>
+          <td>User peer (owner) representation + conversation summary. <code>peerPerspective</code> on context call.</td>
+        </tr>
+        <tr>
+          <td><strong>Session naming</strong></td>
+          <td>per-directory / global / manual map / title-based.</td>
+          <td>Derived from platform session key.</td>
+        </tr>
+        <tr>
+          <td><strong>Multi-agent</strong></td>
+          <td>Single-agent only.</td>
+          <td>Parent observer hierarchy via <code>subagent_spawned</code>.</td>
+        </tr>
+        <tr>
+          <td><strong>Tool surface</strong></td>
+          <td>Single <code>query_user_context</code> tool (on-demand dialectic).</td>
+          <td>6 tools: session, profile, search, context (fast) + recall, analyze (LLM).</td>
+        </tr>
+        <tr>
+          <td><strong>Platform metadata</strong></td>
+          <td>Not stripped.</td>
+          <td>Explicitly stripped before Honcho storage.</td>
+        </tr>
+        <tr>
+          <td><strong>Message dedup</strong></td>
+          <td>None (sends on every save cycle).</td>
+          <td><code>lastSavedIndex</code> in session metadata prevents re-sending.</td>
+        </tr>
+        <tr>
+          <td><strong>CLI surface in prompt</strong></td>
+          <td>Management commands injected into system prompt. Agent knows its own CLI.</td>
+          <td>Not injected.</td>
+        </tr>
+        <tr>
+          <td><strong>AI peer name in identity</strong></td>
+          <td>Replaces "Hermes Agent" in DEFAULT_AGENT_IDENTITY when configured.</td>
+          <td>Not implemented.</td>
+        </tr>
+        <tr>
+          <td><strong>QMD / local file search</strong></td>
+          <td>Not implemented.</td>
+          <td>Passthrough tools when QMD backend configured.</td>
+        </tr>
+        <tr>
+          <td><strong>Workspace metadata</strong></td>
+          <td>Not implemented.</td>
+          <td><code>agentPeerMap</code> in workspace metadata tracks agent&#8594;peer ID.</td>
+        </tr>
+      </tbody>
+    </table>
+  </div>
+</section>
+
+<!-- PATTERNS -->
+<section id="patterns">
+  <h2>Hermes patterns to port</h2>
+
+  <p>Six patterns from Hermes are worth adopting in any Honcho integration. They are described below as integration-agnostic interfaces — the implementation will differ per runtime, but the contract is the same.</p>
+
+  <div class="compare">
+    <div class="compare-card">
+      <h4>Patterns Hermes contributes</h4>
+      <ul>
+        <li>Async prefetch (zero-latency)</li>
+        <li>Dynamic reasoning level</li>
+        <li>Per-peer memory modes</li>
+        <li>AI peer identity formation</li>
+        <li>Session naming strategies</li>
+        <li>CLI surface injection</li>
+      </ul>
+    </div>
+    <div class="compare-card after">
+      <h4>Patterns openclaw contributes back</h4>
+      <ul>
+        <li>lastSavedIndex dedup</li>
+        <li>Platform metadata stripping</li>
+        <li>Multi-agent observer hierarchy</li>
+        <li>peerPerspective on context()</li>
+        <li>Tiered tool surface (fast/LLM)</li>
+        <li>Workspace agentPeerMap</li>
+      </ul>
+    </div>
+  </div>
+</section>
+
+<!-- SPEC: ASYNC PREFETCH -->
+<section id="spec-async">
+  <h2>Spec: async prefetch</h2>
+
+  <h3>Problem</h3>
+  <p>Calling <code>session.context()</code> and <code>peer.chat()</code> synchronously before each LLM call adds 200–800ms of Honcho round-trip latency to every turn. Users experience this as the agent "thinking slowly."</p>
+
+  <h3>Pattern</h3>
+  <p>Fire both calls as non-blocking background work at the <strong>end</strong> of each turn. Store results in a per-session cache keyed by session ID. At the <strong>start</strong> of the next turn, pop from cache — the HTTP is already done. First turn is cold (empty cache); all subsequent turns are zero-latency on the response path.</p>
+
+  <h3>Interface contract</h3>
+  <pre><code><span class="cm">// TypeScript (openclaw / nanobot plugin shape)</span>
+
+<span class="kw">interface</span> <span class="key">AsyncPrefetch</span> {
+  <span class="cm">// Fire context + dialectic fetches at turn end. Non-blocking.</span>
+  firePrefetch(sessionId: <span class="str">string</span>, userMessage: <span class="str">string</span>): <span class="kw">void</span>;
+
+  <span class="cm">// Pop cached results at turn start. Returns empty if cache is cold.</span>
+  popContextResult(sessionId: <span class="str">string</span>): ContextResult | <span class="kw">null</span>;
+  popDialecticResult(sessionId: <span class="str">string</span>): <span class="str">string</span> | <span class="kw">null</span>;
+}
+
+<span class="kw">type</span> <span class="key">ContextResult</span> = {
+  representation: <span class="str">string</span>;
+  card: <span class="str">string</span>[];
+  aiRepresentation?: <span class="str">string</span>;  <span class="cm">// AI peer context if enabled</span>
+  summary?: <span class="str">string</span>;            <span class="cm">// conversation summary if fetched</span>
+};</code></pre>
+
+  <h3>Implementation notes</h3>
+  <ul>
+    <li>Python: <code>threading.Thread(daemon=True)</code>. Write to <code>dict[session_id, result]</code> — GIL makes this safe for simple writes.</li>
+    <li>TypeScript: <code>Promise</code> stored in <code>Map&lt;string, Promise&lt;ContextResult&gt;&gt;</code>. Await at pop time. If not resolved yet, skip (return null) — do not block.</li>
+    <li>The pop is destructive: clears the cache entry after reading so stale data never accumulates.</li>
+    <li>Prefetch should also fire on first turn (even though it won't be consumed until turn 2) — this ensures turn 2 is never cold.</li>
+  </ul>
+
+  <h3>openclaw-honcho adoption</h3>
+  <p>Move <code>session.context()</code> from <code>before_prompt_build</code> to a post-<code>agent_end</code> background task. Store result in <code>state.contextCache</code>. In <code>before_prompt_build</code>, read from cache instead of calling Honcho. If cache is empty (turn 1), inject nothing — the prompt is still valid without Honcho context on the first turn.</p>
+</section>
+
+<!-- SPEC: DYNAMIC REASONING LEVEL -->
+<section id="spec-reasoning">
+  <h2>Spec: dynamic reasoning level</h2>
+
+  <h3>Problem</h3>
+  <p>Honcho's dialectic endpoint supports reasoning levels from <code>minimal</code> to <code>max</code>. A fixed level per tool wastes budget on simple queries and under-serves complex ones.</p>
+
+  <h3>Pattern</h3>
+  <p>Select the reasoning level dynamically based on the user's message. Use the configured default as a floor. Bump by message length. Cap auto-selection at <code>high</code> — never select <code>max</code> automatically.</p>
+
+  <h3>Interface contract</h3>
+  <pre><code><span class="cm">// Shared helper — identical logic in any language</span>
+
+<span class="kw">const</span> LEVELS = [<span class="str">"minimal"</span>, <span class="str">"low"</span>, <span class="str">"medium"</span>, <span class="str">"high"</span>, <span class="str">"max"</span>];
+
+<span class="kw">function</span> <span class="key">dynamicReasoningLevel</span>(
+  query: <span class="str">string</span>,
+  configDefault: <span class="str">string</span> = <span class="str">"low"</span>
+): <span class="str">string</span> {
+  <span class="kw">const</span> baseIdx = Math.max(<span class="num">0</span>, LEVELS.indexOf(configDefault));
+  <span class="kw">const</span> n = query.length;
+  <span class="kw">const</span> bump = n &lt; <span class="num">120</span> ? <span class="num">0</span> : n &lt; <span class="num">400</span> ? <span class="num">1</span> : <span class="num">2</span>;
+  <span class="kw">return</span> LEVELS[Math.min(baseIdx + bump, <span class="num">3</span>)]; <span class="cm">// cap at "high" (idx 3)</span>
+}</code></pre>
+
+  <h3>Config key</h3>
+  <p>Add a <code>dialecticReasoningLevel</code> config field (string, default <code>"low"</code>). This sets the floor. Users can raise or lower it. The dynamic bump always applies on top.</p>
+
+  <h3>openclaw-honcho adoption</h3>
+  <p>Apply in <code>honcho_recall</code> and <code>honcho_analyze</code>: replace the fixed <code>reasoningLevel</code> with the dynamic selector. <code>honcho_recall</code> should use floor <code>"minimal"</code> and <code>honcho_analyze</code> floor <code>"medium"</code> — both still bump with message length.</p>
+</section>
+
+<!-- SPEC: PER-PEER MEMORY MODES -->
+<section id="spec-modes">
+  <h2>Spec: per-peer memory modes</h2>
+
+  <h3>Problem</h3>
+  <p>Users want independent control over whether user context and agent context are written locally, to Honcho, or both. A single <code>memoryMode</code> shorthand is not granular enough.</p>
+
+  <h3>Pattern</h3>
+  <p>Three modes per peer: <code>hybrid</code> (write both local + Honcho), <code>honcho</code> (Honcho only, disable local files), <code>local</code> (local files only, skip Honcho sync for this peer). Two orthogonal axes: user peer and agent peer.</p>
+
+  <h3>Config schema</h3>
+  <pre><code><span class="cm">// ~/.openclaw/openclaw.json  (or ~/.nanobot/config.json)</span>
+{
+  <span class="str">"plugins"</span>: {
+    <span class="str">"openclaw-honcho"</span>: {
+      <span class="str">"config"</span>: {
+        <span class="str">"apiKey"</span>: <span class="str">"..."</span>,
+        <span class="str">"memoryMode"</span>: <span class="str">"hybrid"</span>,          <span class="cm">// shorthand: both peers</span>
+        <span class="str">"userMemoryMode"</span>: <span class="str">"honcho"</span>,       <span class="cm">// override for user peer</span>
+        <span class="str">"agentMemoryMode"</span>: <span class="str">"hybrid"</span>       <span class="cm">// override for agent peer</span>
+      }
+    }
+  }
+}</code></pre>
+
+  <h3>Resolution order</h3>
+  <ol>
+    <li>Per-peer field (<code>userMemoryMode</code> / <code>agentMemoryMode</code>) — wins if present.</li>
+    <li>Shorthand <code>memoryMode</code> — applies to both peers as default.</li>
+    <li>Hardcoded default: <code>"hybrid"</code>.</li>
+  </ol>
+
+  <h3>Effect on Honcho sync</h3>
+  <ul>
+    <li><code>userMemoryMode=local</code>: skip adding user peer messages to Honcho.</li>
+    <li><code>agentMemoryMode=local</code>: skip adding assistant peer messages to Honcho.</li>
+    <li>Both local: skip <code>session.addMessages()</code> entirely.</li>
+    <li><code>userMemoryMode=honcho</code>: disable local USER.md writes.</li>
+    <li><code>agentMemoryMode=honcho</code>: disable local MEMORY.md / SOUL.md writes.</li>
+  </ul>
+</section>
+
+<!-- SPEC: AI PEER IDENTITY -->
+<section id="spec-identity">
+  <h2>Spec: AI peer identity formation</h2>
+
+  <h3>Problem</h3>
+  <p>Honcho builds the user's representation organically by observing what the user says. The same mechanism exists for the AI peer — but only if <code>observe_me=True</code> is set for the agent peer. Without it, the agent peer accumulates nothing and Honcho's AI-side model never forms.</p>
+
+  <p>Additionally, existing persona files (SOUL.md, IDENTITY.md) should seed the AI peer's Honcho representation at first activation, rather than waiting for it to emerge from scratch.</p>
+
+  <h3>Part A: observe_me=True for agent peer</h3>
+  <pre><code><span class="cm">// TypeScript — in session.addPeers() call</span>
+<span class="kw">await</span> session.addPeers([
+  [ownerPeer.id, { observeMe: <span class="kw">true</span>,  observeOthers: <span class="kw">false</span> }],
+  [agentPeer.id, { observeMe: <span class="kw">true</span>,  observeOthers: <span class="kw">true</span>  }], <span class="cm">// was false</span>
+]);</code></pre>
+
+  <p>This is a one-line change but foundational. Without it, Honcho's AI peer representation stays empty regardless of what the agent says.</p>
+
+  <h3>Part B: seedAiIdentity()</h3>
+  <pre><code><span class="kw">async function</span> <span class="key">seedAiIdentity</span>(
+  session: HonchoSession,
+  agentPeer: Peer,
+  content: <span class="str">string</span>,
+  source: <span class="str">string</span>
+): Promise&lt;<span class="kw">boolean</span>&gt; {
+  <span class="kw">const</span> wrapped = [
+    <span class="str">`&lt;ai_identity_seed&gt;`</span>,
+    <span class="str">`&lt;source&gt;${source}&lt;/source&gt;`</span>,
+    <span class="str">``</span>,
+    content.trim(),
+    <span class="str">`&lt;/ai_identity_seed&gt;`</span>,
+  ].join(<span class="str">"\n"</span>);
+
+  <span class="kw">await</span> agentPeer.addMessage(<span class="str">"assistant"</span>, wrapped);
+  <span class="kw">return true</span>;
+}</code></pre>
+
+  <h3>Part C: migrate agent files at setup</h3>
+  <p>During <code>openclaw honcho setup</code>, upload agent-self files (SOUL.md, IDENTITY.md, AGENTS.md, BOOTSTRAP.md) to the agent peer using <code>seedAiIdentity()</code> instead of <code>session.uploadFile()</code>. This routes the content through Honcho's observation pipeline rather than the file store.</p>
+
+  <h3>Part D: AI peer name in identity</h3>
+  <p>When the agent has a configured name (non-default), inject it into the agent's self-identity prefix. In OpenClaw this means adding to the injected system prompt section:</p>
+  <pre><code><span class="cm">// In context hook return value</span>
+<span class="kw">return</span> {
+  systemPrompt: [
+    agentName ? <span class="str">`You are ${agentName}.`</span> : <span class="str">""</span>,
+    <span class="str">"## User Memory Context"</span>,
+    ...sections,
+  ].filter(Boolean).join(<span class="str">"\n\n"</span>)
+};</code></pre>
+
+  <h3>CLI surface: honcho identity subcommand</h3>
+  <pre><code>openclaw honcho identity &lt;file&gt;    <span class="cm"># seed from file</span>
+openclaw honcho identity --show    <span class="cm"># show current AI peer representation</span></code></pre>
+</section>
+
+<!-- SPEC: SESSION NAMING -->
+<section id="spec-sessions">
+  <h2>Spec: session naming strategies</h2>
+
+  <h3>Problem</h3>
+  <p>When Honcho is used across multiple projects or directories, a single global session means every project shares the same context. Per-directory sessions provide isolation without requiring users to name sessions manually.</p>
+
+  <h3>Strategies</h3>
+  <div class="table-wrap">
+    <table>
+      <thead><tr><th>Strategy</th><th>Session key</th><th>When to use</th></tr></thead>
+      <tbody>
+        <tr><td><code>per-directory</code></td><td>basename of CWD</td><td>Default. Each project gets its own session.</td></tr>
+        <tr><td><code>global</code></td><td>fixed string <code>"global"</code></td><td>Single cross-project session.</td></tr>
+        <tr><td>manual map</td><td>user-configured per path</td><td><code>sessions</code> config map overrides directory basename.</td></tr>
+        <tr><td>title-based</td><td>sanitized session title</td><td>When agent supports named sessions; title set mid-conversation.</td></tr>
+      </tbody>
+    </table>
+  </div>
+
+  <h3>Config schema</h3>
+  <pre><code>{
+  <span class="str">"sessionStrategy"</span>: <span class="str">"per-directory"</span>,   <span class="cm">// "per-directory" | "global"</span>
+  <span class="str">"sessionPeerPrefix"</span>: <span class="kw">false</span>,            <span class="cm">// prepend peer name to session key</span>
+  <span class="str">"sessions"</span>: {                            <span class="cm">// manual overrides</span>
+    <span class="str">"/home/user/projects/foo"</span>: <span class="str">"foo-project"</span>
+  }
+}</code></pre>
+
+  <h3>CLI surface</h3>
+  <pre><code>openclaw honcho sessions              <span class="cm"># list all mappings</span>
+openclaw honcho map &lt;name&gt;           <span class="cm"># map cwd to session name</span>
+openclaw honcho map                   <span class="cm"># no-arg = list mappings</span></code></pre>
+
+  <p>Resolution order: manual map wins &rarr; session title &rarr; directory basename &rarr; platform key.</p>
+</section>
+
+<!-- SPEC: CLI SURFACE INJECTION -->
+<section id="spec-cli">
+  <h2>Spec: CLI surface injection</h2>
+
+  <h3>Problem</h3>
+  <p>When a user asks "how do I change my memory settings?" or "what Honcho commands are available?" the agent either hallucinates or says it doesn't know. The agent should know its own management interface.</p>
+
+  <h3>Pattern</h3>
+  <p>When Honcho is active, append a compact command reference to the system prompt. The agent can cite these commands directly instead of guessing.</p>
+
+  <pre><code><span class="cm">// In context hook, append to systemPrompt</span>
+<span class="kw">const</span> honchoSection = [
+  <span class="str">"# Honcho memory integration"</span>,
+  <span class="str">`Active. Session: ${sessionKey}. Mode: ${mode}.`</span>,
+  <span class="str">"Management commands:"</span>,
+  <span class="str">"  openclaw honcho status                    — show config + connection"</span>,
+  <span class="str">"  openclaw honcho mode [hybrid|honcho|local] — show or set memory mode"</span>,
+  <span class="str">"  openclaw honcho sessions                  — list session mappings"</span>,
+  <span class="str">"  openclaw honcho map &lt;name&gt;                — map directory to session"</span>,
+  <span class="str">"  openclaw honcho identity [file] [--show]  — seed or show AI identity"</span>,
+  <span class="str">"  openclaw honcho setup                     — full interactive wizard"</span>,
+].join(<span class="str">"\n"</span>);</code></pre>
+
+  <div class="callout warn">
+    <strong>Keep it compact.</strong> This section is injected every turn. Keep it under 300 chars of context. List commands, not explanations — the agent can explain them on request.
+  </div>
+</section>
+
+<!-- OPENCLAW CHECKLIST -->
+<section id="openclaw-checklist">
+  <h2>openclaw-honcho checklist</h2>
+
+  <p>Ordered by impact. Each item maps to a spec section above.</p>
+
+  <ul class="checklist">
+    <li class="todo"><strong>Async prefetch</strong> — move <code>session.context()</code> out of <code>before_prompt_build</code> into post-<code>agent_end</code> background Promise. Pop from cache at prompt build. (<a href="#spec-async">spec</a>)</li>
+    <li class="todo"><strong>observe_me=True for agent peer</strong> — one-line change in <code>session.addPeers()</code> config for agent peer. (<a href="#spec-identity">spec</a>)</li>
+    <li class="todo"><strong>Dynamic reasoning level</strong> — add <code>dynamicReasoningLevel()</code> helper; apply in <code>honcho_recall</code> and <code>honcho_analyze</code>. Add <code>dialecticReasoningLevel</code> to config schema. (<a href="#spec-reasoning">spec</a>)</li>
+    <li class="todo"><strong>Per-peer memory modes</strong> — add <code>userMemoryMode</code> / <code>agentMemoryMode</code> to config; gate Honcho sync and local writes accordingly. (<a href="#spec-modes">spec</a>)</li>
+    <li class="todo"><strong>seedAiIdentity()</strong> — add helper; apply during setup migration for SOUL.md / IDENTITY.md instead of <code>session.uploadFile()</code>. (<a href="#spec-identity">spec</a>)</li>
+    <li class="todo"><strong>Session naming strategies</strong> — add <code>sessionStrategy</code>, <code>sessions</code> map, <code>sessionPeerPrefix</code> to config; implement resolution function. (<a href="#spec-sessions">spec</a>)</li>
+    <li class="todo"><strong>CLI surface injection</strong> — append command reference to <code>before_prompt_build</code> return value when Honcho is active. (<a href="#spec-cli">spec</a>)</li>
+    <li class="todo"><strong>honcho identity subcommand</strong> — add <code>openclaw honcho identity</code> CLI command. (<a href="#spec-identity">spec</a>)</li>
+    <li class="todo"><strong>AI peer name injection</strong> — if <code>aiPeer</code> name configured, prepend to injected system prompt. (<a href="#spec-identity">spec</a>)</li>
+    <li class="todo"><strong>honcho mode / honcho sessions / honcho map</strong> — CLI parity with Hermes. (<a href="#spec-sessions">spec</a>)</li>
+  </ul>
+
+  <div class="callout success">
+    <strong>Already done in openclaw-honcho (do not re-implement):</strong> lastSavedIndex dedup, platform metadata stripping, multi-agent parent observer hierarchy, peerPerspective on context(), tiered tool surface (fast/LLM), workspace agentPeerMap, QMD passthrough, self-hosted Honcho support.
+  </div>
+</section>
+
+<!-- NANOBOT CHECKLIST -->
+<section id="nanobot-checklist">
+  <h2>nanobot-honcho checklist</h2>
+
+  <p>nanobot-honcho is a greenfield integration. Start from openclaw-honcho's architecture (hook-based, dual peer) and apply all Hermes patterns from day one rather than retrofitting. Priority order:</p>
+
+  <h3>Phase 1 — core correctness</h3>
+  <ul class="checklist">
+    <li class="todo">Dual peer model (owner + agent peer), both with <code>observe_me=True</code></li>
+    <li class="todo">Message capture at turn end with <code>lastSavedIndex</code> dedup</li>
+    <li class="todo">Platform metadata stripping before Honcho storage</li>
+    <li class="todo">Async prefetch from day one — do not implement blocking context injection</li>
+    <li class="todo">Legacy file migration at first activation (USER.md → owner peer, SOUL.md → <code>seedAiIdentity()</code>)</li>
+  </ul>
+
+  <h3>Phase 2 — configuration</h3>
+  <ul class="checklist">
+    <li class="todo">Config schema: <code>apiKey</code>, <code>workspaceId</code>, <code>baseUrl</code>, <code>memoryMode</code>, <code>userMemoryMode</code>, <code>agentMemoryMode</code>, <code>dialecticReasoningLevel</code>, <code>sessionStrategy</code>, <code>sessions</code></li>
+    <li class="todo">Per-peer memory mode gating</li>
+    <li class="todo">Dynamic reasoning level</li>
+    <li class="todo">Session naming strategies</li>
+  </ul>
+
+  <h3>Phase 3 — tools and CLI</h3>
+  <ul class="checklist">
+    <li class="todo">Tool surface: <code>honcho_profile</code>, <code>honcho_recall</code>, <code>honcho_analyze</code>, <code>honcho_search</code>, <code>honcho_context</code></li>
+    <li class="todo">CLI: <code>setup</code>, <code>status</code>, <code>sessions</code>, <code>map</code>, <code>mode</code>, <code>identity</code></li>
+    <li class="todo">CLI surface injection into system prompt</li>
+    <li class="todo">AI peer name wired into agent identity</li>
+  </ul>
+</section>
+
+</div>
+
+<script type="module">
+  import mermaid from 'https://cdn.jsdelivr.net/npm/mermaid@11/dist/mermaid.esm.min.mjs';
+  mermaid.initialize({ startOnLoad: true, securityLevel: 'loose', fontFamily: 'Departure Mono, Noto Emoji, monospace' });
+</script>
+<script>
+  window.addEventListener('scroll', () => {
+    const bar = document.getElementById('progress');
+    const max = document.documentElement.scrollHeight - window.innerHeight;
+    bar.style.width = (max > 0 ? (window.scrollY / max) * 100 : 0) + '%';
+  });
+</script>
+</body>
+</html>

docs/honcho-integration-spec.md

@@ -0,0 +1,377 @@
+# honcho-integration-spec
+
+Comparison of Hermes Agent vs. openclaw-honcho — and a porting spec for bringing Hermes patterns into other Honcho integrations.
+
+---
+
+## Overview
+
+Two independent Honcho integrations have been built for two different agent runtimes: **Hermes Agent** (Python, baked into the runner) and **openclaw-honcho** (TypeScript plugin via hook/tool API). Both use the same Honcho peer paradigm — dual peer model, `session.context()`, `peer.chat()` — but they made different tradeoffs at every layer.
+
+This document maps those tradeoffs and defines a porting spec: a set of Hermes-originated patterns, each stated as an integration-agnostic interface, that any Honcho integration can adopt regardless of runtime or language.
+
+> **Scope** Both integrations work correctly today. This spec is about the delta — patterns in Hermes that are worth propagating and patterns in openclaw-honcho that Hermes should eventually adopt. The spec is additive, not prescriptive.
+
+---
+
+## Architecture comparison
+
+### Hermes: baked-in runner
+
+Honcho is initialised directly inside `AIAgent.__init__`. There is no plugin boundary. Session management, context injection, async prefetch, and CLI surface are all first-class concerns of the runner. Context is injected once per session (baked into `_cached_system_prompt`) and never re-fetched mid-session — this maximises prefix cache hits at the LLM provider.
+
+Turn flow:
+
+```
+user message
+  → _honcho_prefetch()       (reads cache — no HTTP)
+  → _build_system_prompt()   (first turn only, cached)
+  → LLM call
+  → response
+  → _honcho_fire_prefetch()  (daemon threads, turn end)
+       → prefetch_context() thread  ──┐
+       → prefetch_dialectic() thread ─┴→ _context_cache / _dialectic_cache
+```
+
+### openclaw-honcho: hook-based plugin
+
+The plugin registers hooks against OpenClaw's event bus. Context is fetched synchronously inside `before_prompt_build` on every turn. Message capture happens in `agent_end`. The multi-agent hierarchy is tracked via `subagent_spawned`. This model is correct but every turn pays a blocking Honcho round-trip before the LLM call can begin.
+
+Turn flow:
+
+```
+user message
+  → before_prompt_build (BLOCKING HTTP — every turn)
+       → session.context()
+  → system prompt assembled
+  → LLM call
+  → response
+  → agent_end hook
+       → session.addMessages()
+       → session.setMetadata()
+```
+
+---
+
+## Diff table
+
+| Dimension | Hermes Agent | openclaw-honcho |
+|---|---|---|
+| **Context injection timing** | Once per session (cached). Zero HTTP on response path after turn 1. | Every turn, blocking. Fresh context per turn but adds latency. |
+| **Prefetch strategy** | Daemon threads fire at turn end; consumed next turn from cache. | None. Blocking call at prompt-build time. |
+| **Dialectic (peer.chat)** | Prefetched async; result injected into system prompt next turn. | On-demand via `honcho_recall` / `honcho_analyze` tools. |
+| **Reasoning level** | Dynamic: scales with message length. Floor = config default. Cap = "high". | Fixed per tool: recall=minimal, analyze=medium. |
+| **Memory modes** | `user_memory_mode` / `agent_memory_mode`: hybrid / honcho / local. | None. Always writes to Honcho. |
+| **Write frequency** | async (background queue), turn, session, N turns. | After every agent_end (no control). |
+| **AI peer identity** | `observe_me=True`, `seed_ai_identity()`, `get_ai_representation()`, SOUL.md → AI peer. | Agent files uploaded to agent peer at setup. No ongoing self-observation. |
+| **Context scope** | User peer + AI peer representation, both injected. | User peer (owner) representation + conversation summary. `peerPerspective` on context call. |
+| **Session naming** | per-directory / global / manual map / title-based. | Derived from platform session key. |
+| **Multi-agent** | Single-agent only. | Parent observer hierarchy via `subagent_spawned`. |
+| **Tool surface** | Single `query_user_context` tool (on-demand dialectic). | 6 tools: session, profile, search, context (fast) + recall, analyze (LLM). |
+| **Platform metadata** | Not stripped. | Explicitly stripped before Honcho storage. |
+| **Message dedup** | None. | `lastSavedIndex` in session metadata prevents re-sending. |
+| **CLI surface in prompt** | Management commands injected into system prompt. Agent knows its own CLI. | Not injected. |
+| **AI peer name in identity** | Replaces "Hermes Agent" in DEFAULT_AGENT_IDENTITY when configured. | Not implemented. |
+| **QMD / local file search** | Not implemented. | Passthrough tools when QMD backend configured. |
+| **Workspace metadata** | Not implemented. | `agentPeerMap` in workspace metadata tracks agent→peer ID. |
+
+---
+
+## Patterns
+
+Six patterns from Hermes are worth adopting in any Honcho integration. Each is described as an integration-agnostic interface.
+
+**Hermes contributes:**
+- Async prefetch (zero-latency)
+- Dynamic reasoning level
+- Per-peer memory modes
+- AI peer identity formation
+- Session naming strategies
+- CLI surface injection
+
+**openclaw-honcho contributes back (Hermes should adopt):**
+- `lastSavedIndex` dedup
+- Platform metadata stripping
+- Multi-agent observer hierarchy
+- `peerPerspective` on `context()`
+- Tiered tool surface (fast/LLM)
+- Workspace `agentPeerMap`
+
+---
+
+## Spec: async prefetch
+
+### Problem
+
+Calling `session.context()` and `peer.chat()` synchronously before each LLM call adds 200–800ms of Honcho round-trip latency to every turn.
+
+### Pattern
+
+Fire both calls as non-blocking background work at the **end** of each turn. Store results in a per-session cache keyed by session ID. At the **start** of the next turn, pop from cache — the HTTP is already done. First turn is cold (empty cache); all subsequent turns are zero-latency on the response path.
+
+### Interface contract
+
+```typescript
+interface AsyncPrefetch {
+  // Fire context + dialectic fetches at turn end. Non-blocking.
+  firePrefetch(sessionId: string, userMessage: string): void;
+
+  // Pop cached results at turn start. Returns empty if cache is cold.
+  popContextResult(sessionId: string): ContextResult | null;
+  popDialecticResult(sessionId: string): string | null;
+}
+
+type ContextResult = {
+  representation: string;
+  card: string[];
+  aiRepresentation?: string;  // AI peer context if enabled
+  summary?: string;           // conversation summary if fetched
+};
+```
+
+### Implementation notes
+
+- **Python:** `threading.Thread(daemon=True)`. Write to `dict[session_id, result]` — GIL makes this safe for simple writes.
+- **TypeScript:** `Promise` stored in `Map<string, Promise<ContextResult>>`. Await at pop time. If not resolved yet, return null — do not block.
+- The pop is destructive: clears the cache entry after reading so stale data never accumulates.
+- Prefetch should also fire on first turn (even though it won't be consumed until turn 2).
+
+### openclaw-honcho adoption
+
+Move `session.context()` from `before_prompt_build` to a post-`agent_end` background task. Store result in `state.contextCache`. In `before_prompt_build`, read from cache instead of calling Honcho. If cache is empty (turn 1), inject nothing — the prompt is still valid without Honcho context on the first turn.
+
+---
+
+## Spec: dynamic reasoning level
+
+### Problem
+
+Honcho's dialectic endpoint supports reasoning levels from `minimal` to `max`. A fixed level per tool wastes budget on simple queries and under-serves complex ones.
+
+### Pattern
+
+Select the reasoning level dynamically based on the user's message. Use the configured default as a floor. Bump by message length. Cap auto-selection at `high` — never select `max` automatically.
+
+### Logic
+
+```
+< 120 chars  → default (typically "low")
+120–400 chars → one level above default (cap at "high")
+> 400 chars  → two levels above default (cap at "high")
+```
+
+### Config key
+
+Add `dialecticReasoningLevel` (string, default `"low"`). This sets the floor. The dynamic bump always applies on top.
+
+### openclaw-honcho adoption
+
+Apply in `honcho_recall` and `honcho_analyze`: replace fixed `reasoningLevel` with the dynamic selector. `honcho_recall` uses floor `"minimal"`, `honcho_analyze` uses floor `"medium"` — both still bump with message length.
+
+---
+
+## Spec: per-peer memory modes
+
+### Problem
+
+Users want independent control over whether user context and agent context are written locally, to Honcho, or both.
+
+### Modes
+
+| Mode | Effect |
+|---|---|
+| `hybrid` | Write to both local files and Honcho (default) |
+| `honcho` | Honcho only — disable corresponding local file writes |
+| `local` | Local files only — skip Honcho sync for this peer |
+
+### Config schema
+
+```json
+{
+  "memoryMode": "hybrid",
+  "userMemoryMode": "honcho",
+  "agentMemoryMode": "hybrid"
+}
+```
+
+Resolution order: per-peer field wins → shorthand `memoryMode` → default `"hybrid"`.
+
+### Effect on Honcho sync
+
+- `userMemoryMode=local`: skip adding user peer messages to Honcho
+- `agentMemoryMode=local`: skip adding assistant peer messages to Honcho
+- Both local: skip `session.addMessages()` entirely
+- `userMemoryMode=honcho`: disable local USER.md writes
+- `agentMemoryMode=honcho`: disable local MEMORY.md / SOUL.md writes
+
+---
+
+## Spec: AI peer identity formation
+
+### Problem
+
+Honcho builds the user's representation organically by observing what the user says. The same mechanism exists for the AI peer — but only if `observe_me=True` is set for the agent peer. Without it, the agent peer accumulates nothing.
+
+Additionally, existing persona files (SOUL.md, IDENTITY.md) should seed the AI peer's Honcho representation at first activation.
+
+### Part A: observe_me=True for agent peer
+
+```typescript
+await session.addPeers([
+  [ownerPeer.id, { observeMe: true,  observeOthers: false }],
+  [agentPeer.id, { observeMe: true,  observeOthers: true  }], // was false
+]);
+```
+
+One-line change. Foundational. Without it, the AI peer representation stays empty regardless of what the agent says.
+
+### Part B: seedAiIdentity()
+
+```typescript
+async function seedAiIdentity(
+  agentPeer: Peer,
+  content: string,
+  source: string
+): Promise<boolean> {
+  const wrapped = [
+    `<ai_identity_seed>`,
+    `<source>${source}</source>`,
+    ``,
+    content.trim(),
+    `</ai_identity_seed>`,
+  ].join("\n");
+
+  await agentPeer.addMessage("assistant", wrapped);
+  return true;
+}
+```
+
+### Part C: migrate agent files at setup
+
+During `honcho setup`, upload agent-self files (SOUL.md, IDENTITY.md, AGENTS.md) to the agent peer via `seedAiIdentity()` instead of `session.uploadFile()`. This routes content through Honcho's observation pipeline.
+
+### Part D: AI peer name in identity
+
+When the agent has a configured name, prepend it to the injected system prompt:
+
+```typescript
+const namePrefix = agentName ? `You are ${agentName}.\n\n` : "";
+return { systemPrompt: namePrefix + "## User Memory Context\n\n" + sections };
+```
+
+### CLI surface
+
+```
+honcho identity <file>    # seed from file
+honcho identity --show    # show current AI peer representation
+```
+
+---
+
+## Spec: session naming strategies
+
+### Problem
+
+A single global session means every project shares the same Honcho context. Per-directory sessions provide isolation without requiring users to name sessions manually.
+
+### Strategies
+
+| Strategy | Session key | When to use |
+|---|---|---|
+| `per-directory` | basename of CWD | Default. Each project gets its own session. |
+| `global` | fixed string `"global"` | Single cross-project session. |
+| manual map | user-configured per path | `sessions` config map overrides directory basename. |
+| title-based | sanitized session title | When agent supports named sessions set mid-conversation. |
+
+### Config schema
+
+```json
+{
+  "sessionStrategy": "per-directory",
+  "sessionPeerPrefix": false,
+  "sessions": {
+    "/home/user/projects/foo": "foo-project"
+  }
+}
+```
+
+### CLI surface
+
+```
+honcho sessions              # list all mappings
+honcho map <name>            # map cwd to session name
+honcho map                   # no-arg = list mappings
+```
+
+Resolution order: manual map → session title → directory basename → platform key.
+
+---
+
+## Spec: CLI surface injection
+
+### Problem
+
+When a user asks "how do I change my memory settings?" the agent either hallucinates or says it doesn't know. The agent should know its own management interface.
+
+### Pattern
+
+When Honcho is active, append a compact command reference to the system prompt. Keep it under 300 chars.
+
+```
+# Honcho memory integration
+Active. Session: {sessionKey}. Mode: {mode}.
+Management commands:
+  honcho status                    — show config + connection
+  honcho mode [hybrid|honcho|local] — show or set memory mode
+  honcho sessions                  — list session mappings
+  honcho map <name>                — map directory to session
+  honcho identity [file] [--show]  — seed or show AI identity
+  honcho setup                     — full interactive wizard
+```
+
+---
+
+## openclaw-honcho checklist
+
+Ordered by impact:
+
+- [ ] **Async prefetch** — move `session.context()` out of `before_prompt_build` into post-`agent_end` background Promise
+- [ ] **observe_me=True for agent peer** — one-line change in `session.addPeers()`
+- [ ] **Dynamic reasoning level** — add helper; apply in `honcho_recall` and `honcho_analyze`; add `dialecticReasoningLevel` to config
+- [ ] **Per-peer memory modes** — add `userMemoryMode` / `agentMemoryMode` to config; gate Honcho sync and local writes
+- [ ] **seedAiIdentity()** — add helper; use during setup migration for SOUL.md / IDENTITY.md
+- [ ] **Session naming strategies** — add `sessionStrategy`, `sessions` map, `sessionPeerPrefix`
+- [ ] **CLI surface injection** — append command reference to `before_prompt_build` return value
+- [ ] **honcho identity subcommand** — seed from file or `--show` current representation
+- [ ] **AI peer name injection** — if `aiPeer` name configured, prepend to injected system prompt
+- [ ] **honcho mode / sessions / map** — CLI parity with Hermes
+
+Already done in openclaw-honcho (do not re-implement): `lastSavedIndex` dedup, platform metadata stripping, multi-agent parent observer, `peerPerspective` on `context()`, tiered tool surface, workspace `agentPeerMap`, QMD passthrough, self-hosted Honcho.
+
+---
+
+## nanobot-honcho checklist
+
+Greenfield integration. Start from openclaw-honcho's architecture and apply all Hermes patterns from day one.
+
+### Phase 1 — core correctness
+
+- [ ] Dual peer model (owner + agent peer), both with `observe_me=True`
+- [ ] Message capture at turn end with `lastSavedIndex` dedup
+- [ ] Platform metadata stripping before Honcho storage
+- [ ] Async prefetch from day one — do not implement blocking context injection
+- [ ] Legacy file migration at first activation (USER.md → owner peer, SOUL.md → `seedAiIdentity()`)
+
+### Phase 2 — configuration
+
+- [ ] Config schema: `apiKey`, `workspaceId`, `baseUrl`, `memoryMode`, `userMemoryMode`, `agentMemoryMode`, `dialecticReasoningLevel`, `sessionStrategy`, `sessions`
+- [ ] Per-peer memory mode gating
+- [ ] Dynamic reasoning level
+- [ ] Session naming strategies
+
+### Phase 3 — tools and CLI
+
+- [ ] Tool surface: `honcho_profile`, `honcho_recall`, `honcho_analyze`, `honcho_search`, `honcho_context`
+- [ ] CLI: `setup`, `status`, `sessions`, `map`, `mode`, `identity`
+- [ ] CLI surface injection into system prompt
+- [ ] AI peer name wired into agent identity

docs/migration/openclaw.md

@@ -0,0 +1,110 @@
+# Migrating from OpenClaw to Hermes Agent
+
+This guide covers how to import your OpenClaw settings, memories, skills, and API keys into Hermes Agent.
+
+## Three Ways to Migrate
+
+### 1. Automatic (during first-time setup)
+
+When you run `hermes setup` for the first time and Hermes detects `~/.openclaw`, it automatically offers to import your OpenClaw data before configuration begins. Just accept the prompt and everything is handled for you.
+
+### 2. CLI Command (quick, scriptable)
+
+```bash
+hermes claw migrate                      # Full migration with confirmation prompt
+hermes claw migrate --dry-run            # Preview what would happen
+hermes claw migrate --preset user-data   # Migrate without API keys/secrets
+hermes claw migrate --yes                # Skip confirmation prompt
+```
+
+**All options:**
+
+| Flag | Description |
+|------|-------------|
+| `--source PATH` | Path to OpenClaw directory (default: `~/.openclaw`) |
+| `--dry-run` | Preview only — no files are modified |
+| `--preset {user-data,full}` | Migration preset (default: `full`). `user-data` excludes secrets |
+| `--overwrite` | Overwrite existing files (default: skip conflicts) |
+| `--migrate-secrets` | Include allowlisted secrets (auto-enabled with `full` preset) |
+| `--workspace-target PATH` | Copy workspace instructions (AGENTS.md) to this absolute path |
+| `--skill-conflict {skip,overwrite,rename}` | How to handle skill name conflicts (default: `skip`) |
+| `--yes`, `-y` | Skip confirmation prompts |
+
+### 3. Agent-Guided (interactive, with previews)
+
+Ask the agent to run the migration for you:
+
+```
+> Migrate my OpenClaw setup to Hermes
+```
+
+The agent will use the `openclaw-migration` skill to:
+1. Run a dry-run first to preview changes
+2. Ask about conflict resolution (SOUL.md, skills, etc.)
+3. Let you choose between `user-data` and `full` presets
+4. Execute the migration with your choices
+5. Print a detailed summary of what was migrated
+
+## What Gets Migrated
+
+### `user-data` preset
+| Item | Source | Destination |
+|------|--------|-------------|
+| SOUL.md | `~/.openclaw/workspace/SOUL.md` | `~/.hermes/SOUL.md` |
+| Memory entries | `~/.openclaw/workspace/MEMORY.md` | `~/.hermes/memories/MEMORY.md` |
+| User profile | `~/.openclaw/workspace/USER.md` | `~/.hermes/memories/USER.md` |
+| Skills | `~/.openclaw/workspace/skills/` | `~/.hermes/skills/openclaw-imports/` |
+| Command allowlist | `~/.openclaw/workspace/exec_approval_patterns.yaml` | Merged into `~/.hermes/config.yaml` |
+| Messaging settings | `~/.openclaw/config.yaml` (TELEGRAM_ALLOWED_USERS, MESSAGING_CWD) | `~/.hermes/.env` |
+| TTS assets | `~/.openclaw/workspace/tts/` | `~/.hermes/tts/` |
+
+### `full` preset (adds to `user-data`)
+| Item | Source | Destination |
+|------|--------|-------------|
+| Telegram bot token | `~/.openclaw/config.yaml` | `~/.hermes/.env` |
+| OpenRouter API key | `~/.openclaw/.env` or config | `~/.hermes/.env` |
+| OpenAI API key | `~/.openclaw/.env` or config | `~/.hermes/.env` |
+| Anthropic API key | `~/.openclaw/.env` or config | `~/.hermes/.env` |
+| ElevenLabs API key | `~/.openclaw/.env` or config | `~/.hermes/.env` |
+
+Only these 6 allowlisted secrets are ever imported. Other credentials are skipped and reported.
+
+## Conflict Handling
+
+By default, the migration **will not overwrite** existing Hermes data:
+
+- **SOUL.md** — skipped if one already exists in `~/.hermes/`
+- **Memory entries** — skipped if memories already exist (to avoid duplicates)
+- **Skills** — skipped if a skill with the same name already exists
+- **API keys** — skipped if the key is already set in `~/.hermes/.env`
+
+To overwrite conflicts, use `--overwrite`. The migration creates backups before overwriting.
+
+For skills, you can also use `--skill-conflict rename` to import conflicting skills under a new name (e.g., `skill-name-imported`).
+
+## Migration Report
+
+Every migration (including dry runs) produces a report showing:
+- **Migrated items** — what was successfully imported
+- **Conflicts** — items skipped because they already exist
+- **Skipped items** — items not found in the source
+- **Errors** — items that failed to import
+
+For execute runs, the full report is saved to `~/.hermes/migration/openclaw/<timestamp>/`.
+
+## Troubleshooting
+
+### "OpenClaw directory not found"
+The migration looks for `~/.openclaw` by default. If your OpenClaw is installed elsewhere, use `--source`:
+```bash
+hermes claw migrate --source /path/to/.openclaw
+```
+
+### "Migration script not found"
+The migration script ships with Hermes Agent. If you installed via pip (not git clone), the `optional-skills/` directory may not be present. Install the skill from the Skills Hub:
+```bash
+hermes skills install openclaw-migration
+```
+
+### Memory overflow
+If your OpenClaw MEMORY.md or USER.md exceeds Hermes' character limits, excess entries are exported to an overflow file in the migration report directory. You can manually review and add the most important ones.

docs/plans/2026-03-16-pricing-accuracy-architecture-design.md

@@ -0,0 +1,608 @@
+# Pricing Accuracy Architecture
+
+Date: 2026-03-16
+
+## Goal
+
+Hermes should only show dollar costs when they are backed by an official source for the user's actual billing path.
+
+This design replaces the current static, heuristic pricing flow in:
+
+- `run_agent.py`
+- `agent/usage_pricing.py`
+- `agent/insights.py`
+- `cli.py`
+
+with a provider-aware pricing system that:
+
+- handles cache billing correctly
+- distinguishes `actual` vs `estimated` vs `included` vs `unknown`
+- reconciles post-hoc costs when providers expose authoritative billing data
+- supports direct providers, OpenRouter, subscriptions, enterprise pricing, and custom endpoints
+
+## Problems In The Current Design
+
+Current Hermes behavior has four structural issues:
+
+1. It stores only `prompt_tokens` and `completion_tokens`, which is insufficient for providers that bill cache reads and cache writes separately.
+2. It uses a static model price table and fuzzy heuristics, which can drift from current official pricing.
+3. It assumes public API list pricing matches the user's real billing path.
+4. It has no distinction between live estimates and reconciled billed cost.
+
+## Design Principles
+
+1. Normalize usage before pricing.
+2. Never fold cached tokens into plain input cost.
+3. Track certainty explicitly.
+4. Treat the billing path as part of the model identity.
+5. Prefer official machine-readable sources over scraped docs.
+6. Use post-hoc provider cost APIs when available.
+7. Show `n/a` rather than inventing precision.
+
+## High-Level Architecture
+
+The new system has four layers:
+
+1. `usage_normalization`
+   Converts raw provider usage into a canonical usage record.
+2. `pricing_source_resolution`
+   Determines the billing path, source of truth, and applicable pricing source.
+3. `cost_estimation_and_reconciliation`
+   Produces an immediate estimate when possible, then replaces or annotates it with actual billed cost later.
+4. `presentation`
+   `/usage`, `/insights`, and the status bar display cost with certainty metadata.
+
+## Canonical Usage Record
+
+Add a canonical usage model that every provider path maps into before any pricing math happens.
+
+Suggested structure:
+
+```python
+@dataclass
+class CanonicalUsage:
+    provider: str
+    billing_provider: str
+    model: str
+    billing_route: str
+
+    input_tokens: int = 0
+    output_tokens: int = 0
+    cache_read_tokens: int = 0
+    cache_write_tokens: int = 0
+    reasoning_tokens: int = 0
+    request_count: int = 1
+
+    raw_usage: dict[str, Any] | None = None
+    raw_usage_fields: dict[str, str] | None = None
+    computed_fields: set[str] | None = None
+
+    provider_request_id: str | None = None
+    provider_generation_id: str | None = None
+    provider_response_id: str | None = None
+```
+
+Rules:
+
+- `input_tokens` means non-cached input only.
+- `cache_read_tokens` and `cache_write_tokens` are never merged into `input_tokens`.
+- `output_tokens` excludes cache metrics.
+- `reasoning_tokens` is telemetry unless a provider officially bills it separately.
+
+This is the same normalization pattern used by `opencode`, extended with provenance and reconciliation ids.
+
+## Provider Normalization Rules
+
+### OpenAI Direct
+
+Source usage fields:
+
+- `prompt_tokens`
+- `completion_tokens`
+- `prompt_tokens_details.cached_tokens`
+
+Normalization:
+
+- `cache_read_tokens = cached_tokens`
+- `input_tokens = prompt_tokens - cached_tokens`
+- `cache_write_tokens = 0` unless OpenAI exposes it in the relevant route
+- `output_tokens = completion_tokens`
+
+### Anthropic Direct
+
+Source usage fields:
+
+- `input_tokens`
+- `output_tokens`
+- `cache_read_input_tokens`
+- `cache_creation_input_tokens`
+
+Normalization:
+
+- `input_tokens = input_tokens`
+- `output_tokens = output_tokens`
+- `cache_read_tokens = cache_read_input_tokens`
+- `cache_write_tokens = cache_creation_input_tokens`
+
+### OpenRouter
+
+Estimate-time usage normalization should use the response usage payload with the same rules as the underlying provider when possible.
+
+Reconciliation-time records should also store:
+
+- OpenRouter generation id
+- native token fields when available
+- `total_cost`
+- `cache_discount`
+- `upstream_inference_cost`
+- `is_byok`
+
+### Gemini / Vertex
+
+Use official Gemini or Vertex usage fields where available.
+
+If cached content tokens are exposed:
+
+- map them to `cache_read_tokens`
+
+If a route exposes no cache creation metric:
+
+- store `cache_write_tokens = 0`
+- preserve the raw usage payload for later extension
+
+### DeepSeek And Other Direct Providers
+
+Normalize only the fields that are officially exposed.
+
+If a provider does not expose cache buckets:
+
+- do not infer them unless the provider explicitly documents how to derive them
+
+### Subscription / Included-Cost Routes
+
+These still use the canonical usage model.
+
+Tokens are tracked normally. Cost depends on billing mode, not on whether usage exists.
+
+## Billing Route Model
+
+Hermes must stop keying pricing solely by `model`.
+
+Introduce a billing route descriptor:
+
+```python
+@dataclass
+class BillingRoute:
+    provider: str
+    base_url: str | None
+    model: str
+    billing_mode: str
+    organization_hint: str | None = None
+```
+
+`billing_mode` values:
+
+- `official_cost_api`
+- `official_generation_api`
+- `official_models_api`
+- `official_docs_snapshot`
+- `subscription_included`
+- `user_override`
+- `custom_contract`
+- `unknown`
+
+Examples:
+
+- OpenAI direct API with Costs API access: `official_cost_api`
+- Anthropic direct API with Usage & Cost API access: `official_cost_api`
+- OpenRouter request before reconciliation: `official_models_api`
+- OpenRouter request after generation lookup: `official_generation_api`
+- GitHub Copilot style subscription route: `subscription_included`
+- local OpenAI-compatible server: `unknown`
+- enterprise contract with configured rates: `custom_contract`
+
+## Cost Status Model
+
+Every displayed cost should have:
+
+```python
+@dataclass
+class CostResult:
+    amount_usd: Decimal | None
+    status: Literal["actual", "estimated", "included", "unknown"]
+    source: Literal[
+        "provider_cost_api",
+        "provider_generation_api",
+        "provider_models_api",
+        "official_docs_snapshot",
+        "user_override",
+        "custom_contract",
+        "none",
+    ]
+    label: str
+    fetched_at: datetime | None
+    pricing_version: str | None
+    notes: list[str]
+```
+
+Presentation rules:
+
+- `actual`: show dollar amount as final
+- `estimated`: show dollar amount with estimate labeling
+- `included`: show `included` or `$0.00 (included)` depending on UX choice
+- `unknown`: show `n/a`
+
+## Official Source Hierarchy
+
+Resolve cost using this order:
+
+1. Request-level or account-level official billed cost
+2. Official machine-readable model pricing
+3. Official docs snapshot
+4. User override or custom contract
+5. Unknown
+
+The system must never skip to a lower level if a higher-confidence source exists for the current billing route.
+
+## Provider-Specific Truth Rules
+
+### OpenAI Direct
+
+Preferred truth:
+
+1. Costs API for reconciled spend
+2. Official pricing page for live estimate
+
+### Anthropic Direct
+
+Preferred truth:
+
+1. Usage & Cost API for reconciled spend
+2. Official pricing docs for live estimate
+
+### OpenRouter
+
+Preferred truth:
+
+1. `GET /api/v1/generation` for reconciled `total_cost`
+2. `GET /api/v1/models` pricing for live estimate
+
+Do not use underlying provider public pricing as the source of truth for OpenRouter billing.
+
+### Gemini / Vertex
+
+Preferred truth:
+
+1. official billing export or billing API for reconciled spend when available for the route
+2. official pricing docs for estimate
+
+### DeepSeek
+
+Preferred truth:
+
+1. official machine-readable cost source if available in the future
+2. official pricing docs snapshot today
+
+### Subscription-Included Routes
+
+Preferred truth:
+
+1. explicit route config marking the model as included in subscription
+
+These should display `included`, not an API list-price estimate.
+
+### Custom Endpoint / Local Model
+
+Preferred truth:
+
+1. user override
+2. custom contract config
+3. unknown
+
+These should default to `unknown`.
+
+## Pricing Catalog
+
+Replace the current `MODEL_PRICING` dict with a richer pricing catalog.
+
+Suggested record:
+
+```python
+@dataclass
+class PricingEntry:
+    provider: str
+    route_pattern: str
+    model_pattern: str
+
+    input_cost_per_million: Decimal | None = None
+    output_cost_per_million: Decimal | None = None
+    cache_read_cost_per_million: Decimal | None = None
+    cache_write_cost_per_million: Decimal | None = None
+    request_cost: Decimal | None = None
+    image_cost: Decimal | None = None
+
+    source: str = "official_docs_snapshot"
+    source_url: str | None = None
+    fetched_at: datetime | None = None
+    pricing_version: str | None = None
+```
+
+The catalog should be route-aware:
+
+- `openai:gpt-5`
+- `anthropic:claude-opus-4-6`
+- `openrouter:anthropic/claude-opus-4.6`
+- `copilot:gpt-4o`
+
+This avoids conflating direct-provider billing with aggregator billing.
+
+## Pricing Sync Architecture
+
+Introduce a pricing sync subsystem instead of manually maintaining a single hardcoded table.
+
+Suggested modules:
+
+- `agent/pricing/catalog.py`
+- `agent/pricing/sources.py`
+- `agent/pricing/sync.py`
+- `agent/pricing/reconcile.py`
+- `agent/pricing/types.py`
+
+### Sync Sources
+
+- OpenRouter models API
+- official provider docs snapshots where no API exists
+- user overrides from config
+
+### Sync Output
+
+Cache pricing entries locally with:
+
+- source URL
+- fetch timestamp
+- version/hash
+- confidence/source type
+
+### Sync Frequency
+
+- startup warm cache
+- background refresh every 6 to 24 hours depending on source
+- manual `hermes pricing sync`
+
+## Reconciliation Architecture
+
+Live requests may produce only an estimate initially. Hermes should reconcile them later when a provider exposes actual billed cost.
+
+Suggested flow:
+
+1. Agent call completes.
+2. Hermes stores canonical usage plus reconciliation ids.
+3. Hermes computes an immediate estimate if a pricing source exists.
+4. A reconciliation worker fetches actual cost when supported.
+5. Session and message records are updated with `actual` cost.
+
+This can run:
+
+- inline for cheap lookups
+- asynchronously for delayed provider accounting
+
+## Persistence Changes
+
+Session storage should stop storing only aggregate prompt/completion totals.
+
+Add fields for both usage and cost certainty:
+
+- `input_tokens`
+- `output_tokens`
+- `cache_read_tokens`
+- `cache_write_tokens`
+- `reasoning_tokens`
+- `estimated_cost_usd`
+- `actual_cost_usd`
+- `cost_status`
+- `cost_source`
+- `pricing_version`
+- `billing_provider`
+- `billing_mode`
+
+If schema expansion is too large for one PR, add a new pricing events table:
+
+```text
+session_cost_events
+  id
+  session_id
+  request_id
+  provider
+  model
+  billing_mode
+  input_tokens
+  output_tokens
+  cache_read_tokens
+  cache_write_tokens
+  estimated_cost_usd
+  actual_cost_usd
+  cost_status
+  cost_source
+  pricing_version
+  created_at
+  updated_at
+```
+
+## Hermes Touchpoints
+
+### `run_agent.py`
+
+Current responsibility:
+
+- parse raw provider usage
+- update session token counters
+
+New responsibility:
+
+- build `CanonicalUsage`
+- update canonical counters
+- store reconciliation ids
+- emit usage event to pricing subsystem
+
+### `agent/usage_pricing.py`
+
+Current responsibility:
+
+- static lookup table
+- direct cost arithmetic
+
+New responsibility:
+
+- move or replace with pricing catalog facade
+- no fuzzy model-family heuristics
+- no direct pricing without billing-route context
+
+### `cli.py`
+
+Current responsibility:
+
+- compute session cost directly from prompt/completion totals
+
+New responsibility:
+
+- display `CostResult`
+- show status badges:
+  - `actual`
+  - `estimated`
+  - `included`
+  - `n/a`
+
+### `agent/insights.py`
+
+Current responsibility:
+
+- recompute historical estimates from static pricing
+
+New responsibility:
+
+- aggregate stored pricing events
+- prefer actual cost over estimate
+- surface estimates only when reconciliation is unavailable
+
+## UX Rules
+
+### Status Bar
+
+Show one of:
+
+- `$1.42`
+- `~$1.42`
+- `included`
+- `cost n/a`
+
+Where:
+
+- `$1.42` means `actual`
+- `~$1.42` means `estimated`
+- `included` means subscription-backed or explicitly zero-cost route
+- `cost n/a` means unknown
+
+### `/usage`
+
+Show:
+
+- token buckets
+- estimated cost
+- actual cost if available
+- cost status
+- pricing source
+
+### `/insights`
+
+Aggregate:
+
+- actual cost totals
+- estimated-only totals
+- unknown-cost sessions count
+- included-cost sessions count
+
+## Config And Overrides
+
+Add user-configurable pricing overrides in config:
+
+```yaml
+pricing:
+  mode: hybrid
+  sync_on_startup: true
+  sync_interval_hours: 12
+  overrides:
+    - provider: openrouter
+      model: anthropic/claude-opus-4.6
+      billing_mode: custom_contract
+      input_cost_per_million: 4.25
+      output_cost_per_million: 22.0
+      cache_read_cost_per_million: 0.5
+      cache_write_cost_per_million: 6.0
+  included_routes:
+    - provider: copilot
+      model: "*"
+    - provider: codex-subscription
+      model: "*"
+```
+
+Overrides must win over catalog defaults for the matching billing route.
+
+## Rollout Plan
+
+### Phase 1
+
+- add canonical usage model
+- split cache token buckets in `run_agent.py`
+- stop pricing cache-inflated prompt totals
+- preserve current UI with improved backend math
+
+### Phase 2
+
+- add route-aware pricing catalog
+- integrate OpenRouter models API sync
+- add `estimated` vs `included` vs `unknown`
+
+### Phase 3
+
+- add reconciliation for OpenRouter generation cost
+- add actual cost persistence
+- update `/insights` to prefer actual cost
+
+### Phase 4
+
+- add direct OpenAI and Anthropic reconciliation paths
+- add user overrides and contract pricing
+- add pricing sync CLI command
+
+## Testing Strategy
+
+Add tests for:
+
+- OpenAI cached token subtraction
+- Anthropic cache read/write separation
+- OpenRouter estimated vs actual reconciliation
+- subscription-backed models showing `included`
+- custom endpoints showing `n/a`
+- override precedence
+- stale catalog fallback behavior
+
+Current tests that assume heuristic pricing should be replaced with route-aware expectations.
+
+## Non-Goals
+
+- exact enterprise billing reconstruction without an official source or user override
+- backfilling perfect historical cost for old sessions that lack cache bucket data
+- scraping arbitrary provider web pages at request time
+
+## Recommendation
+
+Do not expand the existing `MODEL_PRICING` dict.
+
+That path cannot satisfy the product requirement. Hermes should instead migrate to:
+
+- canonical usage normalization
+- route-aware pricing sources
+- estimate-then-reconcile cost lifecycle
+- explicit certainty states in the UI
+
+This is the minimum architecture that makes the statement "Hermes pricing is backed by official sources where possible, and otherwise clearly labeled" defensible.

docs/skins/example-skin.yaml

@@ -0,0 +1,89 @@
+# ============================================================================
+# Hermes Agent — Example Skin Template
+# ============================================================================
+#
+# Copy this file to ~/.hermes/skins/<name>.yaml to create a custom skin.
+# All fields are optional — missing values inherit from the default skin.
+# Activate with: /skin <name>  or  display.skin: <name> in config.yaml
+#
+# See hermes_cli/skin_engine.py for the full schema reference.
+# ============================================================================
+
+# Required: unique skin name (used in /skin command and config)
+name: example
+description: An example custom skin — copy and modify this template
+
+# ── Colors ──────────────────────────────────────────────────────────────────
+# Hex color values for Rich markup. These control the CLI's visual palette.
+colors:
+  # Banner panel (the startup welcome box)
+  banner_border: "#CD7F32"        # Panel border
+  banner_title: "#FFD700"         # Panel title text
+  banner_accent: "#FFBF00"        # Section headers (Available Tools, Skills, etc.)
+  banner_dim: "#B8860B"           # Dim/muted text (separators, model info)
+  banner_text: "#FFF8DC"          # Body text (tool names, skill names)
+
+  # UI elements
+  ui_accent: "#FFBF00"            # General accent color
+  ui_label: "#4dd0e1"             # Labels
+  ui_ok: "#4caf50"                # Success indicators
+  ui_error: "#ef5350"             # Error indicators
+  ui_warn: "#ffa726"              # Warning indicators
+
+  # Input area
+  prompt: "#FFF8DC"               # Prompt text color
+  input_rule: "#CD7F32"           # Horizontal rule around input
+
+  # Response box
+  response_border: "#FFD700"      # Response box border (ANSI color)
+
+  # Session display
+  session_label: "#DAA520"        # Session label
+  session_border: "#8B8682"       # Session ID dim color
+
+# ── Spinner ─────────────────────────────────────────────────────────────────
+# Customize the animated spinner shown during API calls and tool execution.
+spinner:
+  # Faces shown while waiting for the API response
+  waiting_faces:
+    - "(｡◕‿◕｡)"
+    - "(◕‿◕✿)"
+    - "٩(◕‿◕｡)۶"
+
+  # Faces shown during extended thinking/reasoning
+  thinking_faces:
+    - "(｡•́︿•̀｡)"
+    - "(◔_◔)"
+    - "(¬‿¬)"
+
+  # Verbs used in spinner messages (e.g., "pondering your request...")
+  thinking_verbs:
+    - "pondering"
+    - "contemplating"
+    - "musing"
+    - "ruminating"
+
+  # Optional: left/right decorations around the spinner
+  # Each entry is a [left, right] pair. Omit entirely for no wings.
+  # wings:
+  #   - ["⟪⚔", "⚔⟫"]
+  #   - ["⟪▲", "▲⟫"]
+
+# ── Branding ────────────────────────────────────────────────────────────────
+# Text strings used throughout the CLI interface.
+branding:
+  agent_name: "Hermes Agent"          # Banner title, about display
+  welcome: "Welcome! Type your message or /help for commands."
+  goodbye: "Goodbye! ⚕"              # Exit message
+  response_label: " ⚕ Hermes "       # Response box header label
+  prompt_symbol: "❯ "                 # Input prompt symbol
+  help_header: "(^_^)? Available Commands"  # /help header text
+
+# ── Tool Output ─────────────────────────────────────────────────────────────
+# Character used as the prefix for tool output lines.
+# Default is "┊" (thin dotted vertical line). Some alternatives:
+#   "╎" (light triple dash vertical)
+#   "▏" (left one-eighth block)
+#   "│" (box drawing light vertical)
+#   "┃" (box drawing heavy vertical)
+tool_prefix: "┊"

environments/README.md

@@ -0,0 +1,324 @@
+# Hermes-Agent Atropos Environments
+
+This directory contains the integration layer between **hermes-agent's** tool-calling capabilities and the **Atropos** RL training framework. It provides everything needed to run agentic LLMs through multi-turn tool-calling loops, score their output with arbitrary reward functions, and feed results into Atropos for training or evaluation.
+
+## Architecture Overview
+
+```
+                        Atropos Framework
+                    ┌───────────────────────┐
+                    │       BaseEnv          │  (atroposlib)
+                    │  - Server management   │
+                    │  - Worker scheduling   │
+                    │  - Wandb logging       │
+                    │  - CLI (serve/process/ │
+                    │    evaluate)           │
+                    └───────────┬───────────┘
+                                │ inherits
+                    ┌───────────┴───────────┐
+                    │  HermesAgentBaseEnv    │  hermes_base_env.py
+                    │  - Terminal backend    │
+                    │  - Tool resolution     │
+                    │  - Agent loop          │
+                    │  - ToolContext          │
+                    │  - Async patches       │
+                    └───────────┬───────────┘
+                                │ inherits
+              ┌─────────────────┼─────────────────┐
+              │                 │                  │
+     TerminalTestEnv     HermesSweEnv    TerminalBench2EvalEnv
+     (stack testing)     (SWE training)   (TB2 benchmark eval)
+```
+
+### Inheritance Chain
+
+**BaseEnv** (from `atroposlib`) is the Atropos base class. It provides:
+- Server management (OpenAI-compatible API servers, VLLM, SGLang)
+- Worker scheduling for parallel rollouts
+- Wandb integration for metrics and rollout logging
+- CLI interface with three subcommands: `serve`, `process`, `evaluate`
+- `evaluate_log()` for saving eval results to JSON + samples.jsonl
+
+**HermesAgentBaseEnv** (`hermes_base_env.py`) extends BaseEnv with hermes-agent specifics:
+- Sets `os.environ["TERMINAL_ENV"]` to configure the terminal backend (local, docker, modal, daytona, ssh, singularity)
+- Resolves hermes-agent toolsets via `_resolve_tools_for_group()` (calls `get_tool_definitions()` which queries `tools/registry.py`)
+- Implements `collect_trajectory()` which runs the full agent loop and computes rewards
+- Supports two-phase operation (Phase 1: OpenAI server, Phase 2: VLLM ManagedServer)
+- Applies monkey patches for async-safe tool operation at import time
+
+Concrete environments inherit from `HermesAgentBaseEnv` and implement:
+- `setup()` -- Load dataset, initialize state
+- `get_next_item()` -- Return the next item for rollout
+- `format_prompt()` -- Convert a dataset item into the user message
+- `compute_reward()` -- Score the rollout using ToolContext
+- `evaluate()` -- Periodic evaluation logic
+
+## Core Components
+
+### Agent Loop (`agent_loop.py`)
+
+`HermesAgentLoop` is the reusable multi-turn agent engine. It runs the same pattern as hermes-agent's `run_agent.py`:
+
+1. Send messages + tools to the API via `server.chat_completion()`
+2. If the response contains `tool_calls`, execute each one via `handle_function_call()` (which delegates to `tools/registry.py`'s `dispatch()`)
+3. Append tool results to the conversation and go back to step 1
+4. If the response has no tool_calls, the agent is done
+
+Tool calls are executed in a thread pool (`run_in_executor`) so backends that use `asyncio.run()` internally (Modal, Docker) don't deadlock inside Atropos's event loop.
+
+Returns an `AgentResult` containing the full conversation history, turn count, reasoning content per turn, tool errors, and optional ManagedServer state (for Phase 2).
+
+### Tool Context (`tool_context.py`)
+
+`ToolContext` is a per-rollout handle that gives reward/verification functions direct access to **all** hermes-agent tools, scoped to the rollout's `task_id`. The same `task_id` means the terminal/browser session is the SAME one the model used during its rollout -- all state (files, processes, browser tabs) is preserved.
+
+```python
+async def compute_reward(self, item, result, ctx: ToolContext):
+    # Run tests in the model's terminal sandbox
+    test = ctx.terminal("pytest -v")
+    if test["exit_code"] == 0:
+        return 1.0
+
+    # Check if a file was created
+    content = ctx.read_file("/workspace/solution.py")
+    if content.get("content"):
+        return 0.5
+
+    # Download files locally for verification (binary-safe)
+    ctx.download_file("/remote/output.bin", "/local/output.bin")
+
+    return 0.0
+```
+
+Available methods:
+- **Terminal**: `terminal(command, timeout)` -- run shell commands
+- **Files**: `read_file(path)`, `write_file(path, content)`, `search(query, path)`
+- **Transfers**: `upload_file()`, `upload_dir()`, `download_file()`, `download_dir()` -- binary-safe file transfers between host and sandbox
+- **Web**: `web_search(query)`, `web_extract(urls)`
+- **Browser**: `browser_navigate(url)`, `browser_snapshot()`
+- **Generic**: `call_tool(name, args)` -- call any hermes-agent tool by name
+- **Cleanup**: `cleanup()` -- release all resources (called automatically after `compute_reward`)
+
+### Patches (`patches.py`)
+
+**Problem**: Some hermes-agent tools use `asyncio.run()` internally (e.g., the Modal backend). This crashes when called from inside Atropos's event loop because `asyncio.run()` cannot be nested.
+
+**Solution**: `ModalEnvironment` uses a dedicated `_AsyncWorker` background thread with its own event loop. The calling code sees a sync interface, but internally all async Modal SDK calls happen on the worker thread so they don't conflict with Atropos's loop. This is built directly into `tools/environments/modal.py` — no monkey-patching required.
+
+`patches.py` is now a no-op (kept for backward compatibility with imports).
+
+### Tool Call Parsers (`tool_call_parsers/`)
+
+Client-side parsers that extract structured `tool_calls` from raw model output text. Used in **Phase 2** (VLLM server type) where ManagedServer's `/generate` endpoint returns raw text without tool call parsing.
+
+Each parser is a standalone reimplementation of the corresponding VLLM parser's `extract_tool_calls()` logic. No VLLM dependency -- only standard library (`re`, `json`, `uuid`) and `openai` types.
+
+Available parsers:
+- `hermes` -- Hermes/ChatML `<tool_call>` XML format
+- `mistral` -- Mistral `[TOOL_CALLS]` format
+- `llama3_json` -- Llama 3 JSON tool calling
+- `qwen` -- Qwen tool calling format
+- `qwen3_coder` -- Qwen3 Coder format
+- `deepseek_v3` -- DeepSeek V3 format
+- `deepseek_v3_1` -- DeepSeek V3.1 format
+- `kimi_k2` -- Kimi K2 format
+- `longcat` -- Longcat format
+- `glm45` / `glm47` -- GLM model formats
+
+Usage:
+```python
+from environments.tool_call_parsers import get_parser
+
+parser = get_parser("hermes")
+content, tool_calls = parser.parse(raw_model_output)
+```
+
+In Phase 1 (OpenAI server type), these parsers are not needed -- the server handles tool call parsing natively.
+
+## Two-Phase Operation
+
+### Phase 1: OpenAI Server (Evaluation / SFT Data Generation)
+
+Uses `server.chat_completion()` with `tools=` parameter. The server (VLLM, SGLang, OpenRouter, OpenAI) handles tool call parsing natively. Returns `ChatCompletion` objects with structured `tool_calls`.
+
+- Good for: evaluation, SFT data generation, testing
+- Run with: `serve` (with `run-api`), `process`, or `evaluate` subcommands
+- Placeholder tokens are created for the Atropos pipeline
+
+### Phase 2: VLLM ManagedServer (Full RL Training)
+
+Uses ManagedServer for exact token IDs + logprobs via `/generate`. Client-side tool call parser (from `tool_call_parsers/`) reconstructs structured `tool_calls` from raw output.
+
+- Good for: full RL training with GRPO/PPO
+- Run with: `serve` subcommand
+- Real tokens, masks, and logprobs flow through the pipeline
+
+## Directory Structure
+
+```
+environments/
+├── README.md                     # This file
+├── __init__.py                   # Package exports
+├── hermes_base_env.py            # Abstract base (HermesAgentBaseEnv)
+├── agent_loop.py                 # Multi-turn agent engine (HermesAgentLoop)
+├── tool_context.py               # Per-rollout tool access for reward functions
+├── patches.py                    # Async-safety patches for Modal backend
+│
+├── tool_call_parsers/            # Phase 2 client-side parsers
+│   ├── __init__.py               # Registry + base class
+│   ├── hermes_parser.py
+│   ├── mistral_parser.py
+│   ├── llama_parser.py
+│   ├── qwen_parser.py
+│   ├── qwen3_coder_parser.py
+│   ├── deepseek_v3_parser.py
+│   ├── deepseek_v3_1_parser.py
+│   ├── kimi_k2_parser.py
+│   ├── longcat_parser.py
+│   ├── glm45_parser.py
+│   └── glm47_parser.py
+│
+├── terminal_test_env/            # Stack validation environment
+│   └── terminal_test_env.py
+│
+├── hermes_swe_env/               # SWE-bench style training environment
+│   └── hermes_swe_env.py
+│
+└── benchmarks/                   # Evaluation benchmarks
+    ├── terminalbench_2/          # 89 terminal tasks, Modal sandboxes
+    │   └── terminalbench2_env.py
+    ├── tblite/                   # 100 calibrated tasks (fast TB2 proxy)
+    │   └── tblite_env.py
+    └── yc_bench/                 # Long-horizon strategic benchmark
+        └── yc_bench_env.py
+```
+
+## Concrete Environments
+
+### TerminalTestEnv (`terminal_test_env/`)
+
+A self-contained environment with inline tasks (no external dataset needed) for validating the full stack end-to-end. Each task asks the model to create a file at a known path, and the verifier checks the content matches.
+
+```bash
+# Serve mode (needs run-api)
+run-api
+python environments/terminal_test_env/terminal_test_env.py serve
+
+# Process mode (no run-api, saves to JSONL)
+python environments/terminal_test_env/terminal_test_env.py process \
+    --env.data_path_to_save_groups terminal_test_output.jsonl
+```
+
+### HermesSweEnv (`hermes_swe_env/`)
+
+SWE-bench style training environment. The model gets a coding task, uses terminal + file + web tools to solve it, and the reward function runs tests in the same Modal sandbox.
+
+```bash
+python environments/hermes_swe_env/hermes_swe_env.py serve \
+    --openai.model_name YourModel \
+    --env.dataset_name bigcode/humanevalpack \
+    --env.terminal_backend modal
+```
+
+### TerminalBench2EvalEnv (`benchmarks/terminalbench_2/`)
+
+**Eval-only** environment for the Terminal-Bench 2.0 benchmark (89 tasks). Each task gets a pre-built Docker Hub image, a natural language instruction, and a test suite. The agent uses terminal + file tools to solve the task, then the test suite verifies correctness.
+
+Follows the standard Atropos eval pattern (like GPQA, MMLU, etc.):
+- Run via `evaluate` subcommand (no `run-api` needed)
+- `setup()` loads the dataset, `evaluate()` runs all tasks
+- `rollout_and_score_eval()` handles per-task agent loop + test verification
+- Downloads verifier output locally for reliable reward checking (Harbor pattern)
+
+```bash
+# Run full benchmark
+python environments/benchmarks/terminalbench_2/terminalbench2_env.py evaluate \
+    --openai.model_name anthropic/claude-opus-4.6
+
+# Run subset of tasks
+python environments/benchmarks/terminalbench_2/terminalbench2_env.py evaluate \
+    --openai.model_name anthropic/claude-opus-4.6 \
+    --env.task_filter fix-git,git-multibranch
+
+# Skip specific tasks
+python environments/benchmarks/terminalbench_2/terminalbench2_env.py evaluate \
+    --openai.model_name anthropic/claude-opus-4.6 \
+    --env.skip_tasks heavy-task,slow-task
+```
+
+## Creating a New Environment
+
+### Training Environment
+
+1. Create a new directory under `environments/`
+2. Create your env file inheriting from `HermesAgentBaseEnv`
+3. Implement the four abstract methods + `evaluate()`
+
+```python
+from environments.hermes_base_env import HermesAgentBaseEnv, HermesAgentEnvConfig
+
+class MyEnvConfig(HermesAgentEnvConfig):
+    pass  # Add custom fields as needed
+
+class MyEnv(HermesAgentBaseEnv):
+    name = "my-env"
+    env_config_cls = MyEnvConfig
+
+    @classmethod
+    def config_init(cls):
+        env_config = MyEnvConfig(
+            enabled_toolsets=["terminal", "file"],
+            terminal_backend="modal",
+            # ... other config
+        )
+        server_configs = [APIServerConfig(...)]
+        return env_config, server_configs
+
+    async def setup(self):
+        self.dataset = load_dataset(...)
+        self.iter = 0
+
+    async def get_next_item(self):
+        item = self.dataset[self.iter % len(self.dataset)]
+        self.iter += 1
+        return item
+
+    def format_prompt(self, item):
+        return item["instruction"]
+
+    async def compute_reward(self, item, result, ctx):
+        # ctx gives you full tool access to the rollout's sandbox
+        test = ctx.terminal("pytest -v")
+        return 1.0 if test["exit_code"] == 0 else 0.0
+
+    async def evaluate(self, *args, **kwargs):
+        # Periodic evaluation logic
+        ...
+
+if __name__ == "__main__":
+    MyEnv.cli()
+```
+
+### Eval-Only Environment (Benchmark)
+
+For eval benchmarks, follow the pattern in `terminalbench2_env.py`:
+1. Create under `environments/benchmarks/your-benchmark/`
+2. Inherit from `HermesAgentBaseEnv`
+3. Set eval-only config: `eval_handling=STOP_TRAIN`, `steps_per_eval=1`, `total_steps=1`
+4. Stub the training methods (`collect_trajectories`, `score`)
+5. Implement `rollout_and_score_eval()` and `evaluate()`
+6. Run with `evaluate` subcommand
+
+## Key Config Fields
+
+| Field | Description | Default |
+|-------|-------------|---------|
+| `enabled_toolsets` | Which hermes toolsets to enable | `None` (all) |
+| `disabled_toolsets` | Toolsets to disable | `None` |
+| `distribution` | Probabilistic toolset distribution name | `None` |
+| `max_agent_turns` | Max LLM calls per rollout | `30` |
+| `agent_temperature` | Sampling temperature | `1.0` |
+| `terminal_backend` | `local`, `docker`, `modal`, `daytona`, `ssh`, `singularity` | `local` |
+| `system_prompt` | System message for the agent | `None` |
+| `tool_call_parser` | Parser name for Phase 2 | `hermes` |
+| `eval_handling` | `STOP_TRAIN`, `LIMIT_TRAIN`, `NONE` | `STOP_TRAIN` |

environments/__init__.py

@@ -0,0 +1,36 @@
+"""
+Hermes-Agent Atropos Environments
+
+Provides a layered integration between hermes-agent's tool-calling capabilities
+and the Atropos RL training framework.
+
+Core layers:
+    - agent_loop: Reusable multi-turn agent loop with standard OpenAI-spec tool calling
+    - tool_context: Per-rollout tool access handle for reward/verification functions
+    - hermes_base_env: Abstract base environment (BaseEnv subclass) for Atropos
+    - tool_call_parsers: Client-side tool call parser registry for Phase 2 (VLLM /generate)
+
+Concrete environments:
+    - terminal_test_env/: Simple file-creation tasks for testing the stack
+    - hermes_swe_env/: SWE-bench style tasks with Modal sandboxes
+
+Benchmarks (eval-only):
+    - benchmarks/terminalbench_2/: Terminal-Bench 2.0 evaluation
+"""
+
+try:
+    from environments.agent_loop import AgentResult, HermesAgentLoop
+    from environments.tool_context import ToolContext
+    from environments.hermes_base_env import HermesAgentBaseEnv, HermesAgentEnvConfig
+except ImportError:
+    # atroposlib not installed — environments are unavailable but
+    # submodules like tool_call_parsers can still be imported directly.
+    pass
+
+__all__ = [
+    "AgentResult",
+    "HermesAgentLoop",
+    "ToolContext",
+    "HermesAgentBaseEnv",
+    "HermesAgentEnvConfig",
+]

environments/agent_loop.py

@@ -0,0 +1,511 @@
+"""
+HermesAgentLoop -- Reusable Multi-Turn Agent Engine
+
+Runs the hermes-agent tool-calling loop using standard OpenAI-spec tool calling.
+Works with any server that returns ChatCompletion objects with tool_calls:
+    - Phase 1: OpenAI server type (VLLM, SGLang, OpenRouter, OpenAI API)
+    - Phase 2: ManagedServer with client-side tool call parser
+
+The loop passes tools= and checks response.choices[0].message.tool_calls,
+identical to hermes-agent's run_agent.py. Tool execution is dispatched via
+handle_function_call() from model_tools.py.
+"""
+
+import asyncio
+import concurrent.futures
+import json
+import logging
+import os
+import uuid
+from dataclasses import dataclass, field
+from typing import Any, Dict, List, Optional, Set
+
+from model_tools import handle_function_call
+
+# Thread pool for running sync tool calls that internally use asyncio.run()
+# (e.g., the Modal/Docker/Daytona terminal backends). Running them in a separate
+# thread gives them a clean event loop so they don't deadlock inside Atropos's loop.
+# Size must be large enough for concurrent eval tasks (e.g., 89 TB2 tasks all
+# making tool calls). Too small = thread pool starvation, tasks queue for minutes.
+# Resized at runtime by HermesAgentBaseEnv.__init__ via resize_tool_pool().
+_tool_executor = concurrent.futures.ThreadPoolExecutor(max_workers=128)
+
+
+def resize_tool_pool(max_workers: int):
+    """
+    Replace the global tool executor with a new one of the given size.
+
+    Called by HermesAgentBaseEnv.__init__ based on config.tool_pool_size.
+    Safe to call before any tasks are submitted.
+    """
+    global _tool_executor
+    old_executor = _tool_executor
+    _tool_executor = concurrent.futures.ThreadPoolExecutor(max_workers=max_workers)
+    old_executor.shutdown(wait=False)
+    logger.info("Tool thread pool resized to %d workers", max_workers)
+
+logger = logging.getLogger(__name__)
+
+
+@dataclass
+class ToolError:
+    """Record of a tool execution error during the agent loop."""
+
+    turn: int                  # Which turn the error occurred on
+    tool_name: str             # Which tool was called
+    arguments: str             # The arguments passed (truncated)
+    error: str                 # The error message
+    tool_result: str           # The raw result returned to the model
+
+
+@dataclass
+class AgentResult:
+    """Result of running the agent loop."""
+
+    # Full conversation history in OpenAI message format
+    messages: List[Dict[str, Any]]
+    # ManagedServer.get_state() if available (Phase 2), None otherwise
+    managed_state: Optional[Dict[str, Any]] = None
+    # How many LLM calls were made
+    turns_used: int = 0
+    # True if model stopped calling tools naturally (vs hitting max_turns)
+    finished_naturally: bool = False
+    # Extracted reasoning content per turn (from PR #297 helpers)
+    reasoning_per_turn: List[Optional[str]] = field(default_factory=list)
+    # Tool errors encountered during the loop
+    tool_errors: List[ToolError] = field(default_factory=list)
+
+
+def _extract_reasoning_from_message(message) -> Optional[str]:
+    """
+    Extract reasoning content from a ChatCompletion message.
+
+    Handles multiple provider formats:
+    1. message.reasoning_content field (some providers)
+    2. message.reasoning field (some providers)
+    3. message.reasoning_details[].text (OpenRouter style)
+
+    Note: <think> block extraction from content is NOT done here -- that's
+    handled by the response already in Phase 1 (server does it) or by
+    ManagedServer's patch in Phase 2.
+
+    Args:
+        message: The assistant message from ChatCompletion response
+
+    Returns:
+        Extracted reasoning text, or None if not found
+    """
+    # Check reasoning_content field (common across providers)
+    if hasattr(message, "reasoning_content") and message.reasoning_content:
+        return message.reasoning_content
+
+    # Check reasoning field
+    if hasattr(message, "reasoning") and message.reasoning:
+        return message.reasoning
+
+    # Check reasoning_details (OpenRouter style)
+    if hasattr(message, "reasoning_details") and message.reasoning_details:
+        for detail in message.reasoning_details:
+            if hasattr(detail, "text") and detail.text:
+                return detail.text
+            if isinstance(detail, dict) and detail.get("text"):
+                return detail["text"]
+
+    return None
+
+
+class HermesAgentLoop:
+    """
+    Runs hermes-agent's tool-calling loop using standard OpenAI-spec tool calling.
+
+    Same pattern as run_agent.py:
+    - Pass tools= to the API
+    - Check response.choices[0].message.tool_calls
+    - Dispatch via handle_function_call()
+
+    Works identically with any server type -- OpenAI, VLLM, SGLang, OpenRouter,
+    or ManagedServer with a parser. The server determines how tool_calls get
+    populated on the response.
+    """
+
+    def __init__(
+        self,
+        server,
+        tool_schemas: List[Dict[str, Any]],
+        valid_tool_names: Set[str],
+        max_turns: int = 30,
+        task_id: Optional[str] = None,
+        temperature: float = 1.0,
+        max_tokens: Optional[int] = None,
+        extra_body: Optional[Dict[str, Any]] = None,
+    ):
+        """
+        Initialize the agent loop.
+
+        Args:
+            server: Server object with chat_completion() method (OpenAIServer,
+                    ManagedServer, ServerManager, etc.)
+            tool_schemas: OpenAI-format tool definitions from get_tool_definitions()
+            valid_tool_names: Set of tool names the model is allowed to call
+            max_turns: Maximum number of LLM calls before stopping
+            task_id: Unique ID for terminal/browser session isolation
+            temperature: Sampling temperature for generation
+            max_tokens: Max tokens per generation (None for server default)
+            extra_body: Extra parameters passed to the OpenAI client's create() call.
+                        Used for OpenRouter provider preferences, transforms, etc.
+                        e.g. {"provider": {"ignore": ["DeepInfra"]}}
+        """
+        self.server = server
+        self.tool_schemas = tool_schemas
+        self.valid_tool_names = valid_tool_names
+        self.max_turns = max_turns
+        self.task_id = task_id or str(uuid.uuid4())
+        self.temperature = temperature
+        self.max_tokens = max_tokens
+        self.extra_body = extra_body
+
+    async def run(self, messages: List[Dict[str, Any]]) -> AgentResult:
+        """
+        Execute the full agent loop using standard OpenAI tool calling.
+
+        Args:
+            messages: Initial conversation messages (system + user).
+                      Modified in-place as the conversation progresses.
+
+        Returns:
+            AgentResult with full conversation history, managed state, and metadata
+        """
+        reasoning_per_turn = []
+        tool_errors: List[ToolError] = []
+
+        # Per-loop TodoStore for the todo tool (ephemeral, dies with the loop)
+        from tools.todo_tool import TodoStore, todo_tool as _todo_tool
+        _todo_store = TodoStore()
+
+        # Extract user task from first user message for browser_snapshot context
+        _user_task = None
+        for msg in messages:
+            if msg.get("role") == "user":
+                content = msg.get("content", "")
+                if isinstance(content, str) and content.strip():
+                    _user_task = content.strip()[:500]  # Cap to avoid huge strings
+                break
+
+        import time as _time
+
+        for turn in range(self.max_turns):
+            turn_start = _time.monotonic()
+
+            # Build the chat_completion kwargs
+            chat_kwargs = {
+                "messages": messages,
+                "n": 1,
+                "temperature": self.temperature,
+            }
+
+            # Only pass tools if we have them
+            if self.tool_schemas:
+                chat_kwargs["tools"] = self.tool_schemas
+
+            # Only pass max_tokens if explicitly set
+            if self.max_tokens is not None:
+                chat_kwargs["max_tokens"] = self.max_tokens
+
+            # Inject extra_body for provider-specific params (e.g., OpenRouter
+            # provider preferences like banned/preferred providers, transforms)
+            if self.extra_body:
+                chat_kwargs["extra_body"] = self.extra_body
+
+            # Make the API call -- standard OpenAI spec
+            api_start = _time.monotonic()
+            try:
+                response = await self.server.chat_completion(**chat_kwargs)
+            except Exception as e:
+                api_elapsed = _time.monotonic() - api_start
+                logger.error("API call failed on turn %d (%.1fs): %s", turn + 1, api_elapsed, e)
+                return AgentResult(
+                    messages=messages,
+                    managed_state=self._get_managed_state(),
+                    turns_used=turn + 1,
+                    finished_naturally=False,
+                    reasoning_per_turn=reasoning_per_turn,
+                    tool_errors=tool_errors,
+                )
+
+            api_elapsed = _time.monotonic() - api_start
+
+            if not response or not response.choices:
+                logger.warning("Empty response on turn %d (api=%.1fs)", turn + 1, api_elapsed)
+                return AgentResult(
+                    messages=messages,
+                    managed_state=self._get_managed_state(),
+                    turns_used=turn + 1,
+                    finished_naturally=False,
+                    reasoning_per_turn=reasoning_per_turn,
+                    tool_errors=tool_errors,
+                )
+
+            assistant_msg = response.choices[0].message
+
+            # Extract reasoning content from the response (all provider formats)
+            reasoning = _extract_reasoning_from_message(assistant_msg)
+            reasoning_per_turn.append(reasoning)
+
+            # Check for tool calls -- standard OpenAI spec.
+            # Fallback: if response has no structured tool_calls but content
+            # contains raw tool call tags (e.g. <tool_call>), parse them using
+            # hermes-agent's standalone parsers. This handles the case where
+            # ManagedServer's ToolCallTranslator couldn't parse because vLLM
+            # isn't installed.
+            if (
+                not assistant_msg.tool_calls
+                and assistant_msg.content
+                and self.tool_schemas
+                and "<tool_call>" in (assistant_msg.content or "")
+            ):
+                try:
+                    from environments.tool_call_parsers import get_parser
+                    fallback_parser = get_parser("hermes")
+                    parsed_content, parsed_calls = fallback_parser.parse(
+                        assistant_msg.content
+                    )
+                    if parsed_calls:
+                        assistant_msg.tool_calls = parsed_calls
+                        if parsed_content is not None:
+                            assistant_msg.content = parsed_content
+                        logger.debug(
+                            "Fallback parser extracted %d tool calls from raw content",
+                            len(parsed_calls),
+                        )
+                except Exception:
+                    pass  # Fall through to no tool calls
+
+            if assistant_msg.tool_calls:
+                # Normalize tool calls to dicts — they may come as objects
+                # (OpenAI API) or dicts (vLLM ToolCallTranslator).
+                def _tc_to_dict(tc):
+                    if isinstance(tc, dict):
+                        return {
+                            "id": tc.get("id", f"call_{uuid.uuid4().hex[:8]}"),
+                            "type": "function",
+                            "function": {
+                                "name": tc.get("function", {}).get("name", tc.get("name", "")),
+                                "arguments": tc.get("function", {}).get("arguments", tc.get("arguments", "{}")),
+                            },
+                        }
+                    return {
+                        "id": tc.id,
+                        "type": "function",
+                        "function": {
+                            "name": tc.function.name,
+                            "arguments": tc.function.arguments,
+                        },
+                    }
+
+                # Build the assistant message dict for conversation history
+                msg_dict: Dict[str, Any] = {
+                    "role": "assistant",
+                    "content": assistant_msg.content or "",
+                    "tool_calls": [_tc_to_dict(tc) for tc in assistant_msg.tool_calls],
+                }
+
+                # Preserve reasoning_content for multi-turn chat template handling
+                # (e.g., Kimi-K2's template renders <think> blocks differently
+                # for history vs. the latest turn based on this field)
+                if reasoning:
+                    msg_dict["reasoning_content"] = reasoning
+
+                messages.append(msg_dict)
+
+                # Execute each tool call via hermes-agent's dispatch
+                for tc in assistant_msg.tool_calls:
+                    # Handle both object (OpenAI) and dict (vLLM) formats
+                    if isinstance(tc, dict):
+                        tool_name = tc.get("function", {}).get("name", tc.get("name", ""))
+                        tool_args_raw = tc.get("function", {}).get("arguments", tc.get("arguments", "{}"))
+                    else:
+                        tool_name = tc.function.name
+                        tool_args_raw = tc.function.arguments
+
+                    # Validate tool name
+                    if tool_name not in self.valid_tool_names:
+                        tool_result = json.dumps(
+                            {
+                                "error": f"Unknown tool '{tool_name}'. "
+                                f"Available tools: {sorted(self.valid_tool_names)}"
+                            }
+                        )
+                        tool_errors.append(ToolError(
+                            turn=turn + 1, tool_name=tool_name,
+                            arguments=tool_args_raw[:200],
+                            error=f"Unknown tool '{tool_name}'",
+                            tool_result=tool_result,
+                        ))
+                        logger.warning(
+                            "Model called unknown tool '%s' on turn %d",
+                            tool_name, turn + 1,
+                        )
+                    else:
+                        # Parse arguments
+                        try:
+                            args = json.loads(tool_args_raw)
+                        except json.JSONDecodeError as e:
+                            args = None
+                            tool_result = json.dumps(
+                                {"error": f"Invalid JSON in tool arguments: {e}. Please retry with valid JSON."}
+                            )
+                            tool_errors.append(ToolError(
+                                turn=turn + 1, tool_name=tool_name,
+                                arguments=tool_args_raw[:200],
+                                error=f"Invalid JSON: {e}",
+                                tool_result=tool_result,
+                            ))
+                            logger.warning(
+                                "Invalid JSON in tool call arguments for '%s': %s",
+                                tool_name, tool_args_raw[:200],
+                            )
+
+                        # Dispatch tool only if arguments parsed successfully
+                        if args is not None:
+                            try:
+                                if tool_name == "terminal":
+                                    backend = os.getenv("TERMINAL_ENV", "local")
+                                    cmd_preview = args.get("command", "")[:80]
+                                    logger.info(
+                                        "[%s] $ %s", self.task_id[:8], cmd_preview,
+                                    )
+
+                                tool_submit_time = _time.monotonic()
+
+                                # Todo tool -- handle locally (needs per-loop TodoStore)
+                                if tool_name == "todo":
+                                    tool_result = _todo_tool(
+                                        todos=args.get("todos"),
+                                        merge=args.get("merge", False),
+                                        store=_todo_store,
+                                    )
+                                    tool_elapsed = _time.monotonic() - tool_submit_time
+                                elif tool_name == "memory":
+                                    tool_result = json.dumps({"error": "Memory is not available in RL environments."})
+                                    tool_elapsed = _time.monotonic() - tool_submit_time
+                                elif tool_name == "session_search":
+                                    tool_result = json.dumps({"error": "Session search is not available in RL environments."})
+                                    tool_elapsed = _time.monotonic() - tool_submit_time
+                                else:
+                                    # Run tool calls in a thread pool so backends that
+                                    # use asyncio.run() internally (modal, docker, daytona) get
+                                    # a clean event loop instead of deadlocking.
+                                    loop = asyncio.get_event_loop()
+                                    # Capture current tool_name/args for the lambda
+                                    _tn, _ta, _tid = tool_name, args, self.task_id
+                                    tool_result = await loop.run_in_executor(
+                                        _tool_executor,
+                                        lambda: handle_function_call(
+                                            _tn, _ta, task_id=_tid,
+                                            user_task=_user_task,
+                                        ),
+                                    )
+                                    tool_elapsed = _time.monotonic() - tool_submit_time
+
+                                # Log slow tools and thread pool stats for debugging
+                                pool_active = _tool_executor._work_queue.qsize()
+                                if tool_elapsed > 30:
+                                    logger.warning(
+                                        "[%s] turn %d: %s took %.1fs (pool queue=%d)",
+                                        self.task_id[:8], turn + 1, tool_name,
+                                        tool_elapsed, pool_active,
+                                    )
+                            except Exception as e:
+                                tool_result = json.dumps(
+                                    {"error": f"Tool execution failed: {type(e).__name__}: {str(e)}"}
+                                )
+                                tool_errors.append(ToolError(
+                                    turn=turn + 1, tool_name=tool_name,
+                                    arguments=tool_args_raw[:200],
+                                    error=f"{type(e).__name__}: {str(e)}",
+                                    tool_result=tool_result,
+                                ))
+                                logger.error(
+                                    "Tool '%s' execution failed on turn %d: %s",
+                                    tool_name, turn + 1, e,
+                                )
+
+                        # Also check if the tool returned an error in its JSON result
+                        try:
+                            result_data = json.loads(tool_result)
+                            if isinstance(result_data, dict):
+                                err = result_data.get("error")
+                                exit_code = result_data.get("exit_code")
+                                if err and exit_code and exit_code < 0:
+                                    tool_errors.append(ToolError(
+                                        turn=turn + 1, tool_name=tool_name,
+                                        arguments=tool_args_raw[:200],
+                                        error=str(err),
+                                        tool_result=tool_result[:500],
+                                    ))
+                        except (json.JSONDecodeError, TypeError):
+                            pass
+
+                    # Add tool response to conversation
+                    tc_id = tc.get("id", "") if isinstance(tc, dict) else tc.id
+                    messages.append(
+                        {
+                            "role": "tool",
+                            "tool_call_id": tc_id,
+                            "content": tool_result,
+                        }
+                    )
+
+                turn_elapsed = _time.monotonic() - turn_start
+                logger.info(
+                    "[%s] turn %d: api=%.1fs, %d tools, turn_total=%.1fs",
+                    self.task_id[:8], turn + 1, api_elapsed,
+                    len(assistant_msg.tool_calls), turn_elapsed,
+                )
+
+            else:
+                # No tool calls -- model is done
+                msg_dict = {
+                    "role": "assistant",
+                    "content": assistant_msg.content or "",
+                }
+                if reasoning:
+                    msg_dict["reasoning_content"] = reasoning
+                messages.append(msg_dict)
+
+                turn_elapsed = _time.monotonic() - turn_start
+                logger.info(
+                    "[%s] turn %d: api=%.1fs, no tools (finished), turn_total=%.1fs",
+                    self.task_id[:8], turn + 1, api_elapsed, turn_elapsed,
+                )
+
+                return AgentResult(
+                    messages=messages,
+                    managed_state=self._get_managed_state(),
+                    turns_used=turn + 1,
+                    finished_naturally=True,
+                    reasoning_per_turn=reasoning_per_turn,
+                    tool_errors=tool_errors,
+                )
+
+        # Hit max turns without the model stopping
+        logger.info("Agent hit max_turns (%d) without finishing", self.max_turns)
+        return AgentResult(
+            messages=messages,
+            managed_state=self._get_managed_state(),
+            turns_used=self.max_turns,
+            finished_naturally=False,
+            reasoning_per_turn=reasoning_per_turn,
+            tool_errors=tool_errors,
+        )
+
+    def _get_managed_state(self) -> Optional[Dict[str, Any]]:
+        """
+        Get ManagedServer state if the server supports it.
+
+        Returns state dict with SequenceNodes containing tokens/logprobs/masks,
+        or None if the server doesn't support get_state() (e.g., regular OpenAI server).
+        """
+        if hasattr(self.server, "get_state"):
+            return self.server.get_state()
+        return None

environments/benchmarks/taubench/hermes_agent.py

@@ -0,0 +1,324 @@
+"""
+HermesAgent for tau2-bench evaluation.
+
+Implements the tau2 HalfDuplexAgent interface using litellm with OpenRouter,
+matching the inference path used across the rest of the Hermes Agent codebase.
+
+Usage:
+    python environments/benchmarks/taubench/run_eval.py \\
+        --model anthropic/claude-sonnet-4-5 \\
+        --base-url openrouter \\
+        --env retail
+"""
+
+import json
+import os
+import sys
+from pathlib import Path
+from typing import Optional
+
+import litellm
+from pydantic import BaseModel
+
+_repo_root = Path(__file__).resolve().parent.parent.parent.parent
+if str(_repo_root) not in sys.path:
+    sys.path.insert(0, str(_repo_root))
+
+from environments.tool_call_parsers import get_parser
+
+from tau2.agent.base_agent import HalfDuplexAgent, ValidAgentInputMessage
+from tau2.data_model.message import (
+    AssistantMessage,
+    Message,
+    MultiToolMessage,
+    SystemMessage,
+    ToolCall,
+    ToolMessage,
+    UserMessage,
+)
+from tau2.environment.tool import Tool
+
+
+class HermesAgentState(BaseModel):
+    system_messages: list[SystemMessage]
+    messages: list
+
+
+class HermesAgent(HalfDuplexAgent[HermesAgentState]):
+    """
+    tau2 HalfDuplexAgent backed by litellm, using OpenRouter (or any
+    OpenAI-compatible endpoint).
+
+    Registered as "hermes_agent" in the tau2 registry by run_eval.py.
+    """
+
+    SYSTEM_PROMPT = (
+        "You are a customer service agent that helps the user according to the "
+        "<policy> provided below.\n"
+        "In each turn you can either:\n"
+        "- Send a message to the user.\n"
+        "- Make a tool call.\n"
+        "You cannot do both at the same time.\n\n"
+        "Try to be helpful and always follow the policy. "
+        "Always make sure you generate valid JSON only.\n\n"
+        "<policy>\n{domain_policy}\n</policy>"
+    )
+
+    # System prompt variant for qwen3_coder tool format — tools are embedded
+    # directly in the system prompt as <tools> XML instead of passed via the
+    # OpenAI tools= parameter.
+    SYSTEM_PROMPT_QWEN3_CODER = (
+        "You are a customer service agent that helps the user according to the "
+        "<policy> provided below.\n"
+        "In each turn you can either:\n"
+        "- Send a message to the user.\n"
+        "- Make a tool call.\n"
+        "You cannot do both at the same time.\n\n"
+        "Try to be helpful and always follow the policy. "
+        "Always make sure you generate valid JSON only.\n\n"
+        "You may call one or more functions to assist with the user query.\n\n"
+        "You are provided with function signatures within <tools></tools> XML tags:\n"
+        "<tools>\n{tools_json}\n</tools>\n\n"
+        "<policy>\n{domain_policy}\n</policy>"
+    )
+
+    def __init__(
+        self,
+        tools: list[Tool],
+        domain_policy: str,
+        model: str,
+        base_url: Optional[str] = None,
+        api_key: Optional[str] = None,
+        temperature: float = 0.0,
+        max_tokens: Optional[int] = None,
+        top_p: Optional[float] = None,
+        thinking: bool = False,
+        tool_parser: Optional[str] = None,
+    ):
+        super().__init__(tools=tools, domain_policy=domain_policy)
+        self.model = model
+        self.base_url = base_url
+        self.api_key = api_key
+        self.temperature = temperature
+        self.max_tokens = max_tokens
+        self.top_p = top_p
+        self.thinking = thinking
+        self.tool_parser = tool_parser
+        self._parser = get_parser(tool_parser) if tool_parser else None
+
+        # OpenRouter requires specific headers; pass them via litellm extra_headers
+        self._extra_headers: dict = {}
+        if base_url and "openrouter" in base_url.lower():
+            self._extra_headers = {
+                "HTTP-Referer": "https://hermes-agent.nousresearch.com",
+                "X-Title": "Hermes Agent",
+            }
+
+    @property
+    def system_prompt(self) -> str:
+        if self.tool_parser == "qwen3_coder" and self.tools:
+            tools_json = json.dumps(
+                [t.openai_schema for t in self.tools], indent=2, ensure_ascii=False
+            )
+            return self.SYSTEM_PROMPT_QWEN3_CODER.format(
+                tools_json=tools_json,
+                domain_policy=self.domain_policy,
+            )
+        return self.SYSTEM_PROMPT.format(domain_policy=self.domain_policy)
+
+    def get_init_state(
+        self, message_history: Optional[list[Message]] = None
+    ) -> HermesAgentState:
+        return HermesAgentState(
+            system_messages=[SystemMessage(role="system", content=self.system_prompt)],
+            messages=list(message_history or []),
+        )
+
+    def generate_next_message(
+        self, message: ValidAgentInputMessage, state: HermesAgentState
+    ) -> tuple[AssistantMessage, HermesAgentState]:
+        # Append incoming message(s) to history
+        if isinstance(message, MultiToolMessage):
+            state.messages.extend(message.tool_messages)
+        else:
+            state.messages.append(message)
+
+        # Build litellm-compatible message list
+        all_messages = state.system_messages + state.messages
+        lm_messages = [_to_litellm_message(m) for m in all_messages]
+
+        kwargs = dict(
+            model=self.model,
+            messages=lm_messages,
+            temperature=self.temperature,
+        )
+        if self.tools:
+            kwargs["tools"] = [t.openai_schema for t in self.tools]
+        if self.max_tokens is not None:
+            kwargs["max_tokens"] = self.max_tokens
+        if self.top_p is not None:
+            kwargs["top_p"] = self.top_p
+        # Enable thinking/reasoning mode. OpenRouter exposes this as
+        # `include_reasoning` for nemotron (per supported_parameters in the
+        # model metadata). Pass via extra_body to bypass litellm filtering.
+        if self.thinking:
+            kwargs["extra_body"] = {"include_reasoning": True}
+        # Only pass base_url when model doesn't already have a provider prefix
+        # (litellm uses either the prefix OR base_url, not both)
+        if self.base_url and not self.model.startswith("openrouter/"):
+            kwargs["base_url"] = self.base_url
+        if self.api_key:
+            kwargs["api_key"] = self.api_key
+        if self._extra_headers:
+            kwargs["extra_headers"] = self._extra_headers
+
+        response = litellm.completion(**kwargs)
+        assistant_msg = _litellm_response_to_assistant_message(response, parser=self._parser)
+
+        state.messages.append(assistant_msg)
+        return assistant_msg, state
+
+
+# ---------------------------------------------------------------------------
+# Conversion helpers
+# ---------------------------------------------------------------------------
+
+
+def _to_litellm_message(msg) -> dict:
+    """Convert a tau2 message object to a litellm-compatible dict."""
+    if isinstance(msg, SystemMessage):
+        return {"role": "system", "content": msg.content or ""}
+
+    if isinstance(msg, UserMessage):
+        if msg.tool_calls:
+            # User tool calls (tau2 v2 feature — user has tools too)
+            return {
+                "role": "user",
+                "content": msg.content or "",
+                "tool_calls": [_tool_call_to_dict(tc) for tc in msg.tool_calls],
+            }
+        return {"role": "user", "content": msg.content or ""}
+
+    if isinstance(msg, AssistantMessage):
+        d: dict = {"role": "assistant", "content": msg.content or ""}
+        if msg.tool_calls:
+            d["tool_calls"] = [_tool_call_to_dict(tc) for tc in msg.tool_calls]
+        return d
+
+    if isinstance(msg, ToolMessage):
+        return {
+            "role": "tool",
+            "tool_call_id": msg.id,
+            "content": msg.content or "",
+        }
+
+    # Fallback
+    return {"role": getattr(msg, "role", "user"), "content": str(getattr(msg, "content", ""))}
+
+
+def _tool_call_to_dict(tc: ToolCall) -> dict:
+    import json
+    return {
+        "id": tc.id or "call_0",
+        "type": "function",
+        "function": {
+            "name": tc.name,
+            "arguments": json.dumps(tc.arguments),
+        },
+    }
+
+
+def _litellm_response_to_assistant_message(response, parser=None) -> AssistantMessage:
+    """Convert a litellm ModelResponse to a tau2 AssistantMessage."""
+    import json
+
+    choice = response.choices[0]
+    msg = choice.message
+
+    content = msg.content or ""
+    tool_calls_raw = getattr(msg, "tool_calls", None)
+
+    tau2_tool_calls: Optional[list[ToolCall]] = None
+
+    if parser and content:
+        # Use the custom tool parser (e.g. qwen3_coder) to extract tool calls
+        # from the raw text response.
+        parsed_content, parsed_tool_calls = parser.parse(content)
+        if parsed_tool_calls:
+            content = parsed_content or ""
+            tau2_tool_calls = []
+            for tc in parsed_tool_calls:
+                try:
+                    arguments = json.loads(tc.function.arguments or "{}")
+                except json.JSONDecodeError:
+                    arguments = {}
+                tau2_tool_calls.append(
+                    ToolCall(
+                        id=tc.id or "call_0",
+                        name=tc.function.name,
+                        arguments=arguments,
+                        requestor="assistant",
+                    )
+                )
+    elif tool_calls_raw:
+        tau2_tool_calls = []
+        for tc in tool_calls_raw:
+            if hasattr(tc, "function"):
+                name = tc.function.name
+                try:
+                    arguments = json.loads(tc.function.arguments or "{}")
+                except json.JSONDecodeError:
+                    arguments = {}
+                tau2_tool_calls.append(
+                    ToolCall(
+                        id=tc.id or "call_0",
+                        name=name,
+                        arguments=arguments,
+                        requestor="assistant",
+                    )
+                )
+
+    cost = None
+    try:
+        cost = litellm.completion_cost(response)
+    except Exception:
+        pass
+
+    usage = None
+    if hasattr(response, "usage") and response.usage:
+        usage = dict(response.usage)
+
+    return AssistantMessage(
+        role="assistant",
+        content=content if not tau2_tool_calls else None,
+        tool_calls=tau2_tool_calls,
+        cost=cost,
+        usage=usage,
+    )
+
+
+def create_hermes_agent(tools: list[Tool], domain_policy: str, **kwargs) -> HermesAgent:
+    """
+    Factory function registered with the tau2 registry.
+
+    Expected kwargs:
+        model (str): litellm model string
+        base_url (str): API base URL (optional)
+        api_key (str): API key (optional)
+        temperature (float): sampling temperature (default 0.0)
+        top_p (float): nucleus sampling (optional)
+        max_tokens (int): max tokens (optional)
+        thinking (bool): enable reasoning/thinking mode (default False)
+    """
+    return HermesAgent(
+        tools=tools,
+        domain_policy=domain_policy,
+        model=kwargs["model"],
+        base_url=kwargs.get("base_url"),
+        api_key=kwargs.get("api_key"),
+        temperature=kwargs.get("temperature", 0.0),
+        top_p=kwargs.get("top_p"),
+        max_tokens=kwargs.get("max_tokens"),
+        thinking=kwargs.get("thinking", False),
+        tool_parser=kwargs.get("tool_parser"),
+    )

environments/benchmarks/taubench/run_eval.py

@@ -0,0 +1,288 @@
+"""
+tau2-bench evaluation runner for Hermes Agent.
+
+Runs the tau2-bench retail, airline, telecom, or banking_knowledge evaluation
+using HermesAgent backed by litellm — the same inference path used across the
+rest of the Hermes Agent codebase.
+
+Usage:
+    # Against OpenRouter (auto-detects OPENROUTER_API_KEY)
+    python environments/benchmarks/taubench/run_eval.py \\
+        --model openrouter/anthropic/claude-sonnet-4-5 \\
+        --base-url openrouter \\
+        --env retail
+
+    # Against OpenAI directly
+    python environments/benchmarks/taubench/run_eval.py \\
+        --model gpt-4o \\
+        --env retail
+
+    # Local vLLM
+    python environments/benchmarks/taubench/run_eval.py \\
+        --model openai/NousResearch/Hermes-3-Llama-3.1-70B \\
+        --base-url http://localhost:8000/v1 \\
+        --env retail \\
+        --num-trials 3
+
+    # Specific tasks only
+    python environments/benchmarks/taubench/run_eval.py \\
+        --model openrouter/anthropic/claude-sonnet-4-5 \\
+        --base-url openrouter \\
+        --env retail \\
+        --task-ids task_1 task_2 task_5
+
+Results are saved to results/tau2bench/ as JSON.
+
+Dependencies (requires Python 3.12+):
+    pip install "tau2 @ git+https://github.com/sierra-research/tau2-bench.git"
+    # or: pip install -e ".[tau2bench]"
+"""
+
+import argparse
+import logging
+import os
+import sys
+from pathlib import Path
+from typing import Optional
+
+_repo_root = Path(__file__).resolve().parent.parent.parent.parent
+if str(_repo_root) not in sys.path:
+    sys.path.insert(0, str(_repo_root))
+
+from tau2.data_model.simulation import Results, TextRunConfig
+from tau2.evaluator.evaluator import EvaluationType
+from tau2.registry import registry
+from tau2.runner.batch import run_tasks
+from tau2.runner.helpers import get_tasks
+
+from environments.benchmarks.taubench.hermes_agent import create_hermes_agent
+
+logging.basicConfig(
+    level=logging.INFO, format="%(asctime)s %(levelname)s %(name)s: %(message)s"
+)
+logger = logging.getLogger(__name__)
+
+OPENROUTER_BASE_URL = "https://openrouter.ai/api/v1"
+AGENT_NAME = "hermes_agent"
+
+
+def _register_agent(
+    model: str,
+    base_url: Optional[str],
+    api_key: Optional[str],
+    temperature: float,
+    top_p: Optional[float],
+    max_tokens: Optional[int],
+    thinking: bool,
+    tool_parser: Optional[str],
+) -> None:
+    """Register the HermesAgent factory with the tau2 registry (idempotent)."""
+    if registry.get_agent_factory(AGENT_NAME) is not None:
+        return
+
+    def factory(tools, domain_policy, **kwargs):
+        return create_hermes_agent(
+            tools=tools,
+            domain_policy=domain_policy,
+            model=model,
+            base_url=base_url,
+            api_key=api_key,
+            temperature=temperature,
+            top_p=top_p,
+            max_tokens=max_tokens,
+            thinking=thinking,
+            tool_parser=tool_parser,
+        )
+
+    registry.register_agent_factory(factory=factory, name=AGENT_NAME)
+    logger.info("Registered agent factory: %s (model=%s, thinking=%s, tool_parser=%s)", AGENT_NAME, model, thinking, tool_parser)
+
+
+def run_eval(
+    model: str,
+    base_url: Optional[str],
+    api_key: Optional[str],
+    user_model: str,
+    env_name: str,
+    task_split: Optional[str],
+    num_trials: int,
+    max_concurrency: int,
+    max_steps: int,
+    temperature: float,
+    top_p: Optional[float],
+    max_tokens: Optional[int],
+    thinking: bool,
+    tool_parser: Optional[str],
+    task_ids: Optional[list],
+    start_index: int,
+    end_index: int,
+    log_dir: str,
+    seed: int,
+) -> Results:
+    # Resolve OpenRouter shorthand
+    if base_url and base_url.strip().lower() == "openrouter":
+        base_url = OPENROUTER_BASE_URL
+
+    is_openrouter = base_url and "openrouter" in base_url.lower()
+
+    # litellm requires the "openrouter/" prefix to route correctly
+    if is_openrouter and not model.startswith("openrouter/"):
+        model = f"openrouter/{model}"
+    if is_openrouter and not user_model.startswith("openrouter/"):
+        user_model = f"openrouter/{user_model}"
+
+    # Resolve API key
+    if is_openrouter:
+        api_key = api_key or os.environ.get("OPENROUTER_API_KEY") or os.environ.get("OPENAI_API_KEY")
+        # litellm reads OPENAI_API_KEY for base_url overrides; set it so the
+        # user simulator's generate() call also authenticates correctly.
+        if api_key and not os.environ.get("OPENAI_API_KEY"):
+            os.environ["OPENAI_API_KEY"] = api_key
+    else:
+        api_key = api_key or os.environ.get("OPENAI_API_KEY")
+
+    _register_agent(
+        model=model,
+        base_url=base_url,
+        api_key=api_key,
+        temperature=temperature,
+        top_p=top_p,
+        max_tokens=max_tokens,
+        thinking=thinking,
+        tool_parser=tool_parser,
+    )
+
+    # Load tasks — task_ids in tau2 are strings like "task_1"
+    tasks = get_tasks(
+        task_set_name=env_name,
+        task_split_name=task_split,
+        task_ids=[str(i) for i in task_ids] if task_ids else None,
+    )
+
+    if not task_ids and (end_index != -1 or start_index != 0):
+        end = end_index if end_index != -1 else len(tasks)
+        tasks = tasks[start_index:end]
+
+    logger.info(
+        "Running tau2-%s eval: %d tasks, %d trial(s), concurrency=%d",
+        env_name, len(tasks), num_trials, max_concurrency,
+    )
+
+    save_path = Path(log_dir) / f"tau2-{env_name}-{model.split('/')[-1]}.json"
+    save_path.parent.mkdir(parents=True, exist_ok=True)
+
+    # Pass api_key/base_url to user sim via llm_args so tau2's generate() authenticates.
+    # When using OpenRouter for the user sim, mirror the agent's key + endpoint.
+    user_llm_args: dict = {}
+    if is_openrouter and api_key:
+        user_llm_args["api_key"] = api_key
+        user_llm_args["base_url"] = base_url
+
+    config = TextRunConfig(
+        domain=env_name,
+        agent=AGENT_NAME,
+        user="user_simulator",
+        llm_agent=model,
+        llm_args_agent={},
+        llm_user=user_model,
+        llm_args_user=user_llm_args,
+        num_trials=num_trials,
+        max_steps=max_steps,
+        max_concurrency=max_concurrency,
+        seed=seed,
+    )
+
+    results = run_tasks(
+        config,
+        tasks,
+        save_path=save_path,
+        console_display=True,
+        # ALL: respects each task's reward_basis. NL assertions are skipped
+        # gracefully (scored as pass) rather than raising an error, so tasks
+        # are evaluated only on their actual basis components (DB, ACTION, etc.)
+        evaluation_type=EvaluationType.ALL,
+    )
+
+    logger.info("Results saved to %s", save_path)
+    return results
+
+
+def main():
+    parser = argparse.ArgumentParser(
+        description="Run tau2-bench evaluation with Hermes Agent (requires Python 3.12+)",
+        formatter_class=argparse.ArgumentDefaultsHelpFormatter,
+    )
+    parser.add_argument(
+        "--model", required=True,
+        help="litellm model string, e.g. 'openrouter/anthropic/claude-sonnet-4-5' or 'gpt-4o'",
+    )
+    parser.add_argument(
+        "--base-url", default=None,
+        help="API base URL. Use 'openrouter' as shorthand for https://openrouter.ai/api/v1.",
+    )
+    parser.add_argument("--api-key", default=None, help="API key (falls back to OPENROUTER_API_KEY / OPENAI_API_KEY)")
+    parser.add_argument("--temperature", type=float, default=1.0,
+                        help="Sampling temperature. NVIDIA used 1.0 for nemotron-super.")
+    parser.add_argument("--top-p", type=float, default=0.95,
+                        help="Nucleus sampling. NVIDIA used 0.95 for nemotron-super.")
+    parser.add_argument("--max-tokens", type=int, default=None)
+    parser.add_argument("--thinking", action="store_true", default=False,
+                        help="Enable reasoning/thinking mode (use_reasoning=true). "
+                             "Required to match NVIDIA's reported nemotron-super scores.")
+    parser.add_argument("--tool-parser", default=None,
+                        help="Tool call parser to use (e.g. 'qwen3_coder'). When set, tools are "
+                             "embedded in the system prompt as <tools> XML and responses are parsed "
+                             "from raw text instead of using OpenAI function calling format.")
+    parser.add_argument(
+        "--user-model", default="qwen/qwen3-235b-a22b-2507:nitro",
+        help="litellm model string for the tau2 user simulator. "
+             "Defaults to qwen/qwen3-235b-a22b-2507:nitro (instruct, non-thinking) to match NVIDIA's eval setup. "
+             "When using --base-url openrouter the openrouter/ prefix is added automatically.",
+    )
+    parser.add_argument(
+        "--env", default="retail",
+        choices=["retail", "airline", "telecom", "banking_knowledge", "mock"],
+    )
+    parser.add_argument(
+        "--task-split", default=None,
+        help="Task split name (e.g. 'base'). Defaults to the domain default.",
+    )
+    parser.add_argument("--num-trials", type=int, default=1)
+    parser.add_argument("--max-concurrency", type=int, default=8)
+    parser.add_argument("--max-steps", type=int, default=50)
+    parser.add_argument(
+        "--task-ids", nargs="*", default=None,
+        help="Specific task IDs to run (tau2 task IDs are strings like 'task_1')",
+    )
+    parser.add_argument("--start-index", type=int, default=0)
+    parser.add_argument("--end-index", type=int, default=-1)
+    parser.add_argument("--seed", type=int, default=10)
+    parser.add_argument("--log-dir", default="results/tau2bench")
+
+    args = parser.parse_args()
+
+    run_eval(
+        model=args.model,
+        base_url=args.base_url,
+        api_key=args.api_key,
+        user_model=args.user_model,
+        env_name=args.env,
+        task_split=args.task_split,
+        num_trials=args.num_trials,
+        max_concurrency=args.max_concurrency,
+        max_steps=args.max_steps,
+        temperature=args.temperature,
+        top_p=args.top_p,
+        max_tokens=args.max_tokens,
+        thinking=args.thinking,
+        tool_parser=args.tool_parser,
+        task_ids=args.task_ids,
+        start_index=args.start_index,
+        end_index=args.end_index,
+        log_dir=args.log_dir,
+        seed=args.seed,
+    )
+
+
+if __name__ == "__main__":
+    main()

environments/benchmarks/tblite/README.md

@@ -0,0 +1,73 @@
+# OpenThoughts-TBLite Evaluation Environment
+
+This environment evaluates terminal agents on the [OpenThoughts-TBLite](https://huggingface.co/datasets/open-thoughts/OpenThoughts-TBLite) benchmark, a difficulty-calibrated subset of [Terminal-Bench 2.0](https://www.tbench.ai/leaderboard/terminal-bench/2.0).
+
+## Source
+
+OpenThoughts-TBLite was created by the [OpenThoughts](https://www.openthoughts.ai/) Agent team in collaboration with [Snorkel AI](https://snorkel.ai/) and [Bespoke Labs](https://bespokelabs.ai/). The original dataset and documentation live at:
+
+- **Dataset (source):** [open-thoughts/OpenThoughts-TBLite](https://huggingface.co/datasets/open-thoughts/OpenThoughts-TBLite)
+- **GitHub:** [open-thoughts/OpenThoughts-TBLite](https://github.com/open-thoughts/OpenThoughts-TBLite)
+- **Blog post:** [openthoughts.ai/blog/openthoughts-tblite](https://www.openthoughts.ai/blog/openthoughts-tblite)
+
+## Our Dataset
+
+We converted the source into the same schema used by our Terminal-Bench 2.0 environment (pre-built Docker Hub images, base64-encoded test tarballs, etc.) and published it as:
+
+- **Dataset (ours):** [NousResearch/openthoughts-tblite](https://huggingface.co/datasets/NousResearch/openthoughts-tblite)
+- **Docker images:** `nousresearch/tblite-<task-name>:latest` on Docker Hub (100 images)
+
+The conversion script is at `scripts/prepare_tblite_dataset.py`.
+
+## Why TBLite?
+
+Terminal-Bench 2.0 is one of the strongest frontier evaluations for terminal agents, but when a model scores near the floor (e.g., Qwen 3 8B at <1%), many changes look identical in aggregate score. TBLite addresses this by calibrating task difficulty using Claude Haiku 4.5 as a reference:
+
+| Difficulty | Pass Rate Range | Tasks |
+|------------|----------------|-------|
+| Easy       | >= 70%         | 40    |
+| Medium     | 40-69%         | 26    |
+| Hard       | 10-39%         | 26    |
+| Extreme    | < 10%          | 8     |
+
+This gives enough solvable tasks to detect small improvements quickly, while preserving enough hard tasks to avoid saturation. The correlation between TBLite and TB2 scores is **r = 0.911**.
+
+TBLite also runs 2.6-8x faster than the full TB2, making it practical for iteration loops.
+
+## Usage
+
+```bash
+# Run the full benchmark
+python environments/benchmarks/tblite/tblite_env.py evaluate
+
+# Filter to specific tasks
+python environments/benchmarks/tblite/tblite_env.py evaluate \
+    --env.task_filter "broken-python,pandas-etl"
+
+# Use a different model
+python environments/benchmarks/tblite/tblite_env.py evaluate \
+    --server.model_name "qwen/qwen3-30b"
+```
+
+## Architecture
+
+`TBLiteEvalEnv` is a thin subclass of `TerminalBench2EvalEnv`. All evaluation logic (agent loop, Docker sandbox management, test verification, metrics) is inherited. Only the defaults differ:
+
+| Setting        | TB2                              | TBLite                                  |
+|----------------|----------------------------------|-----------------------------------------|
+| Dataset        | `NousResearch/terminal-bench-2`  | `NousResearch/openthoughts-tblite`      |
+| Tasks          | 89                               | 100                                     |
+| Task timeout   | 1800s (30 min)                   | 1200s (20 min)                          |
+| Wandb name     | `terminal-bench-2`               | `openthoughts-tblite`                   |
+
+## Citation
+
+```bibtex
+@software{OpenThoughts-TBLite,
+  author = {OpenThoughts-Agent team, Snorkel AI, Bespoke Labs},
+  month = Feb,
+  title = {{OpenThoughts-TBLite: A High-Signal Benchmark for Iterating on Terminal Agents}},
+  howpublished = {https://www.openthoughts.ai/blog/openthoughts-tblite},
+  year = {2026}
+}
+```

environments/benchmarks/tblite/default.yaml

@@ -0,0 +1,39 @@
+# OpenThoughts-TBLite Evaluation -- Default Configuration
+#
+# Eval-only environment for the TBLite benchmark (100 difficulty-calibrated
+# terminal tasks, a faster proxy for Terminal-Bench 2.0).
+# Uses Modal terminal backend for per-task cloud-isolated sandboxes
+# and OpenRouter for inference.
+#
+# Usage:
+#   python environments/benchmarks/tblite/tblite_env.py evaluate \
+#       --config environments/benchmarks/tblite/default.yaml
+#
+#   # Override model:
+#   python environments/benchmarks/tblite/tblite_env.py evaluate \
+#       --config environments/benchmarks/tblite/default.yaml \
+#       --openai.model_name anthropic/claude-sonnet-4
+
+env:
+  enabled_toolsets: ["terminal", "file"]
+  max_agent_turns: 60
+  max_token_length: 32000
+  agent_temperature: 0.8
+  terminal_backend: "modal"
+  terminal_timeout: 300        # 5 min per command (builds, pip install)
+  tool_pool_size: 128          # thread pool for 100 parallel tasks
+  dataset_name: "NousResearch/openthoughts-tblite"
+  test_timeout: 600
+  task_timeout: 1200           # 20 min wall-clock per task (TBLite tasks are faster)
+  tokenizer_name: "NousResearch/Hermes-3-Llama-3.1-8B"
+  use_wandb: true
+  wandb_name: "openthoughts-tblite"
+  ensure_scores_are_not_same: false
+  data_dir_to_save_evals: "environments/benchmarks/evals/openthoughts-tblite"
+
+openai:
+  base_url: "https://openrouter.ai/api/v1"
+  model_name: "anthropic/claude-opus-4.6"
+  server_type: "openai"
+  health_check: false
+  # api_key loaded from OPENROUTER_API_KEY in .env

environments/benchmarks/tblite/local.yaml

@@ -0,0 +1,38 @@
+# OpenThoughts-TBLite Evaluation -- Docker Backend (Local Compute)
+#
+# Runs tasks in Docker containers on the local machine.
+# Sandboxed like Modal but no cloud costs. Good for dev/testing.
+#
+# Usage:
+#   python environments/benchmarks/tblite/tblite_env.py evaluate \
+#       --config environments/benchmarks/tblite/local.yaml
+#
+#   # Override concurrency:
+#   python environments/benchmarks/tblite/tblite_env.py evaluate \
+#       --config environments/benchmarks/tblite/local.yaml \
+#       --env.eval_concurrency 4
+
+env:
+  enabled_toolsets: ["terminal", "file"]
+  max_agent_turns: 60
+  max_token_length: 32000
+  agent_temperature: 0.8
+  terminal_backend: "docker"
+  terminal_timeout: 300
+  tool_pool_size: 16
+  dataset_name: "NousResearch/openthoughts-tblite"
+  test_timeout: 600
+  task_timeout: 1200
+  eval_concurrency: 8          # max 8 tasks at once
+  tokenizer_name: "NousResearch/Hermes-3-Llama-3.1-8B"
+  use_wandb: false
+  wandb_name: "openthoughts-tblite-local"
+  ensure_scores_are_not_same: false
+  data_dir_to_save_evals: "environments/benchmarks/evals/openthoughts-tblite-local"
+
+openai:
+  base_url: "https://openrouter.ai/api/v1"
+  model_name: "anthropic/claude-sonnet-4"
+  server_type: "openai"
+  health_check: false
+  # api_key loaded from OPENROUTER_API_KEY in .env

environments/benchmarks/tblite/local_vllm.yaml

@@ -0,0 +1,40 @@
+# OpenThoughts-TBLite Evaluation -- Local vLLM Backend
+#
+# Runs against a local vLLM server with Docker sandboxes.
+#
+# Start the vLLM server from the atropos directory:
+#   python -m example_trainer.vllm_api_server \
+#       --model Qwen/Qwen3-4B-Instruct-2507 \
+#       --port 9001 \
+#       --gpu-memory-utilization 0.8 \
+#       --max-model-len=32000
+#
+# Then run:
+#   python environments/benchmarks/tblite/tblite_env.py evaluate \
+#       --config environments/benchmarks/tblite/local_vllm.yaml
+
+env:
+  enabled_toolsets: ["terminal", "file"]
+  max_agent_turns: 60
+  max_token_length: 16000
+  agent_temperature: 0.6
+  terminal_backend: "docker"
+  terminal_timeout: 300
+  tool_pool_size: 16
+  dataset_name: "NousResearch/openthoughts-tblite"
+  test_timeout: 600
+  task_timeout: 1200
+  eval_concurrency: 8
+  tool_call_parser: "hermes"
+  system_prompt: "You are an expert terminal agent. You MUST use the provided tools to complete tasks. Use the terminal tool to run shell commands, read_file to read files, write_file to write files, search_files to search, and patch to edit files. Do NOT write out solutions as text - execute them using the tools. Always start by exploring the environment with terminal commands."
+  tokenizer_name: "Qwen/Qwen3-4B-Instruct-2507"
+  use_wandb: false
+  wandb_name: "tblite-qwen3-4b-instruct"
+  ensure_scores_are_not_same: false
+  data_dir_to_save_evals: "environments/benchmarks/evals/tblite-qwen3-4b-local"
+
+openai:
+  base_url: "http://localhost:9001"
+  model_name: "Qwen/Qwen3-4B-Instruct-2507"
+  server_type: "vllm"
+  health_check: false

environments/benchmarks/tblite/run_eval.sh

@@ -0,0 +1,42 @@
+#!/bin/bash
+
+# OpenThoughts-TBLite Evaluation
+#
+# Run from repo root:
+#   bash environments/benchmarks/tblite/run_eval.sh
+#
+# Override model:
+#   bash environments/benchmarks/tblite/run_eval.sh \
+#       --openai.model_name anthropic/claude-sonnet-4
+#
+# Run a subset:
+#   bash environments/benchmarks/tblite/run_eval.sh \
+#       --env.task_filter broken-python,pandas-etl
+#
+# All terminal settings (backend, timeout, lifetime, pool size) are
+# configured via env config fields -- no env vars needed.
+
+set -euo pipefail
+
+mkdir -p logs evals/openthoughts-tblite
+LOG_FILE="logs/tblite_$(date +%Y%m%d_%H%M%S).log"
+
+echo "OpenThoughts-TBLite Evaluation"
+echo "Log file: $LOG_FILE"
+echo ""
+
+# Unbuffered python output so logs are written in real-time
+export PYTHONUNBUFFERED=1
+
+# Show INFO-level agent loop timing (api/tool durations per turn)
+# These go to the log file; tqdm + [START]/[PASS]/[FAIL] go to terminal
+export LOGLEVEL=INFO
+
+python tblite_env.py evaluate \
+  --config default.yaml \
+  "$@" \
+  2>&1 | tee "$LOG_FILE"
+
+echo ""
+echo "Log saved to: $LOG_FILE"
+echo "Eval results: evals/openthoughts-tblite/"

environments/benchmarks/tblite/tblite_env.py

@@ -0,0 +1,119 @@
+"""
+OpenThoughts-TBLite Evaluation Environment
+
+A lighter, faster alternative to Terminal-Bench 2.0 for iterating on terminal
+agents. Uses the same evaluation logic as TerminalBench2EvalEnv but defaults
+to the NousResearch/openthoughts-tblite dataset (100 difficulty-calibrated
+tasks vs TB2's 89 harder tasks).
+
+TBLite tasks are a curated subset of TB2 with a difficulty distribution
+designed to give meaningful signal even for smaller models:
+  - Easy (40 tasks):   >= 70% pass rate with Claude Haiku 4.5
+  - Medium (26 tasks): 40-69% pass rate
+  - Hard (26 tasks):   10-39% pass rate
+  - Extreme (8 tasks): < 10% pass rate
+
+Usage:
+    python environments/benchmarks/tblite/tblite_env.py evaluate
+
+    # Filter to specific tasks:
+    python environments/benchmarks/tblite/tblite_env.py evaluate \\
+        --env.task_filter "broken-python,pandas-etl"
+"""
+
+import os
+import sys
+from pathlib import Path
+from typing import List, Tuple
+
+_repo_root = Path(__file__).resolve().parent.parent.parent.parent
+if str(_repo_root) not in sys.path:
+    sys.path.insert(0, str(_repo_root))
+
+from pydantic import Field
+
+from atroposlib.envs.base import EvalHandlingEnum
+from atroposlib.envs.server_handling.server_manager import APIServerConfig
+
+from environments.benchmarks.terminalbench_2.terminalbench2_env import (
+    TerminalBench2EvalConfig,
+    TerminalBench2EvalEnv,
+)
+
+
+class TBLiteEvalConfig(TerminalBench2EvalConfig):
+    """Configuration for the OpenThoughts-TBLite evaluation environment.
+
+    Inherits all TB2 config fields. Only the dataset default and task timeout
+    differ -- TBLite tasks are calibrated to be faster.
+    """
+
+    dataset_name: str = Field(
+        default="NousResearch/openthoughts-tblite",
+        description="HuggingFace dataset containing TBLite tasks.",
+    )
+
+    task_timeout: int = Field(
+        default=1200,
+        description="Maximum wall-clock seconds per task. TBLite tasks are "
+        "generally faster than TB2, so 20 minutes is usually sufficient.",
+    )
+
+
+class TBLiteEvalEnv(TerminalBench2EvalEnv):
+    """OpenThoughts-TBLite evaluation environment.
+
+    Inherits all evaluation logic from TerminalBench2EvalEnv (agent loop,
+    test verification, Docker image resolution, metrics, wandb logging).
+    Only the default configuration differs.
+    """
+
+    name = "openthoughts-tblite"
+    env_config_cls = TBLiteEvalConfig
+
+    @classmethod
+    def config_init(cls) -> Tuple[TBLiteEvalConfig, List[APIServerConfig]]:
+        env_config = TBLiteEvalConfig(
+            enabled_toolsets=["terminal", "file"],
+            disabled_toolsets=None,
+            distribution=None,
+
+            max_agent_turns=60,
+            max_token_length=16000,
+            agent_temperature=0.6,
+            system_prompt=None,
+
+            terminal_backend="modal",
+            terminal_timeout=300,
+
+            test_timeout=180,
+
+            # 100 tasks in parallel
+            tool_pool_size=128,
+
+            eval_handling=EvalHandlingEnum.STOP_TRAIN,
+            group_size=1,
+            steps_per_eval=1,
+            total_steps=1,
+
+            tokenizer_name="NousResearch/Hermes-3-Llama-3.1-8B",
+            use_wandb=True,
+            wandb_name="openthoughts-tblite",
+            ensure_scores_are_not_same=False,
+        )
+
+        server_configs = [
+            APIServerConfig(
+                base_url="https://openrouter.ai/api/v1",
+                model_name="anthropic/claude-sonnet-4",
+                server_type="openai",
+                api_key=os.getenv("OPENROUTER_API_KEY", ""),
+                health_check=False,
+            )
+        ]
+
+        return env_config, server_configs
+
+
+if __name__ == "__main__":
+    TBLiteEvalEnv.cli()

environments/benchmarks/terminalbench_2/default.yaml

@@ -0,0 +1,42 @@
+# Terminal-Bench 2.0 Evaluation -- Default Configuration
+#
+# Eval-only environment for the TB2 benchmark (89 terminal tasks).
+# Uses Modal terminal backend for per-task cloud-isolated sandboxes
+# and OpenRouter for inference.
+#
+# Usage:
+#   python environments/benchmarks/terminalbench_2/terminalbench2_env.py evaluate \
+#       --config environments/benchmarks/terminalbench_2/default.yaml
+#
+#   # Override model:
+#   python environments/benchmarks/terminalbench_2/terminalbench2_env.py evaluate \
+#       --config environments/benchmarks/terminalbench_2/default.yaml \
+#       --openai.model_name anthropic/claude-sonnet-4
+
+env:
+  enabled_toolsets: ["terminal", "file"]
+  max_agent_turns: 60
+  max_token_length: 32000
+  agent_temperature: 0.8
+  terminal_backend: "modal"
+  terminal_timeout: 300        # 5 min per command (builds, pip install)
+  tool_pool_size: 128          # thread pool for 89 parallel tasks
+  dataset_name: "NousResearch/terminal-bench-2"
+  test_timeout: 600
+  task_timeout: 1800           # 30 min wall-clock per task, auto-FAIL if exceeded
+  tokenizer_name: "NousResearch/Hermes-3-Llama-3.1-8B"
+  use_wandb: true
+  wandb_name: "terminal-bench-2"
+  ensure_scores_are_not_same: false
+  data_dir_to_save_evals: "environments/benchmarks/evals/terminal-bench-2"
+  # CRITICAL: Limit concurrent Modal sandbox creations to avoid deadlocks.
+  # Modal's blocking calls (App.lookup, etc.) deadlock when too many sandboxes
+  # are created simultaneously inside thread pool workers via asyncio.run().
+  max_concurrent_tasks: 8
+
+openai:
+  base_url: "https://openrouter.ai/api/v1"
+  model_name: "anthropic/claude-opus-4.6"
+  server_type: "openai"
+  health_check: false
+  # api_key loaded from OPENROUTER_API_KEY in .env

environments/benchmarks/terminalbench_2/run_eval.sh

@@ -0,0 +1,42 @@
+#!/bin/bash
+
+# Terminal-Bench 2.0 Evaluation
+#
+# Run from repo root:
+#   bash environments/benchmarks/terminalbench_2/run_eval.sh
+#
+# Override model:
+#   bash environments/benchmarks/terminalbench_2/run_eval.sh \
+#       --openai.model_name anthropic/claude-sonnet-4
+#
+# Run a subset:
+#   bash environments/benchmarks/terminalbench_2/run_eval.sh \
+#       --env.task_filter fix-git,git-multibranch
+#
+# All terminal settings (backend, timeout, lifetime, pool size) are
+# configured via env config fields -- no env vars needed.
+
+set -euo pipefail
+
+mkdir -p logs evals/terminal-bench-2
+LOG_FILE="logs/terminalbench2_$(date +%Y%m%d_%H%M%S).log"
+
+echo "Terminal-Bench 2.0 Evaluation"
+echo "Log file: $LOG_FILE"
+echo ""
+
+# Unbuffered python output so logs are written in real-time
+export PYTHONUNBUFFERED=1
+
+# Show INFO-level agent loop timing (api/tool durations per turn)
+# These go to the log file; tqdm + [START]/[PASS]/[FAIL] go to terminal
+export LOGLEVEL=INFO
+
+python terminalbench2_env.py evaluate \
+  --config default.yaml \
+  "$@" \
+  2>&1 | tee "$LOG_FILE"
+
+echo ""
+echo "Log saved to: $LOG_FILE"
+echo "Eval results: evals/terminal-bench-2/"