readme edit

.clinerules

@@ -0,0 +1,115 @@
+# Cline's Memory Bank
+
+I am Cline, an expert software engineer with a unique characteristic: my memory resets completely between sessions. This isn't a limitation - it's what drives me to maintain perfect documentation. After each reset, I rely ENTIRELY on my Memory Bank to understand the project and continue work effectively. I MUST read ALL memory bank files at the start of EVERY task - this is not optional.
+
+## Memory Bank Structure
+
+The Memory Bank consists of core files and optional context files, all in Markdown format. Files build upon each other in a clear hierarchy:
+
+flowchart TD
+    PB[projectbrief.md] --> PC[productContext.md]
+    PB --> SP[systemPatterns.md]
+    PB --> TC[techContext.md]
+
+    PC --> AC[activeContext.md]
+    SP --> AC
+    TC --> AC
+
+    AC --> P[progress.md]
+
+### Core Files (Required)
+1. `projectbrief.md`
+   - Foundation document that shapes all other files
+   - Created at project start if it doesn't exist
+   - Defines core requirements and goals
+   - Source of truth for project scope
+
+2. `productContext.md`
+   - Why this project exists
+   - Problems it solves
+   - How it should work
+   - User experience goals
+
+3. `activeContext.md`
+   - Current work focus
+   - Recent changes
+   - Next steps
+   - Active decisions and considerations
+   - Important patterns and preferences
+   - Learnings and project insights
+
+4. `systemPatterns.md`
+   - System architecture
+   - Key technical decisions
+   - Design patterns in use
+   - Component relationships
+   - Critical implementation paths
+
+5. `techContext.md`
+   - Technologies used
+   - Development setup
+   - Technical constraints
+   - Dependencies
+   - Tool usage patterns
+
+6. `progress.md`
+   - What works
+   - What's left to build
+   - Current status
+   - Known issues
+   - Evolution of project decisions
+
+### Additional Context
+Create additional files/folders within memory-bank/ when they help organize:
+- Complex feature documentation
+- Integration specifications
+- API documentation
+- Testing strategies
+- Deployment procedures
+
+## Core Workflows
+
+### Plan Mode
+flowchart TD
+    Start[Start] --> ReadFiles[Read Memory Bank]
+    ReadFiles --> CheckFiles{Files Complete?}
+
+    CheckFiles -->|No| Plan[Create Plan]
+    Plan --> Document[Document in Chat]
+
+    CheckFiles -->|Yes| Verify[Verify Context]
+    Verify --> Strategy[Develop Strategy]
+    Strategy --> Present[Present Approach]
+
+### Act Mode
+flowchart TD
+    Start[Start] --> Context[Check Memory Bank]
+    Context --> Update[Update Documentation]
+    Update --> Execute[Execute Task]
+    Execute --> Document[Document Changes]
+
+## Documentation Updates
+
+Memory Bank updates occur when:
+1. Discovering new project patterns
+2. After implementing significant changes
+3. When user requests with **update memory bank** (MUST review ALL files)
+4. When context needs clarification
+
+flowchart TD
+    Start[Update Process]
+
+    subgraph Process
+        P1[Review ALL Files]
+        P2[Document Current State]
+        P3[Clarify Next Steps]
+        P4[Document Insights & Patterns]
+
+        P1 --> P2 --> P3 --> P4
+    end
+
+    Start --> Process
+
+Note: When triggered by **update memory bank**, I MUST review every memory bank file, even if some don't require updates. Focus particularly on activeContext.md and progress.md as they track current state.
+
+REMEMBER: After every memory reset, I begin completely fresh. The Memory Bank is my only link to previous work. It must be maintained with precision and clarity, as my effectiveness depends entirely on its accuracy.

.cursorrules

@@ -0,0 +1,201 @@
+Hermes-Agent is an agent harness for LLMs with an interactive CLI.
+
+## Development Environment
+
+**IMPORTANT**: Always use the virtual environment if it exists:
+```bash
+source venv/bin/activate  # Before running any Python commands
+```
+
+## Project Structure
+
+- `hermes` - CLI launcher script (run with `./hermes`)
+- `cli.py` - Interactive CLI with Rich UI, prompt_toolkit, animated spinners
+- `cli-config.yaml` - CLI configuration (model, terminal, toolsets, personalities)
+- `tools/` - Individual tool implementations (web, terminal, browser, vision, etc.)
+- `tools/__init__.py` - Exports all tools for importing
+- `model_tools.py` - Consolidates tool schemas and handlers for the agent
+- `toolsets.py` - Groups tools into logical toolsets (web, terminal, browser, etc.)
+- `toolset_distributions.py` - Probability-based tool selection for data generation
+- `run_agent.py` - Primary agent runner with AIAgent class and KawaiiSpinner
+- `batch_runner.py` - Parallel batch processing with checkpointing
+- `tests/` - Test scripts
+
+## File Dependency Chain
+
+```
+tools/*.py → tools/__init__.py → model_tools.py → toolsets.py → toolset_distributions.py
+                                       ↑
+run_agent.py ──────────────────────────┘
+cli.py → run_agent.py (uses AIAgent with quiet_mode=True)
+batch_runner.py → run_agent.py + toolset_distributions.py
+```
+
+Always ensure consistency between tools, model_tools.py, and toolsets.py when changing any of them.
+
+## CLI Architecture (cli.py)
+
+The interactive CLI uses:
+- **Rich** - For the welcome banner and styled panels
+- **prompt_toolkit** - For fixed input area with history and `patch_stdout`
+- **KawaiiSpinner** (in run_agent.py) - Animated feedback during API calls and tool execution
+
+Key components:
+- `HermesCLI` class - Main CLI controller with commands and conversation loop
+- `load_cli_config()` - Loads `cli-config.yaml`, sets environment variables for terminal
+- `build_welcome_banner()` - Displays ASCII art logo, tools, and skills summary
+- `/commands` - Process user commands like `/help`, `/clear`, `/personality`, etc.
+
+CLI uses `quiet_mode=True` when creating AIAgent to suppress verbose logging and enable kawaii-style feedback instead.
+
+### Adding CLI Commands
+
+1. Add to `COMMANDS` dict with description
+2. Add handler in `process_command()` method
+3. For persistent settings, use `save_config_value()` to update `cli-config.yaml`
+
+## Adding a New Tool
+
+Follow this strict order to maintain consistency:
+
+1. Create `tools/your_tool.py` with:
+   - Handler function (sync or async) returning a JSON string via `json.dumps()`
+   - `check_*_requirements()` function to verify dependencies (e.g., API keys)
+   - Schema definition following OpenAI function-calling format
+
+2. Export in `tools/__init__.py`:
+   - Import the handler and check function
+   - Add to `__all__` list
+
+3. Register in `model_tools.py`:
+   - Create `get_*_tool_definitions()` function or add to existing
+   - Add routing in `handle_function_call()` dispatcher
+   - Update `get_all_tool_names()` with the tool name
+   - Update `get_toolset_for_tool()` mapping
+   - Update `get_available_toolsets()` and `check_toolset_requirements()`
+
+4. Add to toolset in `toolsets.py`:
+   - Add to existing toolset or create new one in TOOLSETS dict
+
+5. Optionally add to `toolset_distributions.py` for batch processing
+
+## Tool Implementation Pattern
+
+```python
+# tools/example_tool.py
+import json
+import os
+
+def check_example_requirements() -> bool:
+    """Check if required API keys/dependencies are available."""
+    return bool(os.getenv("EXAMPLE_API_KEY"))
+
+def example_tool(param: str, task_id: str = None) -> str:
+    """Execute the tool and return JSON string result."""
+    try:
+        result = {"success": True, "data": "..."}
+        return json.dumps(result, ensure_ascii=False)
+    except Exception as e:
+        return json.dumps({"error": str(e)}, ensure_ascii=False)
+```
+
+All tool handlers MUST return a JSON string. Never return raw dicts.
+
+## Stateful Tools
+
+Tools that maintain state (terminal, browser) require:
+- `task_id` parameter for session isolation between concurrent tasks
+- `cleanup_*()` function to release resources
+- Cleanup is called automatically in run_agent.py after conversation completes
+
+## Environment Variables
+
+API keys are loaded from `.env` file in repo root:
+- `OPENROUTER_API_KEY` - Main LLM API access (primary provider)
+- `FIRECRAWL_API_KEY` - Web search/extract tools
+- `BROWSERBASE_API_KEY` / `BROWSERBASE_PROJECT_ID` - Browser automation
+- `FAL_KEY` - Image generation (FLUX model)
+- `NOUS_API_KEY` - Vision and Mixture-of-Agents tools
+
+Terminal tool configuration (can also be set in `cli-config.yaml`):
+- `TERMINAL_ENV` - Backend: local, docker, singularity, modal, or ssh
+- `TERMINAL_CWD` - Working directory
+- `TERMINAL_SSH_HOST`, `TERMINAL_SSH_USER`, `TERMINAL_SSH_KEY` - For SSH backend
+
+## Agent Loop (run_agent.py)
+
+The AIAgent class handles:
+- Processing enabled toolsets to provide to the model
+- Piping prompts to the agent
+- Looping LLM calls when tools are invoked, until natural language response
+- Returning the final response
+
+Uses OpenAI-compatible API (primarily OpenRouter) with the OpenAI Python SDK.
+
+## Reasoning Model Support
+
+For models that support chain-of-thought reasoning:
+- Extract `reasoning_content` from API responses
+- Store in `assistant_msg["reasoning"]` for trajectory export
+- Pass back via `reasoning_content` field on subsequent turns
+
+## Trajectory Format
+
+Conversations are saved in ShareGPT format for training:
+```json
+{"from": "system", "value": "System prompt with <tools>...</tools>"}
+{"from": "human", "value": "User message"}
+{"from": "gpt", "value": "<think>reasoning</think>\n<tool_call>{...}</tool_call>"}
+{"from": "tool", "value": "<tool_response>{...}</tool_response>"}
+{"from": "gpt", "value": "Final response"}
+```
+
+Tool calls use `<tool_call>` XML tags, responses use `<tool_response>` tags, reasoning uses `<think>` tags.
+
+## Batch Processing (batch_runner.py)
+
+For processing multiple prompts:
+- Parallel execution with multiprocessing
+- Content-based resume for fault tolerance (matches on prompt text, not indices)
+- Toolset distributions control probabilistic tool availability per prompt
+- Output: `data/<run_name>/trajectories.jsonl` (combined) + individual batch files
+
+## Logging
+
+Trajectories restructure tools as a system prompt for storage in a format suitable for later training use.
+
+## Skills System
+
+Skills are on-demand knowledge documents the agent can load. Located in `skills/` directory:
+
+```
+skills/
+├── mlops/                    # Category folder
+│   ├── axolotl/             # Skill folder
+│   │   ├── SKILL.md         # Main instructions (required)
+│   │   ├── references/      # Additional docs, API specs
+│   │   └── templates/       # Output formats, configs
+│   └── vllm/
+│       └── SKILL.md
+└── example-skill/
+    └── SKILL.md
+```
+
+**Progressive disclosure** (token-efficient):
+1. `skills_categories()` - List category names (~50 tokens)
+2. `skills_list(category)` - Name + description per skill (~3k tokens)
+3. `skill_view(name)` - Full content + tags + linked files
+
+SKILL.md files use YAML frontmatter:
+```yaml
+---
+name: skill-name
+description: Brief description for listing
+tags: [tag1, tag2]
+related_skills: [other-skill]
+version: 1.0.0
+---
+# Skill Content...
+```
+
+Tool files: `tools/skills_tool.py` → `model_tools.py` → `toolsets.py`

.env.example

@@ -1,113 +1,108 @@
 # Hermes Agent Environment Configuration
 # Copy this file to .env and fill in your API keys
 
+# =============================================================================
+# CORE SETTINGS
+# =============================================================================
+# Agent backend:
+# - openai  : default Hermes-Agent loop (OpenAI function-calling via OpenAI SDK)
+# - atropos : Atroposlib ServerManager/ManagedServer-backed loop (training/env integration)
+HERMES_BACKEND=openai
+
+
+# =============================================================================
+# LOCAL / SELF-HOSTED OPENAI-COMPATIBLE ENDPOINTS (vLLM, SGLang, llama.cpp, etc.)
+# =============================================================================
+# For local development (matches the Atropos test env defaults):
+# ATROPOS_SERVER_BASE_URL=http://127.0.0.1:8080
+# ATROPOS_SERVER_MODEL=hermes-4-36b
+# For hosted inference (Nous Research inference API):
+ATROPOS_SERVER_BASE_URL=
+ATROPOS_SERVER_MODEL=
+ATROPOS_TOKENIZER_NAME=
+# Set this to your Nous API key (Bearer token).
+ATROPOS_SERVER_API_KEY=
+
+# Debugging (prints to stdout; use with care)
+# HERMES_DEBUG_ATROPOS_REQUEST=1
+# HERMES_DEBUG_ATROPOS_RESPONSE=1
+# HERMES_DEBUG_OPENAI_REQUEST=1
+# HERMES_DEBUG_OPENAI_RESPONSE=1
+
+# =============================================================================
+# LOCAL / SELF-HOSTED OPENAI-COMPATIBLE ENDPOINTS (vLLM, SGLang, llama.cpp, etc.)
+# =============================================================================
+# If you set ATROPOS_SERVER_BASE_URL or OPENAI_BASE_URL, Hermes will use it instead
+# of OpenRouter.
+#
+# Local server convenience (base URL without /v1):
+# llama.cpp example (see `Hermes-Agent/scripts/launch_llama_cpp_hermes_4_36b.sh`):
+# ATROPOS_SERVER_BASE_URL=http://127.0.0.1:8080
+# ATROPOS_SERVER_MODEL=hermes-4-36b
+# ATROPOS_TOKENIZER_NAME=NousResearch/Hermes-4.3-36B
+# ATROPOS_SERVER_API_KEY=local
+#
+# Hosted Nous inference API:
+# ATROPOS_SERVER_BASE_URL=https://inference-api.nousresearch.com
+# ATROPOS_SERVER_MODEL=Hermes-4.3-36B
+# ATROPOS_TOKENIZER_NAME=NousResearch/Hermes-4.3-36B
+# ATROPOS_SERVER_API_KEY=sk-... (Bearer token)
+#
+# If you plan to run GRPO-style group sampling (e.g. `--env.group_size 4`) against
+# llama.cpp, start the server with at least that many slots, e.g.:
+#   LLAMA_CPP_PARALLEL=4 Hermes-Agent/scripts/launch_llama_cpp_hermes_4_36b.sh
+#
+# Generic OpenAI-compatible (base URL should include /v1):
+# OPENAI_BASE_URL=http://127.0.0.1:8080/v1
+# OPENAI_API_KEY=local
+
 # =============================================================================
 # 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_BASE_URL=https://openrouter.ai/api/v1
 OPENROUTER_API_KEY=
 
 # Default model to use (OpenRouter format: provider/model)
-# Examples: anthropic/claude-opus-4.6, openai/gpt-4o, google/gemini-3-flash-preview, zhipuai/glm-4-plus
-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=
-# OPENCODE_GO_BASE_URL=https://opencode.ai/zen/go/v1  # Override default base URL
+# Examples: anthropic/claude-sonnet-4, openai/gpt-4o, google/gemini-2.0-flash, zhipuai/glm-4-plus
+LLM_MODEL=anthropic/claude-sonnet-4
 
 # =============================================================================
 # TOOL API KEYS
 # =============================================================================
 
-# 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=
 
+# Nous Research API Key - Vision analysis and multi-model reasoning
+# Get at: https://inference-api.nousresearch.com/
+NOUS_API_KEY=
 
 # FAL.ai API Key - Image generation
 # Get at: https://fal.ai/
 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=
-
 # =============================================================================
 # TERMINAL TOOL CONFIGURATION (mini-swe-agent backend)
 # =============================================================================
 # 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
+# - local: Runs directly on your machine (fastest, no isolation)
+# - ssh: Runs on remote server via SSH (great for sandboxing - agent can't touch its own code)
+# - singularity: Runs in Apptainer/Singularity containers (HPC clusters, no root needed)
+# - docker: Runs in Docker containers (isolated, requires Docker + docker group)
+# - modal: Runs in Modal cloud sandboxes (scalable, requires Modal account)
+TERMINAL_ENV=local
 
 # 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
+TERMINAL_DOCKER_IMAGE=python:3.11
+TERMINAL_SINGULARITY_IMAGE=docker://python:3.11
+TERMINAL_MODAL_IMAGE=python:3.11
 
-
-# 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=.
+# Working directory inside the container
+TERMINAL_CWD=/tmp
 
 # Default command timeout in seconds
 TERMINAL_TIMEOUT=60
@@ -149,12 +144,87 @@ TERMINAL_LIFETIME_SECONDS=300
 # SUDO_PASSWORD=your_password_here
 
 # =============================================================================
-# MODAL CLOUD BACKEND (Optional - for TERMINAL_ENV=modal)
+# MODAL CLOUD BACKEND (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.
+# Modal provides cloud sandboxes with per-second billing and auto-scaling.
+# This implementation uses a warm pool of sandboxes for cost efficiency.
+#
+# SETUP:
+#   pip install modal && modal setup
+#   (Authenticates via browser, stores credentials locally)
+#
+# FEATURES:
+# - Auto-scaling warm sandbox pool (no cold start after first use)
+# - Named sandbox recovery (reconnects after restart)
+# - Profile-based heterogeneous environments (CPU, GPU, different images)
+# - Server-side idle_timeout protection against orphaned sandboxes
+
+# Modal app name (groups all sandboxes, used for recovery)
+TERMINAL_MODAL_APP_NAME=hermes-sandbox
+
+# Default profile when none specified
+TERMINAL_MODAL_DEFAULT_PROFILE=default
+
+# Profile config file (optional - YAML format, see modal_profiles.yaml)
+# TERMINAL_MODAL_PROFILES_FILE=modal_profiles.yaml
+
+# --- Default Profile Settings (used if no YAML file) ---
+# These apply when no profile is specified or for the "default" profile
+TERMINAL_MODAL_IMAGE=python:3.11
+TERMINAL_MODAL_MIN_POOL=1
+TERMINAL_MODAL_MAX_POOL=5
+TERMINAL_MODAL_IDLE_TIMEOUT=120
+TERMINAL_MODAL_MAX_LIFETIME=3600
+TERMINAL_MODAL_SCALE_DOWN_IDLE=180
+
+# --- Custom Profile Example: pytorch-gpu ---
+# Uncomment to enable a GPU profile for ML tasks
+# Usage: terminal_tool("python train.py", profile="pytorch-gpu")
+#
+# TERMINAL_MODAL_PROFILE_pytorch_gpu_IMAGE=pytorch/pytorch:2.1.0-cuda12.1-cudnn8-runtime
+# TERMINAL_MODAL_PROFILE_pytorch_gpu_GPU=T4
+# TERMINAL_MODAL_PROFILE_pytorch_gpu_MEMORY=16384
+# TERMINAL_MODAL_PROFILE_pytorch_gpu_MIN_POOL=0
+# TERMINAL_MODAL_PROFILE_pytorch_gpu_MAX_POOL=2
+# TERMINAL_MODAL_PROFILE_pytorch_gpu_IDLE_TIMEOUT=60
+
+# --- Custom Profile Example: node ---
+# Uncomment to enable a Node.js profile
+# Usage: terminal_tool("npm test", profile="node")
+#
+# TERMINAL_MODAL_PROFILE_node_IMAGE=node:18
+# TERMINAL_MODAL_PROFILE_node_MIN_POOL=0
+# TERMINAL_MODAL_PROFILE_node_MAX_POOL=3
+
+# =============================================================================
+# MODAL SECRETS (Secure credential injection)
+# =============================================================================
+# Modal Secrets allow you to securely pass API keys, passwords, and other
+# sensitive data to your sandboxes without exposing them in code or logs.
+#
+# SETUP SECRETS:
+#   1. Via Dashboard: https://modal.com/secrets
+#   2. Via CLI: modal secret create my-secret KEY1=value1 KEY2=value2
+#   3. Via CLI with env: modal secret create my-secret API_KEY="$API_KEY"
+#
+# LIST SECRETS:
+#   modal secret list
+#
+# DELETE SECRETS:
+#   modal secret delete my-secret
+
+# Global secrets applied to ALL profiles (comma-separated secret names)
+# These secrets must be created on Modal dashboard or via CLI first
+# TERMINAL_MODAL_SECRETS=my-api-keys,database-creds
+
+# Per-profile secrets (comma-separated secret names)
+# TERMINAL_MODAL_PROFILE_pytorch_gpu_SECRETS=huggingface-token,wandb-key
+
+# Per-profile environment variables (semicolon-separated KEY=VALUE pairs)
+# TERMINAL_MODAL_PROFILE_default_ENV_VARS=DEBUG=1;LOG_LEVEL=info
+
+# Load local .env file into sandbox (useful for development)
+# TERMINAL_MODAL_PROFILE_default_USE_DOTENV=true
 
 # =============================================================================
 # BROWSER TOOL CONFIGURATION (agent-browser + Browserbase)
@@ -197,55 +267,16 @@ BROWSER_INACTIVITY_TIMEOUT=120
 # Contains full conversation history in trajectory format for debugging/replay
 
 # =============================================================================
-# VOICE TRANSCRIPTION & OPENAI TTS
+# LEGACY/OPTIONAL API KEYS
 # =============================================================================
-# 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-...
+# Morph API Key - For legacy Hecate terminal backend (terminal-hecate tool)
+# Get at: https://morph.so/
+MORPH_API_KEY=
 
-# 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=
-
-# 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)
+# Hecate VM Settings (only if using terminal-hecate tool)
+HECATE_VM_LIFETIME_SECONDS=300
+HECATE_DEFAULT_SNAPSHOT_ID=snapshot_p5294qxt
 
 # =============================================================================
 # DEBUG OPTIONS
@@ -254,69 +285,3 @@ 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

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -1,144 +0,0 @@
-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

@@ -1,11 +0,0 @@
-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

@@ -1,73 +0,0 @@
-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

@@ -1,100 +0,0 @@
-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

@@ -1,75 +0,0 @@
-## 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

@@ -1,60 +0,0 @@
-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:
-    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/docs-site-checks.yml

@@ -1,39 +0,0 @@
-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/tests.yml

@@ -1,42 +0,0 @@
-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

@@ -1,55 +1,64 @@
-/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
+/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
+data/*
+node_modules/
+browser-use/
+agent-browser/
+# Private keys
+*.ppk
+*.pem
+privvy*
+images/
+
+# CLI config (may contain sensitive SSH paths)
+cli-config.yaml
+
+.DS_Store
+
+# artifacts
+*.jsonl
+*.html
+*.json
+*.log
+*.csv
+
+# Singularity/Apptainer images (large binary files)
+*.sif
+
+# Test files
+test_singularity_*.py
+test_*.py
+!tests/test_*.py
+
+# Nomad data
+/tmp/NomadClient*/

.gitmodules

@@ -1,6 +1,3 @@
 [submodule "mini-swe-agent"]
 	path = mini-swe-agent
 	url = https://github.com/SWE-agent/mini-swe-agent
-[submodule "tinker-atropos"]
-	path = tinker-atropos
-	url = https://github.com/nousresearch/tinker-atropos

.plans/openai-api-server.md

@@ -1,291 +0,0 @@
-# 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

@@ -1,705 +0,0 @@
-# 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

@@ -1,385 +0,0 @@
-# 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
-│   ├── 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
-│   └── 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
-
-**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.
-
-**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
-
----
-
-## Known Pitfalls
-
-### 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.
-
-### 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.
-
----
-
-## 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

@@ -1,659 +0,0 @@
-# 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="$(pwd)/venv"
-
-# Install with all extras (messaging, cron, CLI menus, dev tools)
-uv pip install -e ".[all,dev]"
-uv pip install -e "./mini-swe-agent"
-uv pip install -e "./tinker-atropos"
-
-# Optional: browser tools
-npm install
-```
-
-### Configure for development
-
-```bash
-mkdir -p ~/.hermes/{cron,sessions,logs,memories,skills}
-cp cli-config.yaml.example ~/.hermes/config.yaml
-touch ~/.hermes/.env
-
-# Add at minimum an LLM provider key:
-echo 'OPENROUTER_API_KEY=sk-or-v1-your-key' >> ~/.hermes/.env
-```
-
-### Run
-
-```bash
-# Symlink for global access
-mkdir -p ~/.local/bin
-ln -sf "$(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).

LICENSE

@@ -1,21 +0,0 @@
-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.

README.md

@@ -1,177 +1,857 @@
-<p align="center">
-  <img src="assets/banner.png" alt="Hermes Agent" width="100%">
-</p>
+# Hermes Agent
 
-# Hermes Agent ☤
+An AI agent with advanced tool-calling capabilities, featuring a flexible toolsets system for organizing and managing tools.
 
-<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>
+## Features
 
-**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.
+- **Interactive CLI**: Beautiful terminal interface with animated feedback, personalities, and session management
+- **Web Tools**: Search, extract content, and crawl websites
+- **Terminal Tools**: Execute commands via local, Docker, Singularity, Modal, or SSH backends
+- **Browser Tools**: Automate web browsers to navigate, click, type, and extract content
+- **Vision Tools**: Analyze images from URLs
+- **Reasoning Tools**: Advanced multi-model reasoning (Mixture of Agents)
+- **Creative Tools**: Generate images from text prompts
+- **Skills Tools**: On-demand knowledge documents with progressive disclosure
+- **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
 
-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.
-
-<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>
-
----
-
-## Quick Install
+## Quick Start (CLI)
 
 ```bash
-curl -fsSL https://raw.githubusercontent.com/NousResearch/hermes-agent/main/scripts/install.sh | bash
+# After setup (see below), just run:
+./hermes
+
+# Or with options:
+./hermes --model "anthropic/claude-sonnet-4" --toolsets "web,terminal"
 ```
 
-Works on Linux, macOS, and WSL2. The installer handles everything — Python, Node.js, dependencies, and the `hermes` command. No prerequisites except git.
+The CLI provides:
+- Animated spinners during thinking and tool execution
+- Kawaii-style feedback messages
+- `/commands` for configuration, history, and session management
+- Customizable personalities (`/personality kawaii`, `/personality pirate`, etc.)
+- Persistent configuration via `cli-config.yaml`
 
-> **Windows:** Native Windows is not supported. Please install [WSL2](https://learn.microsoft.com/en-us/windows/wsl/install) and run the command above.
-
-After installation:
+## Setup
 
+### 1. Clone the Repository
 ```bash
-source ~/.bashrc    # reload shell (or: source ~/.zshrc)
-hermes              # start chatting!
+# Clone with submodules (recommended)
+git clone --recurse-submodules https://github.com/NousResearch/Hermes-Agent.git
+cd Hermes-Agent
+
+# Or if already cloned without submodules:
+git submodule update --init --recursive
 ```
 
----
-
-## Getting Started
-
+### 2. Install Dependencies
 ```bash
-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
+# Create and activate virtual environment (recommended)
+python3 -m venv venv
+source venv/bin/activate  # On Windows: venv\Scripts\activate
+
+# Install Python packages
+pip install -r requirements.txt
+
+# Install mini-swe-agent for terminal tools
+pip install -e ./mini-swe-agent
+
+# Install Node.js dependencies for browser tools (requires Node.js)
+npm install
 ```
 
-📖 **[Full documentation →](https://hermes-agent.nousresearch.com/docs/)**
-
-## CLI vs Messaging Quick Reference
-
-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.
-
-| 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` |
-
-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).
-
----
-
-## Documentation
-
-All documentation lives at **[hermes-agent.nousresearch.com/docs](https://hermes-agent.nousresearch.com/docs/)**:
-
-| 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 |
-
----
-
-## 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:**
-
+### 3. Configure Environment Variables
 ```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
+# Copy the example environment file
+cp .env.example .env
+
+# Edit .env and add your API keys
+nano .env  # or use your preferred editor
 ```
 
-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`)
+**Required API Keys:**
+- `OPENROUTER_API_KEY` - LLM access via OpenRouter (get at: https://openrouter.ai/keys)
+- `FIRECRAWL_API_KEY` - Web tools (get at: https://firecrawl.dev/)
+- `NOUS_API_KEY` - Vision & reasoning tools (get at: https://inference-api.nousresearch.com/)
+- `FAL_KEY` - Image generation (get at: https://fal.ai/)
 
-See `hermes claw migrate --help` for all options, or use the `openclaw-migration` skill for an interactive agent-guided migration with dry-run previews.
+**Optional API Keys (for specific features):**
+- `BROWSERBASE_API_KEY` - Browser automation (get at: https://browserbase.com/)
+- `BROWSERBASE_PROJECT_ID` - From Browserbase dashboard
+- `MORPH_API_KEY` - For legacy Hecate terminal backend (get at: https://morph.so/)
 
----
+### 4. Configure Terminal Backend
 
-## 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:
+The terminal tool uses **mini-swe-agent** environments. Configure in `.env` or `cli-config.yaml`:
 
 ```bash
-git clone https://github.com/NousResearch/hermes-agent.git
-cd hermes-agent
-git submodule update --init mini-swe-agent   # required terminal backend
-curl -LsSf https://astral.sh/uv/install.sh | sh
-uv venv .venv --python 3.11
+# Backend: "local", "docker", "singularity", "modal", or "ssh"
+TERMINAL_ENV=local          # Default: runs on host machine (no isolation)
+TERMINAL_ENV=ssh            # Remote execution via SSH (agent code stays local)
+TERMINAL_ENV=singularity    # Recommended for HPC: Apptainer/Singularity containers
+TERMINAL_ENV=docker         # Isolated Docker containers
+TERMINAL_ENV=modal          # Cloud execution via Modal
+
+# Container image (for docker/singularity/modal backends)
+TERMINAL_DOCKER_IMAGE=python:3.11-slim
+TERMINAL_SINGULARITY_IMAGE=docker://python:3.11-slim
+TERMINAL_TIMEOUT=60
+
+# SSH backend (for ssh)
+TERMINAL_SSH_HOST=my-server.example.com
+TERMINAL_SSH_USER=myuser
+TERMINAL_SSH_KEY=~/.ssh/id_rsa  # Optional, uses ssh-agent if not set
+```
+
+**Backend Requirements:**
+- **local**: No extra setup (runs directly on your machine, no isolation)
+- **ssh**: SSH access to remote machine (great for sandboxing - agent can't touch its own code)
+- **singularity**: Requires Apptainer or Singularity installed (common on HPC clusters, no root needed)
+- **docker**: Requires Docker installed and user in `docker` group
+- **modal**: Requires Modal account (see setup below)
+
+### Singularity/Apptainer Setup (Recommended for HPC)
+
+Singularity/Apptainer provides rootless container execution, ideal for HPC clusters:
+
+```bash
+# 1. Verify Apptainer is installed
+apptainer --version  # or: singularity --version
+
+# 2. Set up cache directories (important for parallel workers)
+# Use /scratch if available (HPC), otherwise /tmp
+export APPTAINER_CACHEDIR=/scratch/$USER/.apptainer
+export APPTAINER_TMPDIR=/scratch/$USER/.apptainer/tmp
+mkdir -p "$APPTAINER_CACHEDIR" "$APPTAINER_TMPDIR"
+
+# 3. Pre-build SIF image (recommended for parallel batch processing)
+# This avoids race conditions when multiple workers start simultaneously
+apptainer build $APPTAINER_CACHEDIR/python-nodejs.sif docker://nikolaik/python-nodejs:python3.11-nodejs20
+
+# 4. Configure .env to use the local SIF
+TERMINAL_ENV=singularity
+TERMINAL_SINGULARITY_IMAGE=/scratch/$USER/.apptainer/python-nodejs.sif
+```
+
+**Tip:** The batch scripts in `configs/` automatically handle SIF pre-building if `/scratch` is available.
+
+### Modal Cloud Backend Setup
+
+[Modal](https://modal.com) provides serverless cloud compute for running sandboxed environments at scale.
+
+```bash
+# 1. Install Modal and dependencies
+pip install modal boto3
+
+# 2. Authenticate with Modal (opens browser)
+modal setup
+
+# 3. Set terminal backend to modal in .env
+TERMINAL_ENV=modal
+```
+
+Modal uses CLI-based authentication (stored in `~/.modal/`), so no API key is needed in `.env`. After running `modal setup`, commands will automatically execute in Modal's cloud sandboxes.
+
+### Browser Tools Setup
+
+Browser tools enable the agent to navigate websites, fill forms, click buttons, and extract content. They use [agent-browser](https://github.com/vercel-labs/agent-browser) CLI with [Browserbase](https://browserbase.com) cloud execution.
+
+```bash
+# 1. Install Node.js (if not already installed)
+# Use nvm (recommended) or your package manager
+
+# 2. Install agent-browser CLI (choose one option):
+npm install -g agent-browser     # Option A: Global install (recommended)
+npm install                      # Option B: Local install (uses npx fallback)
+
+# 3. Get Browserbase credentials
+# Sign up at https://browserbase.com/ and get your:
+# - API Key (from Settings → API Keys)
+# - Project ID (from your project dashboard)
+
+# 4. Add to your .env file:
+BROWSERBASE_API_KEY=your_api_key_here
+BROWSERBASE_PROJECT_ID=your_project_id_here
+```
+
+**Available Browser Tools:**
+
+| Tool | Description |
+|------|-------------|
+| `browser_navigate` | Navigate to a URL |
+| `browser_snapshot` | Get text-based page snapshot with element refs |
+| `browser_click` | Click an element by ref (e.g., `@e5`) |
+| `browser_type` | Type text into an input field |
+| `browser_scroll` | Scroll up or down |
+| `browser_back` | Go back in browser history |
+| `browser_press` | Press a keyboard key (Enter, Tab, etc.) |
+| `browser_close` | Close the browser session |
+| `browser_get_images` | Get list of images on the page |
+
+**Example Usage:**
+```bash
+# Use browser tools with web search and vision
+python run_agent.py \
+  --query "Go to amazon.com and find the price of the latest Kindle" \
+  --enabled_toolsets=browser,web,vision
+
+# Use browser-focused distribution
+python batch_runner.py \
+  --dataset_file=browser_tasks.jsonl \
+  --distribution=browser_use \
+  --run_name=browser_run
+```
+
+See `.env.example` for all available configuration options including debug settings.
+
+### Skills Tools
+
+Skills are on-demand knowledge documents the agent can load when needed. They follow a **progressive disclosure** pattern to minimize token usage:
+
+```
+skills/
+├── mlops/                    # Category folder
+│   ├── axolotl/             # Skill folder
+│   │   ├── SKILL.md         # Main instructions (required)
+│   │   ├── references/      # Additional docs, API specs
+│   │   └── templates/       # Output formats, configs
+│   └── vllm/
+│       └── SKILL.md
+```
+
+**Available Skills Tools:**
+
+| Tool | Description |
+|------|-------------|
+| `skills_categories` | List available skill categories (~50 tokens) |
+| `skills_list` | List skills with name + description (~3k tokens for 40 skills) |
+| `skill_view` | Load full skill content, tags, and linked files |
+
+**Example Usage:**
+```bash
+# Use skills tools
+python run_agent.py \
+  --query "What skills do you have for fine-tuning? Show me the axolotl skill." \
+  --enabled_toolsets=skills
+```
+
+**Creating Skills:**
+
+Skills use YAML frontmatter for metadata:
+```yaml
+---
+name: my-skill
+description: Brief description shown in skills_list
+tags: [tag1, tag2]
+related_skills: [other-skill]
+version: 1.0.0
+---
+# Skill Content
+
+Instructions, examples, and guidelines here...
+```
+
+Skills can include:
+- `references/` - Additional documentation, API specs, examples
+- `templates/` - Output formats, config files, boilerplate code
+- `scripts/` - Executable helpers (Python, shell scripts)
+
+## Session Logging
+
+Every conversation is automatically logged to `logs/` for debugging and inspection:
+
+```
+logs/
+├── session_20260201_143052_a1b2c3.json
+├── session_20260201_150217_d4e5f6.json
+└── ...
+```
+
+**Log Format:**
+```json
+{
+  "session_id": "20260201_143052_a1b2c3",
+  "model": "anthropic/claude-sonnet-4",
+  "session_start": "2026-02-01T14:30:52.123456",
+  "last_updated": "2026-02-01T14:35:12.789012",
+  "message_count": 8,
+  "conversations": [
+    {"from": "system", "value": "..."},
+    {"from": "human", "value": "..."},
+    {"from": "gpt", "value": "..."},
+    {"from": "tool", "value": "..."}
+  ]
+}
+```
+
+- **Automatic**: Logs are created and updated automatically after each conversation turn
+- **Session ID in Banner**: The CLI displays the session ID in the welcome banner
+- **Trajectory Format**: Uses the same format as batch processing for consistency
+- **Git Ignored**: `logs/` is in `.gitignore` so logs aren't committed
+
+## Interactive CLI
+
+The CLI provides a rich interactive experience for working with the agent.
+
+### Running the CLI
+
+```bash
+# Basic usage
+./hermes
+
+# With specific model
+./hermes --model "anthropic/claude-sonnet-4"
+
+# With specific toolsets
+./hermes --toolsets "web,terminal,skills"
+```
+
+### CLI Commands
+
+| Command | Description |
+|---------|-------------|
+| `/help` | Show available commands |
+| `/tools` | List available tools by toolset |
+| `/toolsets` | List available toolsets |
+| `/model [name]` | Show or change the current model |
+| `/prompt [text]` | View/set custom system prompt |
+| `/personality [name]` | Set a predefined personality |
+| `/clear` | Clear screen and reset conversation |
+| `/reset` | Reset conversation only |
+| `/history` | Show conversation history |
+| `/save` | Save current conversation to file |
+| `/config` | Show current configuration |
+| `/quit` | Exit the CLI |
+
+### Configuration
+
+Copy `cli-config.yaml.example` to `cli-config.yaml` and customize:
+
+```yaml
+# Model settings
+model:
+  default: "anthropic/claude-sonnet-4"
+
+# Terminal backend (local, docker, singularity, modal, or ssh)
+terminal:
+  env_type: "local"
+  cwd: "."  # Use current directory
+
+# Or use SSH for remote execution (keeps agent code isolated)
+# terminal:
+#   env_type: "ssh"
+#   ssh_host: "my-server.example.com"
+#   ssh_user: "myuser"
+#   ssh_key: "~/.ssh/id_rsa"
+#   cwd: "/home/myuser/project"
+
+# Enable specific toolsets
+toolsets:
+  - all  # or: web, terminal, browser, vision, etc.
+
+# Custom personalities (use with /personality command)
+agent:
+  personalities:
+    helpful: "You are a helpful assistant."
+    kawaii: "You are a kawaii assistant! Use cute expressions..."
+```
+
+### Personalities
+
+Built-in personalities available via `/personality`:
+- `helpful`, `concise`, `technical`, `creative`, `teacher`
+- `kawaii`, `catgirl`, `pirate`, `shakespeare`, `surfer`
+- `noir`, `uwu`, `philosopher`, `hype`
+
+## 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
+
+```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
+```
+
+See `toolsets.py` for the complete list of available toolsets and how to create custom ones.
+
+## Basic Usage
+
+### Default (all tools enabled)
+```bash
+# Uses OpenRouter by default - just set OPENROUTER_API_KEY in .env
+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 anthropic/claude-sonnet-4-20250514
+```
+
+### With specific toolset
+```bash
+python run_agent.py \
+  --query "Debug this Python error" \
+  --enabled_toolsets=debugging \
+  --model anthropic/claude-sonnet-4-20250514
+```
+
+### Python API
+```python
+from run_agent import AIAgent
+
+# Uses OpenRouter by default (reads OPENROUTER_API_KEY from .env)
+agent = AIAgent(
+    model="anthropic/claude-sonnet-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:
+
+```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
+```
+
+**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
+
+Use `--list_distributions` to see available toolset distributions for varied data generation.
+
+### Trajectory Compression
+
+Post-process trajectories to fit within token budgets for training:
+
+```bash
+# Compress a directory of JSONL files
+python trajectory_compressor.py --input=data/my_run
+
+# Compress a single JSONL file
+python trajectory_compressor.py --input=data/trajectories.jsonl
+
+# Compress a 15% sample (useful for creating smaller training sets)
+python trajectory_compressor.py --input=data/trajectories.jsonl --sample_percent=15
+
+# Custom output and token target
+python trajectory_compressor.py \
+  --input=data/trajectories.jsonl \
+  --output=data/compressed.jsonl \
+  --target_max_tokens=16000
+```
+
+**Features:**
+- Protects first turns (system, human, first GPT response, first tool call)
+- Protects last N turns (configurable)
+- Summarizes middle turns using LLM to fit target token budget
+- Supports both directory and single file input
+- Optional random sampling with `--sample_percent`
+- Configurable via `configs/trajectory_compression.yaml`
+
+### 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."
+```
+
+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.
+
+The ephemeral prompt influences model behavior during execution, but **only the standard tool-calling system prompt** is saved in trajectory files.
+
+## Command Line Arguments
+
+**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
+
+**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
+
+## Environment Variables
+
+All environment variables can be configured in the `.env` file (copy from `.env.example`).
+
+**LLM Provider (OpenRouter):**
+- `OPENROUTER_API_KEY`: Primary LLM access via OpenRouter (supports Claude, GPT-4, Gemini, etc.)
+- `LLM_MODEL`: Default model (e.g., `anthropic/claude-sonnet-4`, `openai/gpt-4o`)
+
+**Tool API Keys:**
+- `FIRECRAWL_API_KEY`: Web tools (search, extract, crawl)
+- `NOUS_API_KEY`: Vision and reasoning tools
+- `FAL_KEY`: Image generation tools
+
+**Terminal Tool Configuration (mini-swe-agent backend):**
+- `TERMINAL_ENV`: Backend type - `local`, `docker`, `singularity`, `modal`, or `ssh` (default: `local`)
+- `TERMINAL_DOCKER_IMAGE`: Docker image for docker backend (default: `python:3.11-slim`)
+- `TERMINAL_SINGULARITY_IMAGE`: Singularity/Apptainer image (can be `docker://...` URL or local `.sif` path)
+- `TERMINAL_TIMEOUT`: Command timeout in seconds (default: `60`)
+- `TERMINAL_LIFETIME_SECONDS`: Cleanup inactive environments after this time (default: `300`)
+- `TERMINAL_CWD`: Working directory inside containers (default: `/tmp`)
+- `TERMINAL_SCRATCH_DIR`: Custom scratch directory for sandbox storage (optional, auto-detects `/scratch`)
+- `SUDO_PASSWORD`: Enable sudo commands by piping password via `sudo -S` (works with all backends)
+  - If unset in CLI mode, you'll be prompted interactively when sudo is needed (45s timeout)
+
+**SSH Backend Configuration (for remote execution):**
+- `TERMINAL_SSH_HOST`: Remote server hostname or IP
+- `TERMINAL_SSH_USER`: SSH username
+- `TERMINAL_SSH_PORT`: SSH port (default: `22`)
+- `TERMINAL_SSH_KEY`: Path to SSH private key (optional, uses ssh-agent if not set)
+
+**Browser Tool Configuration (agent-browser + Browserbase):**
+- `BROWSERBASE_API_KEY`: Browserbase API key for cloud browser execution
+- `BROWSERBASE_PROJECT_ID`: Browserbase project ID
+- `BROWSER_SESSION_TIMEOUT`: Session timeout in seconds (default: `300`)
+
+**Legacy Hecate Terminal Backend (optional):**
+- `MORPH_API_KEY`: For Hecate/MorphCloud terminal backend
+- `HECATE_VM_LIFETIME_SECONDS`: VM lifetime (default: 300)
+- `HECATE_DEFAULT_SNAPSHOT_ID`: Default snapshot (default: snapshot_p5294qxt)
+
+**Debug Options:**
+- `WEB_TOOLS_DEBUG`, `VISION_TOOLS_DEBUG`, `MOA_TOOLS_DEBUG`, `IMAGE_TOOLS_DEBUG`: Enable debug logging
+
+## Key Files
+
+| File | Purpose |
+|------|---------|
+| `hermes` | CLI launcher script (run with `./hermes`) |
+| `cli.py` | Interactive CLI implementation |
+| `cli-config.yaml` | CLI configuration (copy from `.example`) |
+| `run_agent.py` | Main agent runner - single query execution |
+| `batch_runner.py` | Parallel batch processing with checkpointing |
+| `model_tools.py` | Core tool definitions and handlers |
+| `toolsets.py` | Toolset definitions and composition |
+| `toolset_distributions.py` | Probability distributions for data generation |
+| `trajectory_compressor.py` | Post-process trajectories for training |
+| `tools/` | Individual tool implementations |
+| `tools/skills_tool.py` | Skills system with progressive disclosure |
+| `skills/` | On-demand knowledge documents |
+| `docs/` | Documentation |
+| `configs/` | Example batch run scripts |
+
+# Atropos Integrations & RL Training
+
+Atropos is an RL training framework that uses Hermes-Agent for agent-based environments. This section covers setting up the sandbox infrastructure with either Docker or Singularity backends.
+
+## Prerequisites
+
+### 1. Install Nomad
+Nomad is a workload orchestrator that manages the sandbox containers:
+
+```bash
+# Install Nomad (Linux)
+curl -fsSL https://apt.releases.hashicorp.com/gpg | sudo gpg --dearmor -o /usr/share/keyrings/hashicorp-archive-keyring.gpg
+echo "deb [signed-by=/usr/share/keyrings/hashicorp-archive-keyring.gpg] https://apt.releases.hashicorp.com $(lsb_release -cs) main" | sudo tee /etc/apt/sources.list.d/hashicorp.list
+sudo apt update && sudo apt install nomad
+
+# Verify installation
+nomad --version
+```
+
+For other platforms, see: https://developer.hashicorp.com/nomad/docs/install
+
+### 2. Install Atropos Dependencies
+
+```bash
+python3 -m venv .venv
 source .venv/bin/activate
-uv pip install -e ".[all,dev]"
-uv pip install -e "./mini-swe-agent"
-python -m pytest tests/ -q
+pip install -e '.[atropos]'
 ```
 
-> **RL Training (optional):** To work on the RL/Tinker-Atropos integration, also run:
-> ```bash
-> git submodule update --init tinker-atropos
-> uv pip install -e "./tinker-atropos"
-> ```
+## Backend Options
+
+Atropos supports two container backends for the sandbox environment:
+
+| Backend | Use Case | Requirements |
+|---------|----------|--------------|
+| **Docker** | Development, servers with Docker | Docker installed, user in `docker` group |
+| **Singularity** | HPC clusters, rootless environments | Apptainer/Singularity installed (no root needed) |
 
 ---
 
-## Community
+## Docker Backend (Default)
 
-- 💬 [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)
+### 1. Build the Sandbox Image
+
+```bash
+cd atropos
+docker build -t atropos-sandbox:local .
+```
+
+### 2. Start Nomad (Development Mode)
+
+```bash
+# Start Nomad with Docker driver
+nomad agent -dev -config=nomad-dev.hcl
+```
+
+Or create `nomad-dev.hcl`:
+```hcl
+client {
+  enabled = true
+  options {
+    "driver.allowlist" = "docker"
+  }
+}
+```
+
+### 3. Run with Docker Backend
+
+```bash
+source .venv/bin/activate
+
+# Test the environment
+python -m atropos.envs.swe_smith_oracle_env process \
+    --env.use_wandb false \
+    --env.total_steps 1 \
+    --env.max_items 1 \
+    --env.driver docker
+```
 
 ---
 
-## License
+## Singularity Backend (HPC/Rootless)
 
-MIT — see [LICENSE](LICENSE).
+Singularity/Apptainer is ideal for HPC clusters where Docker requires root privileges.
 
-Built by [Nous Research](https://nousresearch.com).
+### 1. Build the Singularity Image
+
+```bash
+cd atropos
+
+# Option A: Convert from Docker image (if Docker is available)
+docker build -t atropos-sandbox:local .
+apptainer build atropos-sandbox.sif docker-daemon://atropos-sandbox:local
+
+# Option B: Build directly from Dockerfile (requires root or fakeroot)
+apptainer build atropos-sandbox.sif docker://ghcr.io/nousresearch/atropos-sandbox:latest
+```
+
+### 2. Start Nomad with raw_exec Driver
+
+Singularity uses Nomad's `raw_exec` driver. Create `nomad-singularity.hcl`:
+
+```hcl
+client {
+  enabled = true
+  options {
+    "driver.allowlist" = "raw_exec,docker"
+  }
+}
+
+plugin "raw_exec" {
+  config {
+    enabled = true
+  }
+}
+```
+
+Start Nomad:
+```bash
+nomad agent -dev -config=nomad-singularity.hcl
+```
+
+### 3. Run with Singularity Backend
+
+```bash
+source .venv/bin/activate
+
+# Basic test
+python -m atropos.envs.swe_smith_oracle_env process \
+    --env.use_wandb false \
+    --env.total_steps 1 \
+    --env.max_items 1 \
+    --env.driver singularity \
+    --env.singularity_image /path/to/atropos-sandbox.sif
+
+# Full example with all options
+python -m atropos.envs.swe_smith_oracle_env process \
+    --env.use_wandb false \
+    --env.total_steps 10 \
+    --env.group_size 4 \
+    --env.max_items 100 \
+    --env.driver singularity \
+    --env.singularity_image /path/to/atropos-sandbox.sif \
+    --env.slots_per_container 10 \
+    --env.min_containers 1 \
+    --env.max_containers 5
+```
+
+---
+
+## CLI Arguments Reference
+
+### Environment Configuration (`--env.*`)
+
+| Argument | Default | Description |
+|----------|---------|-------------|
+| `--env.driver` | `docker` | Container backend: `docker` or `singularity` |
+| `--env.singularity_image` | - | Path to `.sif` file (required for singularity driver) |
+| `--env.sandbox_image` | `atropos-sandbox:local` | Docker image name (for docker driver) |
+| `--env.slots_per_container` | `10` | Number of parallel slots per container |
+| `--env.min_containers` | `1` | Minimum number of containers to run |
+| `--env.max_containers` | `10` | Maximum containers for auto-scaling |
+| `--env.nomad_address` | `http://localhost:4646` | Nomad server address |
+| `--env.privileged` | `false` | Run containers in privileged mode (Docker only) |
+
+### Processing Configuration
+
+| Argument | Default | Description |
+|----------|---------|-------------|
+| `--env.total_steps` | `1` | Number of processing steps |
+| `--env.group_size` | `1` | Items per processing group |
+| `--env.max_items` | `0` | Max dataset items (0 = all) |
+| `--env.use_wandb` | `true` | Enable Weights & Biases logging |
+| `--env.agent_max_steps` | `50` | Max agent steps per trajectory |
+
+---
+
+## Troubleshooting
+
+### Port Already in Use
+```bash
+# Find and kill process on port 8080
+lsof -ti :8080 | xargs kill
+
+# Or use a different port
+--env.port 8081
+```
+
+### Singularity: Permission Denied
+```bash
+# Check Apptainer is installed
+apptainer --version
+
+# Ensure the .sif file is readable
+ls -la /path/to/atropos-sandbox.sif
+```
+
+### Nomad: Job Not Starting
+```bash
+# Check Nomad status
+nomad status
+
+# View job logs
+nomad alloc logs -job atropos-sandbox-agent-env
+
+# Check stderr for errors
+nomad alloc logs -stderr -job atropos-sandbox-agent-env
+```
+
+### OpenAI API Token Error
+If you see `NotImplementedError: OpenAI endpoints do not support token IDs`:
+```bash
+# For testing/evaluation only (not training)
+export ATROPOS_ALLOW_DUMMY_MANAGED_SERVER=1
+```
+
+---
+
+## Example: Full HPC Workflow
+
+```bash
+# 1. Setup environment
+python3 -m venv .venv
+source .venv/bin/activate
+pip install -e '.[atropos]'
+
+# 2. Build Singularity image (on a machine with Docker)
+cd atropos
+docker build -t atropos-sandbox:local .
+apptainer build atropos-sandbox.sif docker-daemon://atropos-sandbox:local
+
+# 3. Transfer .sif to HPC cluster
+scp atropos-sandbox.sif user@hpc-cluster:/scratch/user/
+
+# 4. On HPC cluster: Start Nomad
+nomad agent -dev -config=nomad-singularity.hcl &
+
+# 5. Run training
+python -m atropos.envs.swe_smith_oracle_env process \
+    --env.driver singularity \
+    --env.singularity_image /scratch/user/atropos-sandbox.sif \
+    --env.total_steps 100 \
+    --env.max_items 1000
+```

RELEASE_v0.2.0.md

@@ -1,383 +0,0 @@
-# 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

@@ -1,377 +0,0 @@
-# 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)

TODO.md

@@ -0,0 +1,729 @@
+# Hermes Agent - Future Improvements
+
+> Ideas for enhancing the agent's capabilities, generated from self-analysis of the codebase.
+
+---
+
+## 🚨 HIGH PRIORITY - Immediate Fixes
+
+These items need to be addressed ASAP:
+
+### 1. SUDO Breaking Terminal Tool 🔐 ✅ COMPLETE
+- [x] **Problem:** SUDO commands break the terminal tool execution (hangs indefinitely)
+- [x] **Fix:** Created custom environment wrappers in `tools/terminal_tool.py`
+  - `stdin=subprocess.DEVNULL` prevents hanging on interactive prompts
+  - Sudo fails gracefully with clear error if no password configured
+  - Same UX as Claude Code - agent sees error, tells user to run it themselves
+- [x] **All 5 environments now have consistent behavior:**
+  - `_LocalEnvironment` - local execution
+  - `_DockerEnvironment` - Docker containers
+  - `_SingularityEnvironment` - Singularity/Apptainer containers
+  - `_ModalEnvironment` - Modal cloud sandboxes
+  - `_SSHEnvironment` - remote SSH execution
+- [x] **Optional sudo support via `SUDO_PASSWORD` env var:**
+  - Shared `_transform_sudo_command()` helper used by all environments
+  - If set, auto-transforms `sudo cmd` → pipes password via `sudo -S`
+  - Documented in `.env.example`, `cli-config.yaml`, and README
+  - Works for chained commands: `cmd1 && sudo cmd2`
+- [x] **Interactive sudo prompt in CLI mode:**
+  - When sudo detected and no password configured, prompts user
+  - 45-second timeout (auto-skips if no input)
+  - Hidden password input via `getpass` (password not visible)
+  - Password cached for session (don't ask repeatedly)
+  - Spinner pauses during prompt for clean UX
+  - Uses `HERMES_INTERACTIVE` env var to detect CLI mode
+
+### 2. Fix `browser_get_images` Tool 🖼️ ✅ VERIFIED WORKING
+- [x] **Tested:** Tool works correctly on multiple sites
+- [x] **Results:** Successfully extracts image URLs, alt text, dimensions
+- [x] **Note:** Some sites (Pixabay, etc.) have Cloudflare bot protection that blocks headless browsers - this is expected behavior, not a bug
+
+### 3. Better Action Logging for Debugging 📝 ✅ COMPLETE
+- [x] **Problem:** Need better logging of agent actions for debugging
+- [x] **Implementation:**
+  - Save full session trajectories to `logs/` directory as JSON
+  - Each session gets a unique file: `session_YYYYMMDD_HHMMSS_UUID.json`
+  - Logs all messages, tool calls with inputs/outputs, timestamps
+  - Structured JSON format for easy parsing and replay
+  - Automatic on CLI runs (configurable)
+
+### 4. Stream Thinking Summaries in Real-Time 💭 ⏸️ DEFERRED
+- [ ] **Problem:** Thinking/reasoning summaries not shown while streaming
+- [ ] **Complexity:** This is a significant refactor - leaving for later
+
+**OpenRouter Streaming Info:**
+- Uses `stream=True` with OpenAI SDK
+- Reasoning comes in `choices[].delta.reasoning_details` chunks
+- Types: `reasoning.summary`, `reasoning.text`, `reasoning.encrypted`
+- Tool call arguments stream as partial JSON (need accumulation)
+- Items paradigm: same ID emitted multiple times with updated content
+
+**Key Challenges:**
+- Tool call JSON accumulation (partial `{"query": "wea` → `{"query": "weather"}`)
+- Multiple concurrent outputs (thinking + tool calls + text simultaneously)
+- State management for partial responses
+- Error handling if connection drops mid-stream
+- Deciding when tool calls are "complete" enough to execute
+
+**UX Questions to Resolve:**
+- Show raw thinking text or summarized?
+- Live expanding text vs. spinner replacement?
+- Markdown rendering while streaming?
+- How to handle thinking + tool call display simultaneously?
+
+**Implementation Options:**
+- New `run_conversation_streaming()` method (keep non-streaming as fallback)
+- Wrapper that handles streaming internally
+- Big refactor of existing `run_conversation()`
+
+**References:**
+- https://openrouter.ai/docs/api/reference/streaming
+- https://openrouter.ai/docs/guides/best-practices/reasoning-tokens#streaming-response
+
+---
+
+## 1. Subagent Architecture (Context Isolation) 🎯
+
+**Problem:** Long-running tools (terminal commands, browser automation, complex file operations) consume massive context. A single `ls -la` can add hundreds of lines. Browser snapshots, debugging sessions, and iterative terminal work quickly bloat the main conversation, leaving less room for actual reasoning.
+
+**Solution:** The main agent becomes an **orchestrator** that delegates context-heavy tasks to **subagents**.
+
+**Architecture:**
+```
+┌─────────────────────────────────────────────────────────────────┐
+│  ORCHESTRATOR (main agent)                                      │
+│  - Receives user request                                        │
+│  - Plans approach                                               │
+│  - Delegates heavy tasks to subagents                           │
+│  - Receives summarized results                                  │
+│  - Maintains clean, focused context                             │
+└─────────────────────────────────────────────────────────────────┘
+         │                    │                    │
+         ▼                    ▼                    ▼
+┌─────────────────┐  ┌─────────────────┐  ┌─────────────────┐
+│ TERMINAL AGENT  │  │ BROWSER AGENT   │  │ CODE AGENT      │
+│ - terminal tool │  │ - browser tools │  │ - file tools    │
+│ - file tools    │  │ - web_search    │  │ - terminal      │
+│                 │  │ - web_extract   │  │                 │
+│ Isolated context│  │ Isolated context│  │ Isolated context│
+│ Returns summary │  │ Returns summary │  │ Returns summary │
+└─────────────────┘  └─────────────────┘  └─────────────────┘
+```
+
+**How it works:**
+1. User asks: "Set up a new Python project with FastAPI and tests"
+2. Orchestrator plans: "I need to create files, install deps, write code"
+3. Orchestrator calls: `terminal_task(goal="Create venv, install fastapi pytest", context="New project in ~/myapp")`
+4. **Subagent spawns** with fresh context, only terminal/file tools
+5. Subagent iterates (may take 10+ tool calls, lots of output)
+6. Subagent completes → returns summary: "Created venv, installed fastapi==0.109.0, pytest==8.0.0"
+7. Orchestrator receives **only the summary**, context stays clean
+8. Orchestrator continues with next subtask
+
+**Key tools to implement:**
+- [ ] `terminal_task(goal, context, cwd?)` - Delegate terminal/shell work
+- [ ] `browser_task(goal, context, start_url?)` - Delegate web research/automation  
+- [ ] `code_task(goal, context, files?)` - Delegate code writing/modification
+- [ ] Generic `delegate_task(goal, context, toolsets=[])` - Flexible delegation
+
+**Implementation details:**
+- [ ] Subagent uses same `run_agent.py` but with:
+  - Fresh/empty conversation history
+  - Limited toolset (only what's needed)
+  - Smaller max_iterations (focused task)
+  - Task-specific system prompt
+- [ ] Subagent returns structured result:
+  ```python
+  {
+    "success": True,
+    "summary": "Installed 3 packages, created 2 files",
+    "details": "Optional longer explanation if needed",
+    "artifacts": ["~/myapp/requirements.txt", "~/myapp/main.py"],  # Files created
+    "errors": []  # Any issues encountered
+  }
+  ```
+- [ ] Orchestrator sees only the summary in its context
+- [ ] Full subagent transcript saved separately for debugging
+
+**Benefits:**
+- 🧹 **Clean context** - Orchestrator stays focused, doesn't drown in tool output
+- 📊 **Better token efficiency** - 50 terminal outputs → 1 summary paragraph
+- 🎯 **Focused subagents** - Each agent has just the tools it needs
+- 🔄 **Parallel potential** - Independent subtasks could run concurrently
+- 🐛 **Easier debugging** - Each subtask has its own isolated transcript
+
+**When to use subagents vs direct tools:**
+- **Subagent**: Multi-step tasks, iteration likely, lots of output expected
+- **Direct**: Quick one-off commands, simple file reads, user needs to see output
+
+**Files to modify:** `run_agent.py` (add orchestration mode), new `tools/delegate_tools.py`, new `subagent_runner.py`
+
+---
+
+## 2. Context Management (complements Subagents)
+
+**Problem:** Context grows unbounded during long conversations. Trajectory compression exists for training data post-hoc, but live conversations lack intelligent context management.
+
+**Ideas:**
+- [ ] **Incremental summarization** - Compress old tool outputs on-the-fly during conversations
+  - Trigger when context exceeds threshold (e.g., 80% of max tokens)
+  - Preserve recent turns fully, summarize older tool responses
+  - Could reuse logic from `trajectory_compressor.py`
+  
+- [ ] **Semantic memory retrieval** - Vector store for long conversation recall
+  - Embed important facts/findings as conversation progresses
+  - Retrieve relevant memories when needed instead of keeping everything in context
+  - Consider lightweight solutions: ChromaDB, FAISS, or even a simple embedding cache
+  
+- [ ] **Working vs. episodic memory** distinction
+  - Working memory: Current task state, recent tool results (always in context)
+  - Episodic memory: Past findings, tried approaches (retrieved on demand)
+  - Clear eviction policies for each
+
+**Files to modify:** `run_agent.py` (add memory manager), possibly new `tools/memory_tool.py`
+
+---
+
+## 3. Self-Reflection & Course Correction 🔄
+
+**Problem:** Current retry logic handles malformed outputs but not semantic failures. Agent doesn't reason about *why* something failed.
+
+**Ideas:**
+- [ ] **Meta-reasoning after failures** - When a tool returns an error or unexpected result:
+  ```
+  Tool failed → Reflect: "Why did this fail? What assumptions were wrong?"
+  → Adjust approach → Retry with new strategy
+  ```
+  - Could be a lightweight LLM call or structured self-prompt
+  
+- [ ] **Planning/replanning module** - For complex multi-step tasks:
+  - Generate plan before execution
+  - After each step, evaluate: "Am I on track? Should I revise the plan?"
+  - Store plan in working memory, update as needed
+  
+- [ ] **Approach memory** - Remember what didn't work:
+  - "I tried X for this type of problem and it failed because Y"
+  - Prevents repeating failed strategies in the same conversation
+
+**Files to modify:** `run_agent.py` (add reflection hooks in tool loop), new `tools/reflection_tool.py`
+
+---
+
+## 4. Tool Composition & Learning 🔧
+
+**Problem:** Tools are atomic. Complex tasks require repeated manual orchestration of the same tool sequences.
+
+**Ideas:**
+- [ ] **Macro tools / Tool chains** - Define reusable tool sequences:
+  ```yaml
+  research_topic:
+    description: "Deep research on a topic"
+    steps:
+      - web_search: {query: "$topic"}
+      - web_extract: {urls: "$search_results.urls[:3]"}
+      - summarize: {content: "$extracted"}
+  ```
+  - Could be defined in skills or a new `macros/` directory
+  - Agent can invoke macro as single tool call
+  
+- [ ] **Tool failure patterns** - Learn from failures:
+  - Track: tool, input pattern, error type, what worked instead
+  - Before calling a tool, check: "Has this pattern failed before?"
+  - Persistent across sessions (stored in skills or separate DB)
+  
+- [ ] **Parallel tool execution** - When tools are independent, run concurrently:
+  - Detect independence (no data dependencies between calls)
+  - Use `asyncio.gather()` for parallel execution
+  - Already have async support in some tools, just need orchestration
+
+**Files to modify:** `model_tools.py`, `toolsets.py`, new `tool_macros.py`
+
+---
+
+## 5. Dynamic Skills Expansion 📚
+
+**Problem:** Skills system is elegant but static. Skills must be manually created and added.
+
+**Ideas:**
+- [ ] **Skill acquisition from successful tasks** - After completing a complex task:
+  - "This approach worked well. Save as a skill?"
+  - Extract: goal, steps taken, tools used, key decisions
+  - Generate SKILL.md automatically
+  - Store in user's skills directory
+  
+- [ ] **Skill templates** - Common patterns that can be parameterized:
+  ```markdown
+  # Debug {language} Error
+  1. Reproduce the error
+  2. Search for error message: `web_search("{error_message} {language}")`
+  3. Check common causes: {common_causes}
+  4. Apply fix and verify
+  ```
+  
+- [ ] **Skill chaining** - Combine skills for complex workflows:
+  - Skills can reference other skills as dependencies
+  - "To do X, first apply skill Y, then skill Z"
+  - Directed graph of skill dependencies
+
+**Files to modify:** `tools/skills_tool.py`, `skills/` directory structure, new `skill_generator.py`
+
+---
+
+## 6. Task Continuation Hints 🎯
+
+**Problem:** Could be more helpful by suggesting logical next steps.
+
+**Ideas:**
+- [ ] **Suggest next steps** - At end of a task, suggest logical continuations:
+  - "Code is written. Want me to also write tests / docs / deploy?"
+  - Based on common workflows for task type
+  - Non-intrusive, just offer options
+
+**Files to modify:** `run_agent.py`, response generation logic
+
+---
+
+## 7. Interactive Clarifying Questions Tool ❓
+
+**Problem:** Agent sometimes makes assumptions or guesses when it should ask the user. Currently can only ask via text, which gets lost in long outputs.
+
+**Ideas:**
+- [ ] **Multiple-choice prompt tool** - Let agent present structured choices to user:
+  ```
+  ask_user_choice(
+    question="Should the language switcher enable only German or all languages?",
+    choices=[
+      "Only enable German - works immediately",
+      "Enable all, mark untranslated - show fallback notice",
+      "Let me specify something else"
+    ]
+  )
+  ```
+  - Renders as interactive terminal UI with arrow key / Tab navigation
+  - User selects option, result returned to agent
+  - Up to 4 choices + optional free-text option
+  
+- [ ] **Implementation:**
+  - Use `inquirer` or `questionary` Python library for rich terminal prompts
+  - Tool returns selected option text (or user's custom input)
+  - **CLI-only** - only works when running via `cli.py` (not API/programmatic use)
+  - Graceful fallback: if not in interactive mode, return error asking agent to rephrase as text
+  
+- [ ] **Use cases:**
+  - Clarify ambiguous requirements before starting work
+  - Confirm destructive operations with clear options
+  - Let user choose between implementation approaches
+  - Checkpoint complex multi-step workflows
+
+**Files to modify:** New `tools/ask_user_tool.py`, `cli.py` (detect interactive mode), `model_tools.py`
+
+---
+
+## 8. Resource Awareness & Efficiency 💰
+
+**Problem:** No awareness of costs, time, or resource usage. Could be smarter about efficiency.
+
+**Ideas:**
+- [ ] **Tool result caching** - Don't repeat identical operations:
+  - Cache web searches, extractions within a session
+  - Invalidation based on time-sensitivity of query
+  - Hash-based lookup: same input → cached output
+
+- [ ] **Lazy evaluation** - Don't fetch everything upfront:
+  - Get summaries first, full content only if needed
+  - "I found 5 relevant pages. Want me to deep-dive on any?"
+
+**Files to modify:** `model_tools.py`, new `resource_tracker.py`
+
+---
+
+## 9. Collaborative Problem Solving 🤝
+
+**Problem:** Interaction is command/response. Complex problems benefit from dialogue.
+
+**Ideas:**
+- [ ] **Assumption surfacing** - Make implicit assumptions explicit:
+  - "I'm assuming you want Python 3.11+. Correct?"
+  - "This solution assumes you have sudo access..."
+  - Let user correct before going down wrong path
+
+- [ ] **Checkpoint & confirm** - For high-stakes operations:
+  - "About to delete 47 files. Here's the list - proceed?"
+  - "This will modify your database. Want a backup first?"
+  - Configurable threshold for when to ask
+
+**Files to modify:** `run_agent.py`, system prompt configuration
+
+---
+
+## 10. Project-Local Context 💾
+
+**Problem:** Valuable context lost between sessions.
+
+**Ideas:**
+- [ ] **Project awareness** - Remember project-specific context:
+  - Store `.hermes/context.md` in project directory
+  - "This is a Django project using PostgreSQL"
+  - Coding style preferences, deployment setup, etc.
+  - Load automatically when working in that directory
+
+- [ ] **Handoff notes** - Leave notes for future sessions:
+  - Write to `.hermes/notes.md` in project
+  - "TODO for next session: finish implementing X"
+  - "Known issues: Y doesn't work on Windows"
+
+**Files to modify:** New `project_context.py`, auto-load in `run_agent.py`
+
+---
+
+## 11. Graceful Degradation & Robustness 🛡️
+
+**Problem:** When things go wrong, recovery is limited. Should fail gracefully.
+
+**Ideas:**
+- [ ] **Fallback chains** - When primary approach fails, have backups:
+  - `web_extract` fails → try `browser_navigate` → try `web_search` for cached version
+  - Define fallback order per tool type
+  
+- [ ] **Partial progress preservation** - Don't lose work on failure:
+  - Long task fails midway → save what we've got
+  - "I completed 3/5 steps before the error. Here's what I have..."
+  
+- [ ] **Self-healing** - Detect and recover from bad states:
+  - Browser stuck → close and retry
+  - Terminal hung → timeout and reset
+
+**Files to modify:** `model_tools.py`, tool implementations, new `fallback_manager.py`
+
+---
+
+## 12. Tools & Skills Wishlist 🧰
+
+*Things that would need new tool implementations (can't do well with current tools):*
+
+### High-Impact
+
+- [ ] **Audio/Video Transcription** 🎬 *(See also: Section 16 for detailed spec)*
+  - Transcribe audio files, podcasts, YouTube videos
+  - Extract key moments from video
+  - Voice memo transcription for messaging integrations
+  - *Provider options: Whisper API, Deepgram, local Whisper*
+  
+- [ ] **Diagram Rendering** 📊
+  - Render Mermaid/PlantUML to actual images
+  - Can generate the code, but rendering requires external service or tool
+  - "Show me how these components connect" → actual visual diagram
+
+### Medium-Impact
+
+- [ ] **Canvas / Visual Workspace** 🖼️
+  - Agent-controlled visual panel for rendering interactive UI
+  - Inspired by OpenClaw's Canvas feature
+  - **Capabilities:**
+    - `present` / `hide` - Show/hide the canvas panel
+    - `navigate` - Load HTML files or URLs into the canvas
+    - `eval` - Execute JavaScript in the canvas context
+    - `snapshot` - Capture the rendered UI as an image
+  - **Use cases:**
+    - Display generated HTML/CSS/JS previews
+    - Show interactive data visualizations (charts, graphs)
+    - Render diagrams (Mermaid → rendered output)
+    - Present structured information in rich format
+    - A2UI-style component system for structured agent UI
+  - **Implementation options:**
+    - Electron-based panel for CLI
+    - WebSocket-connected web app
+    - VS Code webview extension
+  - *Would let agent "show" things rather than just describe them*
+
+- [ ] **Document Generation** 📄
+  - Create styled PDFs, Word docs, presentations
+  - *Can do basic PDF via terminal tools, but limited*
+
+- [ ] **Diff/Patch Tool** 📝
+  - Surgical code modifications with preview
+  - "Change line 45-50 to X" without rewriting whole file
+  - Show diffs before applying
+  - *Can use `diff`/`patch` but a native tool would be safer*
+
+### Skills to Create
+
+- [ ] **Domain-specific skill packs:**
+  - DevOps/Infrastructure (Terraform, K8s, AWS)
+  - Data Science workflows (EDA, model training)
+  - Security/pentesting procedures
+  
+- [ ] **Framework-specific skills:**
+  - React/Vue/Angular patterns
+  - Django/Rails/Express conventions
+  - Database optimization playbooks
+
+- [ ] **Troubleshooting flowcharts:**
+  - "Docker container won't start" → decision tree
+  - "Production is slow" → systematic diagnosis
+
+---
+
+## 13. Messaging Platform Integrations 💬
+
+**Problem:** Agent currently only works via `cli.py` which requires direct terminal access. Users may want to interact via messaging apps from their phone or other devices.
+
+**Architecture:**
+- `run_agent.py` already accepts `conversation_history` parameter and returns updated messages ✅
+- Need: persistent session storage, platform monitors, session key resolution
+
+**Implementation approach:**
+```
+┌─────────────────────────────────────────────────────────────┐
+│  Platform Monitor (e.g., telegram_monitor.py)               │
+│  ├─ Long-running daemon connecting to messaging platform    │
+│  ├─ On message: resolve session key → load history from disk│
+│  ├─ Call run_agent.py with loaded history                   │
+│  ├─ Save updated history back to disk (JSONL)               │
+│  └─ Send response back to platform                          │
+└─────────────────────────────────────────────────────────────┘
+```
+
+**Platform support (each user sets up their own credentials):**
+- [ ] **Telegram** - via `python-telegram-bot` or `grammy` equivalent
+  - Bot token from @BotFather
+  - Easiest to set up, good for personal use
+- [ ] **Discord** - via `discord.py`
+  - Bot token from Discord Developer Portal
+  - Can work in servers (group sessions) or DMs
+- [ ] **WhatsApp** - via `baileys` (WhatsApp Web protocol)
+  - QR code scan to authenticate
+  - More complex, but reaches most people
+
+**Session management:**
+- [ ] **Session store** - JSONL persistence per session key
+  - `~/.hermes/sessions/{session_key}.jsonl`
+  - Session keys: `telegram:dm:{user_id}`, `discord:channel:{id}`, etc.
+- [ ] **Session expiry** - Configurable reset policies
+  - Daily reset (default 4am) OR idle timeout (e.g., 2 hours)
+  - Manual reset via `/reset` or `/new` command in chat
+- [ ] **Session continuity** - Conversations persist across messages until reset
+
+**Files to create:** `monitors/telegram_monitor.py`, `monitors/discord_monitor.py`, `monitors/session_store.py`
+
+---
+
+## 14. Scheduled Tasks / Cron Jobs ⏰
+
+**Problem:** Agent only runs on-demand. Some tasks benefit from scheduled execution (daily summaries, monitoring, reminders).
+
+**Ideas:**
+- [ ] **Cron-style scheduler** - Run agent turns on a schedule
+  - Store jobs in `~/.hermes/cron/jobs.json`
+  - Each job: `{ id, schedule, prompt, session_mode, delivery }`
+  - Uses APScheduler or similar Python library
+  
+- [ ] **Session modes:**
+  - `isolated` - Fresh session each run (no history, clean context)
+  - `main` - Append to main session (agent remembers previous scheduled runs)
+  
+- [ ] **Delivery options:**
+  - Write output to file (`~/.hermes/cron/output/{job_id}/{timestamp}.md`)
+  - Send to messaging channel (if integrations enabled)
+  - Both
+  
+- [ ] **CLI interface:**
+  ```bash
+  # List scheduled jobs
+  python cli.py --cron list
+  
+  # Add a job (runs daily at 9am)
+  python cli.py --cron add "Summarize my email inbox" --schedule "0 9 * * *"
+  
+  # Quick syntax for simple intervals  
+  python cli.py --cron add "Check server status" --every 30m
+  
+  # Remove a job
+  python cli.py --cron remove <job_id>
+  ```
+
+- [ ] **Agent self-scheduling** - Let the agent create its own cron jobs
+  - New tool: `schedule_task(prompt, schedule, session_mode)`
+  - "Remind me to check the deployment tomorrow at 9am"
+  - Agent can set follow-up tasks for itself
+
+- [ ] **In-chat command:** `/cronjob {prompt} {frequency}` when using messaging integrations
+
+**Files to create:** `cron/scheduler.py`, `cron/jobs.py`, `tools/schedule_tool.py`
+
+---
+
+## 15. Text-to-Speech (TTS) 🔊
+
+**Problem:** Agent can only respond with text. Some users prefer audio responses (accessibility, hands-free use, podcasts).
+
+**Ideas:**
+- [ ] **TTS tool** - Generate audio files from text
+  ```python
+  tts_generate(text="Here's your summary...", voice="nova", output="summary.mp3")
+  ```
+  - Returns path to generated audio file
+  - For messaging integrations: can send as voice message
+  
+- [ ] **Provider options:**
+  - Edge TTS (free, good quality, many voices)
+  - OpenAI TTS (paid, excellent quality)
+  - ElevenLabs (paid, best quality, voice cloning)
+  - Local options (Coqui TTS, Bark)
+  
+- [ ] **Modes:**
+  - On-demand: User explicitly asks "read this to me"
+  - Auto-TTS: Configurable to always generate audio for responses
+  - Long-text handling: Summarize or chunk very long responses
+  
+- [ ] **Integration with messaging:**
+  - When enabled, can send voice notes instead of/alongside text
+  - User preference per channel
+
+**Files to create:** `tools/tts_tool.py`, config in `cli-config.yaml`
+
+---
+
+## 16. Speech-to-Text / Audio Transcription 🎤
+
+**Problem:** Users may want to send voice memos instead of typing. Agent is blind to audio content.
+
+**Ideas:**
+- [ ] **Voice memo transcription** - For messaging integrations
+  - User sends voice message → transcribe → process as text
+  - Seamless: user speaks, agent responds
+  
+- [ ] **Audio/video file transcription** - Existing idea, expanded:
+  - Transcribe local audio files (mp3, wav, m4a)
+  - Transcribe YouTube videos (download audio → transcribe)
+  - Extract key moments with timestamps
+  
+- [ ] **Provider options:**
+  - OpenAI Whisper API (good quality, cheap)
+  - Deepgram (fast, good for real-time)
+  - Local Whisper (free, runs on GPU)
+  - Groq Whisper (fast, free tier available)
+  
+- [ ] **Tool interface:**
+  ```python
+  transcribe(source="audio.mp3")  # Local file
+  transcribe(source="https://youtube.com/...")  # YouTube
+  transcribe(source="voice_message", data=bytes)  # Voice memo
+  ```
+
+**Files to create:** `tools/transcribe_tool.py`, integrate with messaging monitors
+
+---
+
+## Priority Order (Suggested)
+
+1. **🎯 Subagent Architecture** - Critical for context management, enables everything else
+2. **Memory & Context Management** - Complements subagents for remaining context
+3. **Self-Reflection** - Improves reliability and reduces wasted tool calls  
+4. **Project-Local Context** - Practical win, keeps useful info across sessions
+5. **Messaging Integrations** - Unlocks mobile access, new interaction patterns
+6. **Scheduled Tasks / Cron Jobs** - Enables automation, reminders, monitoring
+7. **Tool Composition** - Quality of life, builds on other improvements
+8. **Dynamic Skills** - Force multiplier for repeated tasks
+9. **Interactive Clarifying Questions** - Better UX for ambiguous tasks
+10. **TTS / Audio Transcription** - Accessibility, hands-free use
+
+---
+
+## Removed Items (Unrealistic)
+
+The following were removed because they're architecturally impossible:
+
+- ~~Proactive suggestions / Prefetching~~ - Agent only runs on user request, can't interject
+- ~~Clipboard integration~~ - No access to user's local system clipboard
+
+The following **moved to active TODO** (now possible with new architecture):
+
+- ~~Session save/restore~~ → See **Messaging Integrations** (session persistence)
+- ~~Voice/TTS playback~~ → See **TTS** (can generate audio files, send via messaging)
+- ~~Set reminders~~ → See **Scheduled Tasks / Cron Jobs**
+
+The following were removed because they're **already possible**:
+
+- ~~HTTP/API Client~~ → Use `curl` or Python `requests` in terminal
+- ~~Structured Data Manipulation~~ → Use `pandas` in terminal
+- ~~Git-Native Operations~~ → Use `git` CLI in terminal
+- ~~Symbolic Math~~ → Use `SymPy` in terminal
+- ~~Code Quality Tools~~ → Run linters (`eslint`, `black`, `mypy`) in terminal
+- ~~Testing Framework~~ → Run `pytest`, `jest`, etc. in terminal
+- ~~Translation~~ → LLM handles this fine, or use translation APIs
+
+---
+
+---
+
+## 🧪 Brainstorm Ideas (Not Yet Fleshed Out)
+
+*These are early-stage ideas that need more thinking before implementation. Captured here so they don't get lost.*
+
+### Remote/Distributed Execution 🌐
+
+**Concept:** Run agent on a powerful remote server while interacting from a thin client.
+
+**Why interesting:**
+- Run on beefy GPU server for local LLM inference
+- Agent has access to remote machine's resources (files, tools, internet)
+- User interacts via lightweight client (phone, low-power laptop)
+
+**Open questions:**
+- How does this differ from just SSH + running cli.py on remote?
+- Would need secure communication channel (WebSocket? gRPC?)
+- How to handle tool outputs that reference remote paths?
+- Credential management for remote execution
+- Latency considerations for interactive use
+
+**Possible architecture:**
+```
+┌─────────────┐         ┌─────────────────────────┐
+│ Thin Client │ ◄─────► │ Remote Hermes Server    │
+│ (phone/web) │  WS/API │ - Full agent + tools    │
+└─────────────┘         │ - GPU for local LLM     │
+                        │ - Access to server files│
+                        └─────────────────────────┘
+```
+
+**Related to:** Messaging integrations (could be the "server" that monitors receive from)
+
+---
+
+### Multi-Agent Parallel Execution 🤖🤖
+
+**Concept:** Extension of Subagent Architecture (Section 1) - run multiple subagents in parallel.
+
+**Why interesting:**
+- Independent subtasks don't need to wait for each other
+- "Research X while setting up Y" - both run simultaneously
+- Faster completion for complex multi-part tasks
+
+**Open questions:**
+- How to detect which tasks are truly independent?
+- Resource management (API rate limits, concurrent connections)
+- How to merge results when parallel tasks have conflicts?
+- Cost implications of multiple parallel LLM calls
+
+*Note: Basic subagent delegation (Section 1) should be implemented first, parallel execution is an optimization on top.*
+
+---
+
+### Plugin/Extension System 🔌
+
+**Concept:** Allow users to add custom tools/skills without modifying core code.
+
+**Why interesting:**
+- Community contributions
+- Organization-specific tools
+- Clean separation of core vs. extensions
+
+**Open questions:**
+- Security implications of loading arbitrary code
+- Versioning and compatibility
+- Discovery and installation UX
+
+---
+
+*Last updated: $(date +%Y-%m-%d)* 🤖

acp_adapter/__init__.py

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

acp_adapter/__main__.py

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

acp_adapter/auth.py

@@ -1,24 +0,0 @@
-"""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

@@ -1,85 +0,0 @@
-"""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
-
-
-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 = Path(os.getenv("HERMES_HOME", Path.home() / ".hermes"))
-    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))
-    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

@@ -1,171 +0,0 @@
-"""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 defaultdict, 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

@@ -1,80 +0,0 @@
-"""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 Any, Callable, Optional
-
-from acp.schema import (
-    AllowedOutcome,
-    DeniedOutcome,
-    PermissionOption,
-    RequestPermissionRequest,
-    SelectedPermissionOutcome,
-)
-
-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

@@ -1,479 +0,0 @@
-"""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,
-    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,
-        client_capabilities: ClientCapabilities | None = None,
-        client_info: Implementation | None = None,
-        **kwargs: Any,
-    ) -> InitializeResponse:
-        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, 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"]
-
-        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
-
-        # Auto-detect provider for the requested model
-        try:
-            from hermes_cli.models import parse_model_input, detect_provider_for_model
-            current_provider = getattr(state.agent, "provider", None) or "openrouter"
-            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,
-        )
-        provider_label = target_provider or getattr(state.agent, "provider", "auto")
-        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()
-        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)
-                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
-    ):
-        """Switch the model for a session (called by ACP protocol)."""
-        state = self.session_manager.get_session(session_id)
-        if state:
-            state.model = model_id
-            state.agent = self.session_manager._make_agent(
-                session_id=session_id,
-                cwd=state.cwd,
-                model=model_id,
-            )
-            logger.info("Session %s: model switched to %s", session_id, model_id)
-        return None

acp_adapter/session.py

@@ -1,203 +0,0 @@
-"""ACP session manager — maps ACP sessions to Hermes AIAgent instances."""
-from __future__ import annotations
-
-import copy
-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."""
-
-    def __init__(self, agent_factory=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.
-        """
-        self._sessions: Dict[str, SessionState] = {}
-        self._lock = Lock()
-        self._agent_factory = agent_factory
-
-    # ---- 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)
-        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``."""
-        with self._lock:
-            return self._sessions.get(session_id)
-
-    def remove_session(self, session_id: str) -> bool:
-        """Remove a session. Returns True if it existed."""
-        with self._lock:
-            existed = self._sessions.pop(session_id, None) is not None
-        if existed:
-            _clear_task_cwd(session_id)
-        return existed
-
-    def fork_session(self, session_id: str, cwd: str = ".") -> Optional[SessionState]:
-        """Deep-copy a session's history into a new session."""
-        import threading
-
-        with self._lock:
-            original = self._sessions.get(session_id)
-            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(),
-            )
-            self._sessions[new_id] = state
-        _register_task_cwd(new_id, cwd)
-        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."""
-        with self._lock:
-            return [
-                {
-                    "session_id": s.session_id,
-                    "cwd": s.cwd,
-                    "model": s.model,
-                    "history_len": len(s.history),
-                }
-                for s in self._sessions.values()
-            ]
-
-    def update_cwd(self, session_id: str, cwd: str) -> Optional[SessionState]:
-        """Update the working directory for a session and its tool overrides."""
-        with self._lock:
-            state = self._sessions.get(session_id)
-            if state is None:
-                return None
-            state.cwd = cwd
-        _register_task_cwd(session_id, cwd)
-        return state
-
-    def cleanup(self) -> None:
-        """Remove all sessions 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)
-
-    # ---- internal -----------------------------------------------------------
-
-    def _make_agent(
-        self,
-        *,
-        session_id: str,
-        cwd: str,
-        model: 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"
-        requested_provider = None
-        if isinstance(model_cfg, dict):
-            default_model = str(model_cfg.get("default") or default_model)
-            requested_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)
-            kwargs.update(
-                {
-                    "provider": runtime.get("provider"),
-                    "api_mode": runtime.get("api_mode"),
-                    "base_url": runtime.get("base_url"),
-                    "api_key": runtime.get("api_key"),
-                }
-            )
-        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

@@ -1,215 +0,0 @@
-"""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

@@ -1,12 +0,0 @@
-{
-  "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

@@ -1,25 +0,0 @@
-<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

@@ -1,6 +0,0 @@
-"""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

@@ -1,383 +0,0 @@
-"""Automatic context window compression for long conversations.
-
-Self-contained class with its own OpenAI client for summarization.
-Uses Gemini Flash (cheap/fast) to summarize middle turns while
-protecting head and tail context.
-"""
-
-import logging
-import os
-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]:"
-
-
-class ContextCompressor:
-    """Compresses conversation context when approaching the model's context limit.
-
-    Algorithm: protect first N + last N turns, summarize everything in between.
-    Token tracking uses actual counts from API responses for accuracy.
-    """
-
-    def __init__(
-        self,
-        model: str,
-        threshold_percent: float = 0.50,
-        protect_first_n: int = 3,
-        protect_last_n: int = 4,
-        summary_target_tokens: int = 2500,
-        quiet_mode: bool = False,
-        summary_model_override: str = None,
-        base_url: str = "",
-        api_key: str = "",
-    ):
-        self.model = model
-        self.base_url = base_url
-        self.api_key = api_key
-        self.threshold_percent = threshold_percent
-        self.protect_first_n = protect_first_n
-        self.protect_last_n = protect_last_n
-        self.summary_target_tokens = summary_target_tokens
-        self.quiet_mode = quiet_mode
-
-        self.context_length = get_model_context_length(model, base_url=base_url, api_key=api_key)
-        self.threshold_tokens = int(self.context_length * threshold_percent)
-        self.compression_count = 0
-        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 ""
-
-    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": (self.last_prompt_tokens / self.context_length * 100) if self.context_length else 0,
-            "compression_count": self.compression_count,
-        }
-
-    def _generate_summary(self, turns_to_summarize: List[Dict[str, Any]]) -> Optional[str]:
-        """Generate a concise summary of conversation turns.
-
-        Tries the auxiliary model first, then falls back to the user's main
-        model.  Returns None if all attempts fail — the caller should drop
-        the middle turns without a summary rather than inject a useless
-        placeholder.
-        """
-        parts = []
-        for msg in turns_to_summarize:
-            role = msg.get("role", "unknown")
-            content = msg.get("content") or ""
-            if len(content) > 2000:
-                content = content[:1000] + "\n...[truncated]...\n" + content[-500:]
-            tool_calls = msg.get("tool_calls", [])
-            if tool_calls:
-                tool_names = [tc.get("function", {}).get("name", "?") for tc in tool_calls if isinstance(tc, dict)]
-                content += f"\n[Tool calls: {', '.join(tool_names)}]"
-            parts.append(f"[{role.upper()}]: {content}")
-
-        content_to_summarize = "\n\n".join(parts)
-        prompt = f"""Create a concise handoff summary for a later assistant that will continue this conversation after earlier turns are compacted.
-
-Describe:
-1. What actions were taken (tool calls, searches, file operations)
-2. Key information or results obtained
-3. Important decisions, constraints, or user preferences
-4. Relevant data, file names, outputs, or next steps needed to continue
-
-Keep it factual, concise, and focused on helping the next assistant resume without repeating work. Target ~{self.summary_target_tokens} tokens.
-
----
-TURNS TO SUMMARIZE:
-{content_to_summarize}
----
-
-Write only the summary body. Do not include any preamble or prefix; the system will add the handoff wrapper."""
-
-        # Use the centralized LLM router — handles provider resolution,
-        # auth, and fallback internally.
-        try:
-            call_kwargs = {
-                "task": "compression",
-                "messages": [{"role": "user", "content": prompt}],
-                "temperature": 0.3,
-                "max_tokens": self.summary_target_tokens * 2,
-                "timeout": 30.0,
-            }
-            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()
-            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 message just before ``idx`` is an assistant message with
-        tool_calls, those tool results will start at ``idx`` and would be
-        separated from their parent.  Move backwards to include the whole
-        group in the summarised region.
-        """
-        if idx <= 0 or idx >= len(messages):
-            return idx
-        prev = messages[idx - 1]
-        if prev.get("role") == "assistant" and prev.get("tool_calls"):
-            # The results for this assistant turn sit at idx..idx+k.
-            # Include the assistant message in the summarised region too.
-            idx -= 1
-        return idx
-
-    def compress(self, messages: List[Dict[str, Any]], current_tokens: int = None) -> List[Dict[str, Any]]:
-        """Compress conversation messages by summarizing middle turns.
-
-        Keeps first N + last N turns, summarizes everything in between.
-        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
-
-        compress_start = self.protect_first_n
-        compress_end = n_messages - self.protect_last_n
-        if compress_start >= compress_end:
-            return messages
-
-        # Adjust boundaries to avoid splitting tool_call/result groups.
-        compress_start = self._align_boundary_forward(messages, compress_start)
-        compress_end = self._align_boundary_backward(messages, compress_end)
-        if compress_start >= compress_end:
-            return messages
-
-        turns_to_summarize = messages[compress_start:compress_end]
-        display_tokens = current_tokens if current_tokens else self.last_prompt_tokens or estimate_messages_tokens_rough(messages)
-
-        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,
-            )
-            logger.info(
-                "Summarizing turns %d-%d (%d turns)",
-                compress_start + 1,
-                compress_end,
-                len(turns_to_summarize),
-            )
-
-        summary = self._generate_summary(turns_to_summarize)
-
-        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/display.py

@@ -1,614 +0,0 @@
-"""CLI presentation -- spinner, kawaii faces, tool preview formatting.
-
-Pure display functions and classes with no AIAgent dependency.
-Used by AIAgent._execute_tool_calls for CLI feedback.
-"""
-
-import json
-import logging
-import os
-import sys
-import threading
-import time
-
-# ANSI escape codes for coloring tool failure indicators
-_RED = "\033[31m"
-_RESET = "\033[0m"
-
-logger = logging.getLogger(__name__)
-
-
-# =========================================================================
-# Skin-aware helpers (lazy import to avoid circular deps)
-# =========================================================================
-
-def _get_skin():
-    """Get the active skin config, or None if not available."""
-    try:
-        from hermes_cli.skin_engine import get_active_skin
-        return get_active_skin()
-    except Exception:
-        return None
-
-
-def get_skin_faces(key: str, default: list) -> list:
-    """Get spinner face list from active skin, falling back to default."""
-    skin = _get_skin()
-    if skin:
-        faces = skin.get_spinner_list(key)
-        if faces:
-            return faces
-    return default
-
-
-def get_skin_verbs() -> list:
-    """Get thinking verbs from active skin."""
-    skin = _get_skin()
-    if skin:
-        verbs = skin.get_spinner_list("thinking_verbs")
-        if verbs:
-            return verbs
-    return KawaiiSpinner.THINKING_VERBS
-
-
-def get_skin_tool_prefix() -> str:
-    """Get tool output prefix character from active skin."""
-    skin = _get_skin()
-    if skin:
-        return skin.tool_prefix
-    return "┊"
-
-
-def get_tool_emoji(tool_name: str, default: str = "⚡") -> str:
-    """Get the display emoji for a tool.
-
-    Resolution order:
-    1. Active skin's ``tool_emojis`` overrides (if a skin is loaded)
-    2. Tool registry's per-tool ``emoji`` field
-    3. *default* fallback
-    """
-    # 1. Skin override
-    skin = _get_skin()
-    if skin and skin.tool_emojis:
-        override = skin.tool_emojis.get(tool_name)
-        if override:
-            return override
-    # 2. Registry default
-    try:
-        from tools.registry import registry
-        emoji = registry.get_emoji(tool_name, default="")
-        if emoji:
-            return emoji
-    except Exception:
-        pass
-    # 3. Hardcoded fallback
-    return default
-
-
-# =========================================================================
-# Tool preview (one-line summary of a tool call's primary argument)
-# =========================================================================
-
-def _oneline(text: str) -> str:
-    """Collapse whitespace (including newlines) to single spaces."""
-    return " ".join(text.split())
-
-
-def build_tool_preview(tool_name: str, args: dict, max_len: int = 40) -> str | None:
-    """Build a short preview of a tool call's primary argument for display."""
-    if not args:
-        return None
-    primary_args = {
-        "terminal": "command", "web_search": "query", "web_extract": "urls",
-        "read_file": "path", "write_file": "path", "patch": "path",
-        "search_files": "pattern", "browser_navigate": "url",
-        "browser_click": "ref", "browser_type": "text",
-        "image_generate": "prompt", "text_to_speech": "text",
-        "vision_analyze": "question", "mixture_of_agents": "user_prompt",
-        "skill_view": "name", "skills_list": "category",
-        "cronjob": "action",
-        "execute_code": "code", "delegate_task": "goal",
-        "clarify": "question", "skill_manage": "name",
-    }
-
-    if tool_name == "process":
-        action = args.get("action", "")
-        sid = args.get("session_id", "")
-        data = args.get("data", "")
-        timeout_val = args.get("timeout")
-        parts = [action]
-        if sid:
-            parts.append(sid[:16])
-        if data:
-            parts.append(f'"{_oneline(data[:20])}"')
-        if timeout_val and action == "wait":
-            parts.append(f"{timeout_val}s")
-        return " ".join(parts) if parts else None
-
-    if tool_name == "todo":
-        todos_arg = args.get("todos")
-        merge = args.get("merge", False)
-        if todos_arg is None:
-            return "reading task list"
-        elif merge:
-            return f"updating {len(todos_arg)} task(s)"
-        else:
-            return f"planning {len(todos_arg)} task(s)"
-
-    if tool_name == "session_search":
-        query = _oneline(args.get("query", ""))
-        return f"recall: \"{query[:25]}{'...' if len(query) > 25 else ''}\""
-
-    if tool_name == "memory":
-        action = args.get("action", "")
-        target = args.get("target", "")
-        if action == "add":
-            content = _oneline(args.get("content", ""))
-            return f"+{target}: \"{content[:25]}{'...' if len(content) > 25 else ''}\""
-        elif action == "replace":
-            return f"~{target}: \"{_oneline(args.get('old_text', '')[:20])}\""
-        elif action == "remove":
-            return f"-{target}: \"{_oneline(args.get('old_text', '')[:20])}\""
-        return action
-
-    if tool_name == "send_message":
-        target = args.get("target", "?")
-        msg = _oneline(args.get("message", ""))
-        if len(msg) > 20:
-            msg = msg[:17] + "..."
-        return f"to {target}: \"{msg}\""
-
-    if tool_name.startswith("rl_"):
-        rl_previews = {
-            "rl_list_environments": "listing envs",
-            "rl_select_environment": args.get("name", ""),
-            "rl_get_current_config": "reading config",
-            "rl_edit_config": f"{args.get('field', '')}={args.get('value', '')}",
-            "rl_start_training": "starting",
-            "rl_check_status": args.get("run_id", "")[:16],
-            "rl_stop_training": f"stopping {args.get('run_id', '')[:16]}",
-            "rl_get_results": args.get("run_id", "")[:16],
-            "rl_list_runs": "listing runs",
-            "rl_test_inference": f"{args.get('num_steps', 3)} steps",
-        }
-        return rl_previews.get(tool_name)
-
-    key = primary_args.get(tool_name)
-    if not key:
-        for fallback_key in ("query", "text", "command", "path", "name", "prompt", "code", "goal"):
-            if fallback_key in args:
-                key = fallback_key
-                break
-
-    if not key or key not in args:
-        return None
-
-    value = args[key]
-    if isinstance(value, list):
-        value = value[0] if value else ""
-
-    preview = _oneline(str(value))
-    if not preview:
-        return None
-    if len(preview) > max_len:
-        preview = preview[:max_len - 3] + "..."
-    return preview
-
-
-# =========================================================================
-# KawaiiSpinner
-# =========================================================================
-
-class KawaiiSpinner:
-    """Animated spinner with kawaii faces for CLI feedback during tool execution."""
-
-    SPINNERS = {
-        'dots': ['⠋', '⠙', '⠹', '⠸', '⠼', '⠴', '⠦', '⠧', '⠇', '⠏'],
-        'bounce': ['⠁', '⠂', '⠄', '⡀', '⢀', '⠠', '⠐', '⠈'],
-        'grow': ['▁', '▂', '▃', '▄', '▅', '▆', '▇', '█', '▇', '▆', '▅', '▄', '▃', '▂'],
-        'arrows': ['←', '↖', '↑', '↗', '→', '↘', '↓', '↙'],
-        'star': ['✶', '✷', '✸', '✹', '✺', '✹', '✸', '✷'],
-        'moon': ['🌑', '🌒', '🌓', '🌔', '🌕', '🌖', '🌗', '🌘'],
-        'pulse': ['◜', '◠', '◝', '◞', '◡', '◟'],
-        'brain': ['🧠', '💭', '💡', '✨', '💫', '🌟', '💡', '💭'],
-        'sparkle': ['⁺', '˚', '*', '✧', '✦', '✧', '*', '˚'],
-    }
-
-    KAWAII_WAITING = [
-        "(｡◕‿◕｡)", "(◕‿◕✿)", "٩(◕‿◕｡)۶", "(✿◠‿◠)", "( ˘▽˘)っ",
-        "♪(´ε` )", "(◕ᴗ◕✿)", "ヾ(＾∇＾)", "(≧◡≦)", "(★ω★)",
-    ]
-
-    KAWAII_THINKING = [
-        "(｡•́︿•̀｡)", "(◔_◔)", "(¬‿¬)", "( •_•)>⌐■-■", "(⌐■_■)",
-        "(´･_･`)", "◉_◉", "(°ロ°)", "( ˘⌣˘)♡", "ヽ(>∀<☆)☆",
-        "٩(๑❛ᴗ❛๑)۶", "(⊙_⊙)", "(¬_¬)", "( ͡° ͜ʖ ͡°)", "ಠ_ಠ",
-    ]
-
-    THINKING_VERBS = [
-        "pondering", "contemplating", "musing", "cogitating", "ruminating",
-        "deliberating", "mulling", "reflecting", "processing", "reasoning",
-        "analyzing", "computing", "synthesizing", "formulating", "brainstorming",
-    ]
-
-    def __init__(self, message: str = "", spinner_type: str = 'dots'):
-        self.message = message
-        self.spinner_frames = self.SPINNERS.get(spinner_type, self.SPINNERS['dots'])
-        self.running = False
-        self.thread = None
-        self.frame_idx = 0
-        self.start_time = None
-        self.last_line_len = 0
-        self._last_flush_time = 0.0  # Rate-limit flushes for patch_stdout compat
-        # Capture stdout NOW, before any redirect_stdout(devnull) from
-        # child agents can replace sys.stdout with a black hole.
-        self._out = sys.stdout
-
-    def _write(self, text: str, end: str = '\n', flush: bool = False):
-        """Write to the stdout captured at spinner creation time."""
-        try:
-            self._out.write(text + end)
-            if flush:
-                self._out.flush()
-        except (ValueError, OSError):
-            pass
-
-    def _animate(self):
-        # Cache skin wings at start (avoid per-frame imports)
-        skin = _get_skin()
-        wings = skin.get_spinner_wings() if skin else []
-
-        while self.running:
-            if os.getenv("HERMES_SPINNER_PAUSE"):
-                time.sleep(0.1)
-                continue
-            frame = self.spinner_frames[self.frame_idx % len(self.spinner_frames)]
-            elapsed = time.time() - self.start_time
-            if wings:
-                left, right = wings[self.frame_idx % len(wings)]
-                line = f"  {left} {frame} {self.message} {right} ({elapsed:.1f}s)"
-            else:
-                line = f"  {frame} {self.message} ({elapsed:.1f}s)"
-            pad = max(self.last_line_len - len(line), 0)
-            # Rate-limit flush() calls to avoid spinner spam under
-            # prompt_toolkit's patch_stdout.  Each flush() pushes a queue
-            # item that may trigger a separate run_in_terminal() call; if
-            # items are processed one-at-a-time the \r overwrite is lost
-            # and every frame appears on its own line.  By flushing at
-            # most every 0.4s we guarantee multiple \r-frames are batched
-            # into a single write, so the terminal collapses them correctly.
-            now = time.time()
-            should_flush = (now - self._last_flush_time) >= 0.4
-            self._write(f"\r{line}{' ' * pad}", end='', flush=should_flush)
-            if should_flush:
-                self._last_flush_time = now
-            self.last_line_len = len(line)
-            self.frame_idx += 1
-            time.sleep(0.12)
-
-    def start(self):
-        if self.running:
-            return
-        self.running = True
-        self.start_time = time.time()
-        self.thread = threading.Thread(target=self._animate, daemon=True)
-        self.thread.start()
-
-    def update_text(self, new_message: str):
-        self.message = new_message
-
-    def print_above(self, text: str):
-        """Print a line above the spinner without disrupting animation.
-
-        Clears the current spinner line, prints the text, and lets the
-        next animation tick redraw the spinner on the line below.
-        Thread-safe: uses the captured stdout reference (self._out).
-        Works inside redirect_stdout(devnull) because _write bypasses
-        sys.stdout and writes to the stdout captured at spinner creation.
-        """
-        if not self.running:
-            self._write(f"  {text}", flush=True)
-            return
-        # Clear spinner line with spaces (not \033[K) to avoid garbled escape
-        # codes when prompt_toolkit's patch_stdout is active — same approach
-        # as stop(). Then print text; spinner redraws on next tick.
-        blanks = ' ' * max(self.last_line_len + 5, 40)
-        self._write(f"\r{blanks}\r  {text}", flush=True)
-
-    def stop(self, final_message: str = None):
-        self.running = False
-        if self.thread:
-            self.thread.join(timeout=0.5)
-        # Clear the spinner line with spaces instead of \033[K to avoid
-        # garbled escape codes when prompt_toolkit's patch_stdout is active.
-        blanks = ' ' * max(self.last_line_len + 5, 40)
-        self._write(f"\r{blanks}\r", end='', flush=True)
-        if final_message:
-            self._write(f"  {final_message}", flush=True)
-
-    def __enter__(self):
-        self.start()
-        return self
-
-    def __exit__(self, exc_type, exc_val, exc_tb):
-        self.stop()
-        return False
-
-
-# =========================================================================
-# Kawaii face arrays (used by AIAgent._execute_tool_calls for spinner text)
-# =========================================================================
-
-KAWAII_SEARCH = [
-    "♪(´ε` )", "(｡◕‿◕｡)", "ヾ(＾∇＾)", "(◕ᴗ◕✿)", "( ˘▽˘)っ",
-    "٩(◕‿◕｡)۶", "(✿◠‿◠)", "♪～(´ε｀ )", "(ノ´ヮ`)ノ*:・゚✧", "＼(◎o◎)／",
-]
-KAWAII_READ = [
-    "φ(゜▽゜*)♪", "( ˘▽˘)っ", "(⌐■_■)", "٩(｡•́‿•̀｡)۶", "(◕‿◕✿)",
-    "ヾ(＠⌒ー⌒＠)ノ", "(✧ω✧)", "♪(๑ᴖ◡ᴖ๑)♪", "(≧◡≦)", "( ´ ▽ ` )ノ",
-]
-KAWAII_TERMINAL = [
-    "ヽ(>∀<☆)ノ", "(ノ°∀°)ノ", "٩(^ᴗ^)۶", "ヾ(⌐■_■)ノ♪", "(•̀ᴗ•́)و",
-    "┗(＾0＾)┓", "(｀・ω・´)", "＼(￣▽￣)／", "(ง •̀_•́)ง", "ヽ(´▽`)/",
-]
-KAWAII_BROWSER = [
-    "(ノ°∀°)ノ", "(☞゚ヮ゚)☞", "( ͡° ͜ʖ ͡°)", "┌( ಠ_ಠ)┘", "(⊙_⊙)？",
-    "ヾ(•ω•`)o", "(￣ω￣)", "( ˇωˇ )", "(ᵔᴥᵔ)", "＼(◎o◎)／",
-]
-KAWAII_CREATE = [
-    "✧*。٩(ˊᗜˋ*)و✧", "(ﾉ◕ヮ◕)ﾉ*:・ﾟ✧", "ヽ(>∀<☆)ノ", "٩(♡ε♡)۶", "(◕‿◕)♡",
-    "✿◕ ‿ ◕✿", "(*≧▽≦)", "ヾ(＾-＾)ノ", "(☆▽☆)", "°˖✧◝(⁰▿⁰)◜✧˖°",
-]
-KAWAII_SKILL = [
-    "ヾ(＠⌒ー⌒＠)ノ", "(๑˃ᴗ˂)ﻭ", "٩(◕‿◕｡)۶", "(✿╹◡╹)", "ヽ(・∀・)ノ",
-    "(ノ´ヮ`)ノ*:・ﾟ✧", "♪(๑ᴖ◡ᴖ๑)♪", "(◠‿◠)", "٩(ˊᗜˋ*)و", "(＾▽＾)",
-    "ヾ(＾∇＾)", "(★ω★)/", "٩(｡•́‿•̀｡)۶", "(◕ᴗ◕✿)", "＼(◎o◎)／",
-    "(✧ω✧)", "ヽ(>∀<☆)ノ", "( ˘▽˘)っ", "(≧◡≦) ♡", "ヾ(￣▽￣)",
-]
-KAWAII_THINK = [
-    "(っ°Д°;)っ", "(；′⌒`)", "(・_・ヾ", "( ´_ゝ`)", "(￣ヘ￣)",
-    "(。-`ω´-)", "( ˘︹˘ )", "(¬_¬)", "ヽ(ー_ー )ノ", "(；一_一)",
-]
-KAWAII_GENERIC = [
-    "♪(´ε` )", "(◕‿◕✿)", "ヾ(＾∇＾)", "٩(◕‿◕｡)۶", "(✿◠‿◠)",
-    "(ノ´ヮ`)ノ*:・ﾟ✧", "ヽ(>∀<☆)ノ", "(☆▽☆)", "( ˘▽˘)っ", "(≧◡≦)",
-]
-
-
-# =========================================================================
-# Cute tool message (completion line that replaces the spinner)
-# =========================================================================
-
-def _detect_tool_failure(tool_name: str, result: str | None) -> tuple[bool, str]:
-    """Inspect a tool result string for signs of failure.
-
-    Returns ``(is_failure, suffix)`` where *suffix* is an informational tag
-    like ``" [exit 1]"`` for terminal failures, or ``" [error]"`` for generic
-    failures.  On success, returns ``(False, "")``.
-    """
-    if result is None:
-        return False, ""
-
-    if tool_name == "terminal":
-        try:
-            data = json.loads(result)
-            exit_code = data.get("exit_code")
-            if exit_code is not None and exit_code != 0:
-                return True, f" [exit {exit_code}]"
-        except (json.JSONDecodeError, TypeError, AttributeError):
-            logger.debug("Could not parse terminal result as JSON for exit code check")
-        return False, ""
-
-    # Memory-specific: distinguish "full" from real errors
-    if tool_name == "memory":
-        try:
-            data = json.loads(result)
-            if data.get("success") is False and "exceed the limit" in data.get("error", ""):
-                return True, " [full]"
-        except (json.JSONDecodeError, TypeError, AttributeError):
-            logger.debug("Could not parse memory result as JSON for capacity check")
-
-    # Generic heuristic for non-terminal tools
-    lower = result[:500].lower()
-    if '"error"' in lower or '"failed"' in lower or result.startswith("Error"):
-        return True, " [error]"
-
-    return False, ""
-
-
-def get_cute_tool_message(
-    tool_name: str, args: dict, duration: float, result: str | None = None,
-) -> str:
-    """Generate a formatted tool completion line for CLI quiet mode.
-
-    Format: ``| {emoji} {verb:9} {detail}  {duration}``
-
-    When *result* is provided the line is checked for failure indicators.
-    Failed tool calls get a red prefix and an informational suffix.
-    """
-    dur = f"{duration:.1f}s"
-    is_failure, failure_suffix = _detect_tool_failure(tool_name, result)
-    skin_prefix = get_skin_tool_prefix()
-
-    def _trunc(s, n=40):
-        s = str(s)
-        return (s[:n-3] + "...") if len(s) > n else s
-
-    def _path(p, n=35):
-        p = str(p)
-        return ("..." + p[-(n-3):]) if len(p) > n else p
-
-    def _wrap(line: str) -> str:
-        """Apply skin tool prefix and failure suffix."""
-        if skin_prefix != "┊":
-            line = line.replace("┊", skin_prefix, 1)
-        if not is_failure:
-            return line
-        return f"{line}{failure_suffix}"
-
-    if tool_name == "web_search":
-        return _wrap(f"┊ 🔍 search    {_trunc(args.get('query', ''), 42)}  {dur}")
-    if tool_name == "web_extract":
-        urls = args.get("urls", [])
-        if urls:
-            url = urls[0] if isinstance(urls, list) else str(urls)
-            domain = url.replace("https://", "").replace("http://", "").split("/")[0]
-            extra = f" +{len(urls)-1}" if len(urls) > 1 else ""
-            return _wrap(f"┊ 📄 fetch     {_trunc(domain, 35)}{extra}  {dur}")
-        return _wrap(f"┊ 📄 fetch     pages  {dur}")
-    if tool_name == "web_crawl":
-        url = args.get("url", "")
-        domain = url.replace("https://", "").replace("http://", "").split("/")[0]
-        return _wrap(f"┊ 🕸️  crawl     {_trunc(domain, 35)}  {dur}")
-    if tool_name == "terminal":
-        return _wrap(f"┊ 💻 $         {_trunc(args.get('command', ''), 42)}  {dur}")
-    if tool_name == "process":
-        action = args.get("action", "?")
-        sid = args.get("session_id", "")[:12]
-        labels = {"list": "ls processes", "poll": f"poll {sid}", "log": f"log {sid}",
-                  "wait": f"wait {sid}", "kill": f"kill {sid}", "write": f"write {sid}", "submit": f"submit {sid}"}
-        return _wrap(f"┊ ⚙️  proc      {labels.get(action, f'{action} {sid}')}  {dur}")
-    if tool_name == "read_file":
-        return _wrap(f"┊ 📖 read      {_path(args.get('path', ''))}  {dur}")
-    if tool_name == "write_file":
-        return _wrap(f"┊ ✍️  write     {_path(args.get('path', ''))}  {dur}")
-    if tool_name == "patch":
-        return _wrap(f"┊ 🔧 patch     {_path(args.get('path', ''))}  {dur}")
-    if tool_name == "search_files":
-        pattern = _trunc(args.get("pattern", ""), 35)
-        target = args.get("target", "content")
-        verb = "find" if target == "files" else "grep"
-        return _wrap(f"┊ 🔎 {verb:9} {pattern}  {dur}")
-    if tool_name == "browser_navigate":
-        url = args.get("url", "")
-        domain = url.replace("https://", "").replace("http://", "").split("/")[0]
-        return _wrap(f"┊ 🌐 navigate  {_trunc(domain, 35)}  {dur}")
-    if tool_name == "browser_snapshot":
-        mode = "full" if args.get("full") else "compact"
-        return _wrap(f"┊ 📸 snapshot  {mode}  {dur}")
-    if tool_name == "browser_click":
-        return _wrap(f"┊ 👆 click     {args.get('ref', '?')}  {dur}")
-    if tool_name == "browser_type":
-        return _wrap(f"┊ ⌨️  type      \"{_trunc(args.get('text', ''), 30)}\"  {dur}")
-    if tool_name == "browser_scroll":
-        d = args.get("direction", "down")
-        arrow = {"down": "↓", "up": "↑", "right": "→", "left": "←"}.get(d, "↓")
-        return _wrap(f"┊ {arrow}  scroll    {d}  {dur}")
-    if tool_name == "browser_back":
-        return _wrap(f"┊ ◀️  back      {dur}")
-    if tool_name == "browser_press":
-        return _wrap(f"┊ ⌨️  press     {args.get('key', '?')}  {dur}")
-    if tool_name == "browser_close":
-        return _wrap(f"┊ 🚪 close     browser  {dur}")
-    if tool_name == "browser_get_images":
-        return _wrap(f"┊ 🖼️  images    extracting  {dur}")
-    if tool_name == "browser_vision":
-        return _wrap(f"┊ 👁️  vision    analyzing page  {dur}")
-    if tool_name == "todo":
-        todos_arg = args.get("todos")
-        merge = args.get("merge", False)
-        if todos_arg is None:
-            return _wrap(f"┊ 📋 plan      reading tasks  {dur}")
-        elif merge:
-            return _wrap(f"┊ 📋 plan      update {len(todos_arg)} task(s)  {dur}")
-        else:
-            return _wrap(f"┊ 📋 plan      {len(todos_arg)} task(s)  {dur}")
-    if tool_name == "session_search":
-        return _wrap(f"┊ 🔍 recall    \"{_trunc(args.get('query', ''), 35)}\"  {dur}")
-    if tool_name == "memory":
-        action = args.get("action", "?")
-        target = args.get("target", "")
-        if action == "add":
-            return _wrap(f"┊ 🧠 memory    +{target}: \"{_trunc(args.get('content', ''), 30)}\"  {dur}")
-        elif action == "replace":
-            return _wrap(f"┊ 🧠 memory    ~{target}: \"{_trunc(args.get('old_text', ''), 20)}\"  {dur}")
-        elif action == "remove":
-            return _wrap(f"┊ 🧠 memory    -{target}: \"{_trunc(args.get('old_text', ''), 20)}\"  {dur}")
-        return _wrap(f"┊ 🧠 memory    {action}  {dur}")
-    if tool_name == "skills_list":
-        return _wrap(f"┊ 📚 skills    list {args.get('category', 'all')}  {dur}")
-    if tool_name == "skill_view":
-        return _wrap(f"┊ 📚 skill     {_trunc(args.get('name', ''), 30)}  {dur}")
-    if tool_name == "image_generate":
-        return _wrap(f"┊ 🎨 create    {_trunc(args.get('prompt', ''), 35)}  {dur}")
-    if tool_name == "text_to_speech":
-        return _wrap(f"┊ 🔊 speak     {_trunc(args.get('text', ''), 30)}  {dur}")
-    if tool_name == "vision_analyze":
-        return _wrap(f"┊ 👁️  vision    {_trunc(args.get('question', ''), 30)}  {dur}")
-    if tool_name == "mixture_of_agents":
-        return _wrap(f"┊ 🧠 reason    {_trunc(args.get('user_prompt', ''), 30)}  {dur}")
-    if tool_name == "send_message":
-        return _wrap(f"┊ 📨 send      {args.get('target', '?')}: \"{_trunc(args.get('message', ''), 25)}\"  {dur}")
-    if tool_name == "cronjob":
-        action = args.get("action", "?")
-        if action == "create":
-            skills = args.get("skills") or ([] if not args.get("skill") else [args.get("skill")])
-            label = args.get("name") or (skills[0] if skills else None) or args.get("prompt", "task")
-            return _wrap(f"┊ ⏰ cron      create {_trunc(label, 24)}  {dur}")
-        if action == "list":
-            return _wrap(f"┊ ⏰ cron      listing  {dur}")
-        return _wrap(f"┊ ⏰ cron      {action} {args.get('job_id', '')}  {dur}")
-    if tool_name.startswith("rl_"):
-        rl = {
-            "rl_list_environments": "list envs", "rl_select_environment": f"select {args.get('name', '')}",
-            "rl_get_current_config": "get config", "rl_edit_config": f"set {args.get('field', '?')}",
-            "rl_start_training": "start training", "rl_check_status": f"status {args.get('run_id', '?')[:12]}",
-            "rl_stop_training": f"stop {args.get('run_id', '?')[:12]}", "rl_get_results": f"results {args.get('run_id', '?')[:12]}",
-            "rl_list_runs": "list runs", "rl_test_inference": "test inference",
-        }
-        return _wrap(f"┊ 🧪 rl        {rl.get(tool_name, tool_name.replace('rl_', ''))}  {dur}")
-    if tool_name == "execute_code":
-        code = args.get("code", "")
-        first_line = code.strip().split("\n")[0] if code.strip() else ""
-        return _wrap(f"┊ 🐍 exec      {_trunc(first_line, 35)}  {dur}")
-    if tool_name == "delegate_task":
-        tasks = args.get("tasks")
-        if tasks and isinstance(tasks, list):
-            return _wrap(f"┊ 🔀 delegate  {len(tasks)} parallel tasks  {dur}")
-        return _wrap(f"┊ 🔀 delegate  {_trunc(args.get('goal', ''), 35)}  {dur}")
-
-    preview = build_tool_preview(tool_name, args) or ""
-    return _wrap(f"┊ ⚡ {tool_name[:9]:9} {_trunc(preview, 35)}  {dur}")
-
-
-# =========================================================================
-# Honcho session line (one-liner with clickable OSC 8 hyperlink)
-# =========================================================================
-
-_DIM = "\033[2m"
-_SKY_BLUE = "\033[38;5;117m"
-_ANSI_RESET = "\033[0m"
-
-
-def honcho_session_url(workspace: str, session_name: str) -> str:
-    """Build a Honcho app URL for a session."""
-    from urllib.parse import quote
-    return (
-        f"https://app.honcho.dev/explore"
-        f"?workspace={quote(workspace, safe='')}"
-        f"&view=sessions"
-        f"&session={quote(session_name, safe='')}"
-    )
-
-
-def _osc8_link(url: str, text: str) -> str:
-    """OSC 8 terminal hyperlink (clickable in iTerm2, Ghostty, WezTerm, etc.)."""
-    return f"\033]8;;{url}\033\\{text}\033]8;;\033\\"
-
-
-def honcho_session_line(workspace: str, session_name: str) -> str:
-    """One-line session indicator: `Honcho session: <clickable name>`."""
-    url = honcho_session_url(workspace, session_name)
-    linked_name = _osc8_link(url, f"{_SKY_BLUE}{session_name}{_ANSI_RESET}")
-    return f"{_DIM}Honcho session:{_ANSI_RESET} {linked_name}"
-
-
-def write_tty(text: str) -> None:
-    """Write directly to /dev/tty, bypassing stdout capture."""
-    try:
-        fd = os.open("/dev/tty", os.O_WRONLY)
-        os.write(fd, text.encode("utf-8"))
-        os.close(fd)
-    except OSError:
-        sys.stdout.write(text)
-        sys.stdout.flush()

agent/insights.py

@@ -1,789 +0,0 @@
-"""
-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")
-
-    def _get_sessions(self, cutoff: float, source: str = None) -> List[Dict]:
-        """Fetch sessions within the time window."""
-        if source:
-            cursor = self._conn.execute(
-                f"""SELECT {self._SESSION_COLS} FROM sessions
-                    WHERE started_at >= ? AND source = ?
-                    ORDER BY started_at DESC""",
-                (cutoff, source),
-            )
-        else:
-            cursor = self._conn.execute(
-                f"""SELECT {self._SESSION_COLS} FROM sessions
-                    WHERE started_at >= ?
-                    ORDER BY started_at DESC""",
-                (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']:,}")
-        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(f"  * 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']:,}")
-        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

@@ -1,496 +0,0 @@
-"""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__)
-
-_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 high and step down on context-length errors until one works.
-CONTEXT_PROBE_TIERS = [
-    2_000_000,
-    1_000_000,
-    512_000,
-    200_000,
-    128_000,
-    64_000,
-    32_000,
-]
-
-DEFAULT_CONTEXT_LENGTHS = {
-    "anthropic/claude-opus-4": 200000,
-    "anthropic/claude-opus-4.5": 200000,
-    "anthropic/claude-opus-4.6": 200000,
-    "anthropic/claude-sonnet-4": 200000,
-    "anthropic/claude-sonnet-4-20250514": 200000,
-    "anthropic/claude-sonnet-4.5": 200000,
-    "anthropic/claude-sonnet-4.6": 200000,
-    "anthropic/claude-haiku-4.5": 200000,
-    # Bare Anthropic model IDs (for native API provider)
-    "claude-opus-4-6": 200000,
-    "claude-sonnet-4-6": 200000,
-    "claude-opus-4-5-20251101": 200000,
-    "claude-sonnet-4-5-20250929": 200000,
-    "claude-opus-4-1-20250805": 200000,
-    "claude-opus-4-20250514": 200000,
-    "claude-sonnet-4-20250514": 200000,
-    "claude-haiku-4-5-20251001": 200000,
-    "openai/gpt-5": 128000,
-    "openai/gpt-4.1": 1047576,
-    "openai/gpt-4.1-mini": 1047576,
-    "openai/gpt-4o": 128000,
-    "openai/gpt-4-turbo": 128000,
-    "openai/gpt-4o-mini": 128000,
-    "google/gemini-3-pro-preview": 1048576,
-    "google/gemini-3-flash": 1048576,
-    "google/gemini-2.5-flash": 1048576,
-    "google/gemini-2.0-flash": 1048576,
-    "google/gemini-2.5-pro": 1048576,
-    "deepseek/deepseek-v3.2": 65536,
-    "meta-llama/llama-3.3-70b-instruct": 131072,
-    "deepseek/deepseek-chat-v3": 65536,
-    "qwen/qwen-2.5-72b-instruct": 32768,
-    "glm-4.7": 202752,
-    "glm-5": 202752,
-    "glm-4.5": 131072,
-    "glm-4.5-flash": 131072,
-    "kimi-for-coding": 262144,
-    "kimi-k2.5": 262144,
-    "kimi-k2-thinking": 262144,
-    "kimi-k2-thinking-turbo": 262144,
-    "kimi-k2-turbo-preview": 262144,
-    "kimi-k2-0905-preview": 131072,
-    "MiniMax-M2.7": 204800,
-    "MiniMax-M2.7-highspeed": 204800,
-    "MiniMax-M2.5": 204800,
-    "MiniMax-M2.5-highspeed": 204800,
-    "MiniMax-M2.1": 204800,
-    # OpenCode Zen models
-    "gpt-5.4-pro": 128000,
-    "gpt-5.4": 128000,
-    "gpt-5.3-codex": 128000,
-    "gpt-5.3-codex-spark": 128000,
-    "gpt-5.2": 128000,
-    "gpt-5.2-codex": 128000,
-    "gpt-5.1": 128000,
-    "gpt-5.1-codex": 128000,
-    "gpt-5.1-codex-max": 128000,
-    "gpt-5.1-codex-mini": 128000,
-    "gpt-5": 128000,
-    "gpt-5-codex": 128000,
-    "gpt-5-nano": 128000,
-    # Bare model IDs without provider prefix (avoid duplicates with entries above)
-    "claude-opus-4-5": 200000,
-    "claude-opus-4-1": 200000,
-    "claude-sonnet-4-5": 200000,
-    "claude-sonnet-4": 200000,
-    "claude-haiku-4-5": 200000,
-    "claude-3-5-haiku": 200000,
-    "gemini-3.1-pro": 1048576,
-    "gemini-3-pro": 1048576,
-    "gemini-3-flash": 1048576,
-    "minimax-m2.5": 204800,
-    "minimax-m2.5-free": 204800,
-    "minimax-m2.1": 204800,
-    "glm-4.6": 202752,
-    "kimi-k2": 262144,
-    "qwen3-coder": 32768,
-    "big-pickle": 128000,
-    # Alibaba Cloud / DashScope Qwen models
-    "qwen3.5-plus": 131072,
-    "qwen3-max": 131072,
-    "qwen3-coder-plus": 131072,
-    "qwen3-coder-next": 131072,
-    "qwen-plus-latest": 131072,
-    "qwen3.5-flash": 131072,
-    "qwen-vl-max": 32768,
-}
-
-_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",
-)
-
-_MAX_COMPLETION_KEYS = (
-    "max_completion_tokens",
-    "max_output_tokens",
-    "max_tokens",
-)
-
-
-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)
-
-
-def _is_known_provider_base_url(base_url: str) -> bool:
-    normalized = _normalize_base_url(base_url)
-    if not normalized:
-        return False
-    parsed = urlparse(normalized if "://" in normalized else f"https://{normalized}")
-    host = parsed.netloc.lower() or parsed.path.lower()
-    known_hosts = (
-        "api.openai.com",
-        "chatgpt.com",
-        "api.anthropic.com",
-        "api.z.ai",
-        "api.moonshot.ai",
-        "api.kimi.com",
-        "api.minimax",
-    )
-    return any(known_host in host for known_host in known_hosts)
-
-
-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)
-
-            _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 get_model_context_length(model: str, base_url: str = "", api_key: str = "") -> int:
-    """Get the context length for a model.
-
-    Resolution order:
-    1. Persistent cache (previously discovered via probing)
-    2. Active endpoint metadata (/models for explicit custom endpoints)
-    3. OpenRouter API metadata
-    4. Hardcoded DEFAULT_CONTEXT_LENGTHS (fuzzy match for hosted routes only)
-    5. First probe tier (2M) — will be narrowed on first context error
-    """
-    # 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 explicit custom routes
-    if _is_custom_endpoint(base_url):
-        endpoint_metadata = fetch_endpoint_model_metadata(base_url, api_key=api_key)
-        if model in endpoint_metadata:
-            context_length = endpoint_metadata[model].get("context_length")
-            if isinstance(context_length, int):
-                return context_length
-        if not _is_known_provider_base_url(base_url):
-            # Explicit third-party endpoints should not borrow fuzzy global
-            # defaults from unrelated providers with similarly named models.
-            return CONTEXT_PROBE_TIERS[0]
-
-    # 3. OpenRouter API metadata
-    metadata = fetch_model_metadata()
-    if model in metadata:
-        return metadata[model].get("context_length", 128000)
-
-    # 4. Hardcoded defaults (fuzzy match — longest key first for specificity)
-    for default_model, length in sorted(
-        DEFAULT_CONTEXT_LENGTHS.items(), key=lambda x: len(x[0]), reverse=True
-    ):
-        if default_model in model or model in default_model:
-            return length
-
-    # 5. Unknown model — start at highest probe tier
-    return CONTEXT_PROBE_TIERS[0]
-
-
-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

agent/prompt_builder.py

@@ -1,546 +0,0 @@
-"""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 logging
-import os
-import re
-from pathlib import Path
-from typing import Optional
-
-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."
-)
-
-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. Your final response is automatically "
-        "delivered to the job's configured destination, so do not use send_message to "
-        "send to that same target again. If you want the user to receive something in "
-        "the scheduled destination, put it directly in your final response. Use "
-        "send_message only for additional or different targets."
-    ),
-    "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 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:
-        from tools.skills_tool import _parse_frontmatter, skill_matches_platform
-
-        raw = skill_file.read_text(encoding="utf-8")[:2000]
-        frontmatter, _ = _parse_frontmatter(raw)
-
-        if not skill_matches_platform(frontmatter):
-            return False, {}, ""
-
-        desc = ""
-        raw_desc = frontmatter.get("description", "")
-        if raw_desc:
-            desc = str(raw_desc).strip().strip("'\"")
-            if len(desc) > 60:
-                desc = desc[:57] + "..."
-
-        return True, frontmatter, desc
-    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:
-        from tools.skills_tool import _parse_frontmatter
-        raw = skill_file.read_text(encoding="utf-8")[:2000]
-        frontmatter, _ = _parse_frontmatter(raw)
-        hermes = frontmatter.get("metadata", {}).get("hermes", {})
-        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", []),
-        }
-    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.
-
-    Scans ~/.hermes/skills/ for SKILL.md files grouped by category.
-    Includes per-skill descriptions from frontmatter so the model can
-    match skills by meaning, not just name.
-    Filters out skills incompatible with the current OS platform.
-    """
-    hermes_home = Path(os.getenv("HERMES_HOME", Path.home() / ".hermes"))
-    skills_dir = hermes_home / "skills"
-
-    if not skills_dir.exists():
-        return ""
-
-    # Collect skills with descriptions, grouped by category.
-    # Each entry: (skill_name, description)
-    # Supports sub-categories: skills/mlops/training/axolotl/SKILL.md
-    # -> category "mlops/training", skill "axolotl"
-    # Load disabled skill names once for the entire scan
-    try:
-        from tools.skills_tool import _get_disabled_skill_names
-        disabled = _get_disabled_skill_names()
-    except Exception:
-        disabled = set()
-
-    skills_by_category: dict[str, list[tuple[str, str]]] = {}
-    for skill_file in skills_dir.rglob("SKILL.md"):
-        is_compatible, frontmatter, desc = _parse_skill_file(skill_file)
-        if not is_compatible:
-            continue
-        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
-        # Respect user's disabled skills config
-        fm_name = frontmatter.get("name", skill_name)
-        if fm_name in disabled or skill_name in disabled:
-            continue
-        # Skip skills whose conditional activation rules exclude them
-        conditions = _read_skill_conditions(skill_file)
-        if not _skill_should_show(conditions, available_tools, available_toolsets):
-            continue
-        skills_by_category.setdefault(category, []).append((skill_name, desc))
-
-    if not skills_by_category:
-        return ""
-
-    # Read category-level descriptions from DESCRIPTION.md
-    # Checks both the exact category path and parent directories
-    category_descriptions = {}
-    for category in skills_by_category:
-        cat_path = Path(category)
-        desc_file = skills_dir / cat_path / "DESCRIPTION.md"
-        if desc_file.exists():
-            try:
-                content = desc_file.read_text(encoding="utf-8")
-                match = re.search(r"^---\s*\n.*?description:\s*(.+?)\s*\n.*?^---", content, re.MULTILINE | re.DOTALL)
-                if match:
-                    category_descriptions[category] = match.group(1).strip()
-            except Exception as e:
-                logger.debug("Could not read skill description %s: %s", desc_file, e)
-
-    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}")
-
-    return (
-        "## 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."
-    )
-
-
-# =========================================================================
-# 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 build_context_files_prompt(cwd: Optional[str] = None) -> str:
-    """Discover and load context files for the system prompt.
-
-    Discovery: AGENTS.md (recursive), .cursorrules / .cursor/rules/*.mdc,
-    and SOUL.md from HERMES_HOME only. Each capped at 20,000 chars.
-    """
-    if cwd is None:
-        cwd = os.getcwd()
-
-    cwd_path = Path(cwd).resolve()
-    sections = []
-
-    # AGENTS.md (hierarchical, recursive)
-    top_level_agents = None
-    for name in ["AGENTS.md", "agents.md"]:
-        candidate = cwd_path / name
-        if candidate.exists():
-            top_level_agents = candidate
-            break
-
-    if top_level_agents:
-        agents_files = []
-        for root, dirs, files in os.walk(cwd_path):
-            dirs[:] = [d for d in dirs if not d.startswith('.') and d not in ('node_modules', '__pycache__', 'venv', '.venv')]
-            for f in files:
-                if f.lower() == "agents.md":
-                    agents_files.append(Path(root) / f)
-        agents_files.sort(key=lambda p: len(p.parts))
-
-        total_agents_content = ""
-        for agents_path in agents_files:
-            try:
-                content = agents_path.read_text(encoding="utf-8").strip()
-                if content:
-                    rel_path = agents_path.relative_to(cwd_path)
-                    content = _scan_context_content(content, str(rel_path))
-                    total_agents_content += f"## {rel_path}\n\n{content}\n\n"
-            except Exception as e:
-                logger.debug("Could not read %s: %s", agents_path, e)
-
-        if total_agents_content:
-            total_agents_content = _truncate_content(total_agents_content, "AGENTS.md")
-            sections.append(total_agents_content)
-
-    # .cursorrules
-    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 cursorrules_content:
-        cursorrules_content = _truncate_content(cursorrules_content, ".cursorrules")
-        sections.append(cursorrules_content)
-
-    # .hermes.md / HERMES.md — per-project agent config (walk to git root)
-    hermes_md_content = ""
-    hermes_md_path = _find_hermes_md(cwd_path)
-    if hermes_md_path:
-        try:
-            content = hermes_md_path.read_text(encoding="utf-8").strip()
-            if content:
-                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)
-                hermes_md_content = f"## {rel}\n\n{content}"
-        except Exception as e:
-            logger.debug("Could not read %s: %s", hermes_md_path, e)
-
-    if hermes_md_content:
-        hermes_md_content = _truncate_content(hermes_md_content, ".hermes.md")
-        sections.append(hermes_md_content)
-
-    # SOUL.md from HERMES_HOME only
-    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 = Path(os.getenv("HERMES_HOME", Path.home() / ".hermes")) / "SOUL.md"
-    if soul_path.exists():
-        try:
-            content = soul_path.read_text(encoding="utf-8").strip()
-            if content:
-                content = _scan_context_content(content, "SOUL.md")
-                content = _truncate_content(content, "SOUL.md")
-                sections.append(content)
-        except Exception as e:
-            logger.debug("Could not read SOUL.md from %s: %s", soul_path, e)
-
-    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

@@ -1,70 +0,0 @@
-"""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) -> None:
-    """Add cache_control to a single message, handling all format variations."""
-    role = msg.get("role", "")
-    content = msg.get("content")
-
-    if role == "tool":
-        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",
-) -> 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)
-        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)
-
-    return messages

agent/redact.py

@@ -1,161 +0,0 @@
-"""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__)
-
-# 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"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
-]
-
-# 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 not text:
-        return text
-    if os.getenv("HERMES_REDACT_SECRETS", "").lower() in ("0", "false", "no", "off"):
-        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

@@ -1,282 +0,0 @@
-"""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:
-        skill_view_target = str(skill_dir.relative_to(SKILLS_DIR))
-        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
-        if not SKILLS_DIR.exists():
-            return _skill_commands
-        disabled = _get_disabled_skill_names()
-        for skill_md in SKILLS_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)
-                # 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
-                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/smart_model_routing.py

@@ -1,184 +0,0 @@
-"""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"),
-            },
-            "label": None,
-            "signature": (
-                primary.get("model"),
-                primary.get("provider"),
-                primary.get("base_url"),
-                primary.get("api_mode"),
-            ),
-        }
-
-    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"),
-            },
-            "label": None,
-            "signature": (
-                primary.get("model"),
-                primary.get("provider"),
-                primary.get("base_url"),
-                primary.get("api_mode"),
-            ),
-        }
-
-    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"),
-        },
-        "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"),
-        ),
-    }

agent/title_generator.py

@@ -1,125 +0,0 @@
-"""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 = 15.0) -> 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

@@ -1,56 +0,0 @@
-"""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

@@ -1,655 +0,0 @@
-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}"
-            text = text.rstrip("0").rstrip(".")
-            return f"{sign}{text}{suffix}"
-
-    return f"{value:,}"

atropos/Dockerfile

@@ -0,0 +1,41 @@
+# Dockerfile for atropos-agent sandbox server
+# Runs inside Nomad containers to handle tool execution
+# Includes bubblewrap for namespace-based slot isolation
+
+FROM python:3.11-slim
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    # Bubblewrap for namespace isolation
+    bubblewrap \
+    # `script` for PTY allocation (used for stable tmux+asciinema startup)
+    util-linux \
+    # Git for SWE-style tasks (cloning repos)
+    git \
+    # tmux for stateful terminal sessions (Phase 4.7+)
+    tmux \
+    # Common tools agents might need
+    curl \
+    wget \
+    jq \
+    # Cleanup
+    && rm -rf /var/lib/apt/lists/*
+
+# Install Python dependencies (sandbox server + optional terminal recording)
+RUN pip install --no-cache-dir aiohttp asciinema
+
+# Copy the sandbox server
+COPY sandbox_server.py /app/sandbox_server.py
+
+WORKDIR /app
+
+# Create data directory for slot workspaces
+RUN mkdir -p /data
+
+# Verify bubblewrap is installed and working
+RUN bwrap --version
+
+EXPOSE 8080
+
+# Default command - can be overridden by Nomad job spec
+CMD ["python", "sandbox_server.py", "--port", "8080", "--slots", "10", "--data-dir", "/data"]

atropos/__init__.py

@@ -0,0 +1,46 @@
+"""
+Atropos integration for Hermes-Agent.
+
+This package is intentionally optional: Hermes-Agent should work without Atropos.
+If you import anything from `atropos.*` without having `atroposlib` installed,
+we raise a clear error with install instructions.
+
+Install (recommended, from repo checkout):
+  uv sync --extra atropos
+
+Or (pip / editable):
+  pip install -e '.[atropos]'
+"""
+
+from __future__ import annotations
+
+
+def _require_atroposlib() -> None:
+    try:
+        import atroposlib  # noqa: F401
+    except ModuleNotFoundError as exc:  # pragma: no cover
+        raise ModuleNotFoundError(
+            "Hermes-Agent Atropos integration requires `atroposlib`, but it is not installed.\n"
+            "Install it with:\n"
+            "  uv sync --extra atropos\n"
+            "or:\n"
+            "  pip install -e '.[atropos]'\n"
+        ) from exc
+
+
+_require_atroposlib()
+
+# Re-export the most commonly used pieces for convenience.
+from .agent import AgentConfig, AgentResult, AgentStep, AtroposAgent, SequenceData  # noqa: E402
+from .envs import AgentEnv, AgentEnvConfig  # noqa: E402
+
+__all__ = [
+    "AtroposAgent",
+    "AgentConfig",
+    "AgentResult",
+    "AgentStep",
+    "SequenceData",
+    "AgentEnv",
+    "AgentEnvConfig",
+]
+

atropos/agent/__init__.py

@@ -0,0 +1,15 @@
+"""
+Agent abstractions for atropos-agent.
+
+Provides the core AtroposAgent class for running ReACT-style agent loops.
+"""
+
+from .atropos_agent import AgentConfig, AgentResult, AgentStep, AtroposAgent, SequenceData
+
+__all__ = [
+    "AtroposAgent",
+    "AgentConfig",
+    "AgentResult",
+    "AgentStep",
+    "SequenceData",
+]

atropos/agent/atropos_agent.py

@@ -0,0 +1,850 @@
+"""
+ReACT-style agent implementation for atropos-agent.
+
+This module provides the core AtroposAgent class that implements a basic
+Reason-Act-Observe loop with tool calling capabilities.
+
+Uses ManagedServer from atroposlib for automatic token/logprob tracking,
+making trajectories ready for RL training.
+
+The agent uses Hermes-style XML tags for tool calls:
+- <think>...</think> for reasoning
+- <tool_call>{"name": "...", "arguments": {...}}</tool_call> for actions
+- <tool_response>...</tool_response> for observations
+"""
+
+import asyncio
+import os
+import json
+import time
+from contextlib import asynccontextmanager
+from dataclasses import dataclass, field
+from uuid import uuid4
+from typing import Any, AsyncGenerator, Awaitable, Callable, Dict, List, Optional, Union
+
+from dotenv import load_dotenv
+import httpx
+
+from ..tools import ToolCall, ToolRegistry, ToolResult
+from atroposlib.envs.server_handling.managed_server import ManagedServer
+
+load_dotenv()
+
+
+# Default system prompt with tool calling instructions.
+AGENT_SYSTEM_PROMPT = """You are a deep thinking AI. You MUST enclose your internal reasoning inside <think>...</think> tags.
+
+You are a function calling AI model.
+
+You are provided with function signatures within <tools></tools> XML tags.
+You must call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions.
+You can ONLY respond without a tool call if you are totally certain you have the final answer to the user's question or task
+After calling & executing a function, you will be provided with function results within <tool_response></tool_response> XML tags.
+
+Here are the available tools:
+<tools>
+{tools_json}
+</tools>
+
+Use the following JSON schema for each tool call you will make:
+{"title": "FunctionCall", "type": "object", "properties": {"name": {"title": "Name", "type": "string"}, "arguments": {"title": "Arguments", "type": "object"}}, "required": ["name", "arguments"]}
+
+## REQUIRED TOOL FORMAT
+
+When you decide to call a tool, your assistant message MUST be:
+1) exactly one <think>...</think> block, followed by
+2) one or more <tool_call>...</tool_call> blocks,
+and NOTHING else in that message.
+
+If you need to explain anything, put it inside <think>. Do NOT write natural language outside <think> or <tool_call>.
+
+For each function call return a JSON object with function name and arguments within <tool_call></tool_call> XML tags as follows:
+<tool_call>
+{"name": "<function-name>", "arguments": {"arg1": "value1"}}
+</tool_call>
+
+Each <tool_call> must be on its own and contain ONLY the JSON object (no extra text).
+The JSON inside <tool_call> MUST be valid JSON with double quotes.
+
+Do NOT output <tool_response> in an assistant message.
+
+After you receive tool results, you may either call more tools (same required format) or provide the final answer.
+When providing the final answer, do NOT include any <tool_call> blocks.
+
+## TERMINAL TOOL NOTES
+
+- Commands execute under POSIX `/bin/sh` (not bash).
+- Each tool call runs in a fresh shell: environment changes (like `cd` or venv activation) do not persist across tool calls.
+- Avoid bash-only features like `source`, `[[ ... ]]`, or process substitution.
+- Prefer explicit venv usage:
+  - `python -m venv .venv && . .venv/bin/activate && python -m pip install -e .` (POSIX `.` activation), or
+  - `.venv/bin/python -m pip install -e .` (no activation required).
+
+## ICL (examples)
+
+User: Show the current directory.
+Assistant:
+<think>I should run pwd.</think>
+<tool_call>
+{"name": "terminal", "arguments": {"command": "pwd"}}
+</tool_call>
+User: <tool_response>{"success": true, "output": "/tmp\\n"}</tool_response>
+Assistant: /tmp
+
+User: List files, then count them.
+Assistant:
+<think>I should count files.</think>
+<tool_call>
+{"name": "terminal", "arguments": {"command": "ls -1 | wc -l"}}
+</tool_call>
+User: <tool_response>{"success": true, "output": "3\\n"}</tool_response>
+Assistant: 3
+
+User: Run pwd, then print ok (two tool calls).
+Assistant:
+<think>I should run two commands.</think>
+<tool_call>
+{"name": "terminal", "arguments": {"command": "pwd"}}
+</tool_call>
+<tool_call>
+{"name": "terminal", "arguments": {"command": "echo ok"}}
+</tool_call>
+User: <tool_response>{"success": true, "output": "/tmp\\n"}</tool_response>
+User: <tool_response>{"success": true, "output": "ok\\n"}</tool_response>
+Assistant: ok
+"""
+
+
+@dataclass
+class AgentConfig:
+    """Configuration for the AtroposAgent."""
+    
+    # Generation parameters
+    temperature: Optional[float] = 0.7
+    # Default to "let the backend decide" (important for tool-tag completions that may be longer).
+    max_tokens: Optional[int] = None
+    
+    # Agent behavior
+    max_steps: int = 50
+    system_prompt: Optional[str] = None
+    tool_delay_s: float = 0.0
+    
+    # Working directory for tools
+    working_dir: Optional[str] = None
+
+
+@dataclass
+class SequenceData:
+    """Token/logprob data from a single completion."""
+    
+    full_text: str
+    tokens: List[int]
+    masked_tokens: List[int]  # -100 for prompt, actual IDs for completion
+    logprobs: List[float]  # 1.0 for prompt, actual values for completion
+    metadata: Optional[Dict[str, Any]] = None
+    
+    @classmethod
+    def from_sequence_node(cls, node) -> "SequenceData":
+        """Create from a ManagedServer SequenceNode."""
+        return cls(
+            full_text=node.full_text,
+            tokens=node.tokens,
+            masked_tokens=node.masked_tokens,
+            logprobs=node.logprobs,
+            metadata=getattr(node, "metadata", None),
+        )
+
+
+@dataclass
+class AgentStep:
+    """A single step in the agent's trajectory."""
+    
+    step_number: int
+    assistant_message: str
+    tool_calls: List[ToolCall] = field(default_factory=list)
+    tool_results: List[ToolResult] = field(default_factory=list)
+    sequence_data: Optional[SequenceData] = None  # Token data from this step
+    
+    @property
+    def has_tool_calls(self) -> bool:
+        return len(self.tool_calls) > 0
+
+
+@dataclass
+class AgentResult:
+    """Result of running an agent trajectory."""
+    
+    success: bool
+    final_response: str
+    steps: List[AgentStep] = field(default_factory=list)
+    total_tokens: int = 0
+    error: Optional[str] = None
+    metadata: Dict[str, Any] = field(default_factory=dict)
+    
+    # Full trajectory token data for RL training
+    trajectory_data: Optional[SequenceData] = None
+    
+    @property
+    def num_steps(self) -> int:
+        return len(self.steps)
+    
+    @property
+    def total_tool_calls(self) -> int:
+        return sum(len(step.tool_calls) for step in self.steps)
+    
+    def to_messages(self) -> List[Dict[str, str]]:
+        """Convert trajectory to messages format for logging."""
+        messages = []
+        for step in self.steps:
+            messages.append({"role": "assistant", "content": step.assistant_message})
+            if step.tool_results:
+                # Combine all tool responses
+                responses = "\n".join(r.to_xml() for r in step.tool_results)
+                messages.append({"role": "user", "content": responses})
+        return messages
+    
+    def to_scored_data(self, score: float) -> Optional[Dict[str, Any]]:
+        """
+        Convert to format suitable for ScoredDataGroup.
+        
+        Args:
+            score: The score for this trajectory
+            
+        Returns:
+            Dict with tokens, masks, scores suitable for training, or None if no data
+        """
+        if self.trajectory_data is None:
+            return None
+        
+        return {
+            "tokens": self.trajectory_data.tokens,
+            "masks": self.trajectory_data.masked_tokens,
+            "scores": score,
+            "logprobs": self.trajectory_data.logprobs,
+        }
+
+
+class AtroposAgent:
+    """
+    A ReACT-style agent that uses LLMs with tool calling.
+    
+    This implementation wraps ManagedServer for automatic token/logprob tracking,
+    making trajectories ready for RL training.
+    
+    Example:
+        # `server` may be an Atropos `ServerManager` (recommended) or a single `APIServer`.
+        # In practice, environments usually construct this via `BaseEnv`.
+        server = ...
+        tools = ToolRegistry()
+        tools.register(BashTool())
+        
+        agent = AtroposAgent(server=server, tools=tools)
+        result = await agent.run("List the files in the current directory")
+        
+        # Access token data for training
+        if result.trajectory_data:
+            print(f"Tokens: {result.trajectory_data.tokens}")
+            print(f"Masked: {result.trajectory_data.masked_tokens}")
+    """
+    
+    def __init__(
+        self,
+        server,  # ServerManager or APIServer
+        tools: Optional[ToolRegistry] = None,
+        config: Optional[AgentConfig] = None,
+        tokenizer: Optional[Any] = None,
+        execute_tool: Optional[Callable[[ToolCall], Awaitable[ToolResult]]] = None,
+    ):
+        self.server = server
+        self.tools = tools or ToolRegistry()
+        self.config = config or AgentConfig()
+        self.tokenizer = tokenizer or getattr(server, "tokenizer", None)
+        self.execute_tool = execute_tool or self.tools.execute
+
+    @asynccontextmanager
+    async def _managed(self) -> AsyncGenerator[Any, None]:
+        """
+        Yield a ManagedServer-like object.
+
+        - If `self.server` is a ServerManager, use its `managed_server()` context manager.
+        - If `self.server` is a single APIServer, wrap it in `ManagedServer` directly.
+        """
+        if os.getenv("ATROPOS_BYPASS_MANAGED_SERVER") == "1":
+            yield _DirectChatCompletionClient(server=self.server)
+            return
+        if hasattr(self.server, "managed_server"):
+            async with self.server.managed_server(tokenizer=self.tokenizer) as managed:
+                yield managed
+        else:
+            managed = ManagedServer(server=self.server, tokenizer=self.tokenizer)
+            try:
+                yield managed
+            finally:
+                managed.reset()
+    
+    def _build_system_prompt(self) -> str:
+        """Build the system prompt with tool descriptions."""
+        if self.config.system_prompt:
+            return self.config.system_prompt
+
+        tools_json = self.tools.get_prompt_tool_definitions_json()
+        # Avoid `str.format()` here because the prompt contains many literal `{}` braces
+        # in JSON examples; we only want to substitute the single `{tools_json}` token.
+        return AGENT_SYSTEM_PROMPT.replace("{tools_json}", tools_json)
+
+    def _infer_server_model_for_debug(self) -> Optional[str]:
+        """
+        Best-effort inference of the configured model name for debug payload saving.
+
+        ManagedServer/server_manager typically injects `model` internally, so `chat_kwargs`
+        may not contain it. For replaying saved payloads via curl, it's useful to persist it.
+        """
+        servers = getattr(self.server, "servers", None)
+        if isinstance(servers, list) and servers:
+            s0 = servers[0]
+            cfg = getattr(s0, "config", None)
+            model = getattr(cfg, "model_name", None) or getattr(s0, "model_name", None)
+            if isinstance(model, str) and model:
+                return model
+        model = getattr(self.server, "model_name", None) or getattr(self.server, "model", None)
+        if isinstance(model, str) and model:
+            return model
+        return None
+
+    def _infer_server_base_url_for_debug(self) -> Optional[str]:
+        """
+        Best-effort inference of the configured base_url for debug logging.
+
+        This is helpful when diagnosing hangs / retries at the transport layer.
+        """
+        servers = getattr(self.server, "servers", None)
+        if isinstance(servers, list) and servers:
+            s0 = servers[0]
+            cfg = getattr(s0, "config", None)
+            base_url = getattr(cfg, "base_url", None) or getattr(s0, "base_url", None)
+            if isinstance(base_url, str) and base_url:
+                return base_url
+        base_url = getattr(self.server, "base_url", None)
+        if isinstance(base_url, str) and base_url:
+            return base_url
+        return None
+
+    def _extract_response_metadata(self, response: Any) -> Dict[str, Any]:
+        """
+        Extract lightweight, JSON-serializable metadata from an OpenAI-style response.
+
+        This is useful for debugging training runs, especially when ManagedServer state
+        tracking is unavailable (e.g. OpenAI-compatible chat endpoints).
+        """
+        meta: Dict[str, Any] = {}
+        try:
+            rid = getattr(response, "id", None)
+            if isinstance(rid, str) and rid:
+                meta["id"] = rid
+            model = getattr(response, "model", None)
+            if isinstance(model, str) and model:
+                meta["model"] = model
+            created = getattr(response, "created", None)
+            if isinstance(created, int):
+                meta["created"] = created
+            system_fingerprint = getattr(response, "system_fingerprint", None)
+            if isinstance(system_fingerprint, str) and system_fingerprint:
+                meta["system_fingerprint"] = system_fingerprint
+
+            choices = getattr(response, "choices", None)
+            if isinstance(choices, list) and choices:
+                fr = getattr(choices[0], "finish_reason", None)
+                if isinstance(fr, str) and fr:
+                    meta["finish_reason"] = fr
+
+            usage = getattr(response, "usage", None)
+            if usage is not None:
+                if hasattr(usage, "model_dump"):
+                    meta["usage"] = usage.model_dump()
+                elif isinstance(usage, dict):
+                    meta["usage"] = usage
+        except Exception:
+            pass
+        return meta
+
+    def _debug_dump_request(self, *, step_num: int, chat_kwargs: Dict[str, Any]) -> None:
+        if os.getenv("ATROPOS_DEBUG_AGENT_REQUEST") != "1":
+            return
+        try:
+            # Avoid dumping megabytes by default; messages can be huge.
+            meta = {
+                "step": step_num,
+                "base_url": self._infer_server_base_url_for_debug(),
+                "model": chat_kwargs.get("model") or self._infer_server_model_for_debug(),
+                "chat_kwargs_keys": sorted(list(chat_kwargs.keys())),
+                "n": chat_kwargs.get("n"),
+                "max_tokens": chat_kwargs.get("max_tokens"),
+                "temperature": chat_kwargs.get("temperature"),
+                "num_messages": len(chat_kwargs.get("messages") or []),
+            }
+            print("\n=== ATROPOS_DEBUG_AGENT_REQUEST ===", flush=True)
+            print(meta, flush=True)
+
+            if os.getenv("ATROPOS_DEBUG_AGENT_REQUEST_FULL") == "1":
+                payload = dict(chat_kwargs)
+                # Make the payload more legible and less huge.
+                try:
+                    dumped = json.dumps(payload, ensure_ascii=False, indent=2)
+                except Exception:
+                    dumped = repr(payload)
+                print("\n=== ATROPOS_DEBUG_AGENT_REQUEST_FULL ===", flush=True)
+                print(dumped[:200_000], flush=True)
+
+            # Optional: save the FULL request payload to disk (no truncation).
+            save_dir = os.getenv("ATROPOS_DEBUG_AGENT_REQUEST_SAVE_DIR")
+            if save_dir:
+                os.makedirs(save_dir, exist_ok=True)
+                payload: Dict[str, Any] = dict(chat_kwargs)
+                if "model" not in payload:
+                    model = self._infer_server_model_for_debug()
+                    if model:
+                        payload["model"] = model
+                # Use a unique filename so parallel trajectories don't clobber each other.
+                fname = os.path.join(
+                    save_dir,
+                    f"atropos_agent_request_step{step_num}_{int(time.time()*1000)}_{os.getpid()}_{uuid4().hex}.json",
+                )
+                with open(fname, "w", encoding="utf-8") as f:
+                    json.dump(payload, f, ensure_ascii=False, indent=2)
+                print(f"[AtroposAgent] saved request payload: {fname}", flush=True)
+        except Exception:
+            return
+
+    def _debug_dump_response(self, *, step_num: int, response: Any) -> None:
+        if os.getenv("ATROPOS_DEBUG_AGENT_RESPONSE") != "1":
+            return
+        print("\n=== ATROPOS_DEBUG_AGENT_RESPONSE ===", flush=True)
+        print({"step": step_num, "type": type(response).__name__}, flush=True)
+        try:
+            dumped = response.model_dump()  # openai pydantic model
+        except Exception:
+            dumped = getattr(response, "__dict__", {"repr": repr(response)})
+        # Keep the dump bounded; we only need enough to see the assistant message content.
+        text = str(dumped)
+        print(text[:200_000], flush=True)
+
+    async def _chat_completion_with_debug(
+        self, *, managed: Any, step_num: int, chat_kwargs: Dict[str, Any]
+    ) -> Any:
+        """
+        Call `managed.chat_completion()` with optional timeout + richer failure logging.
+
+        Debug env vars:
+        - `ATROPOS_AGENT_CHAT_TIMEOUT_S`: if set, wraps the await in `asyncio.wait_for`.
+        - `ATROPOS_DEBUG_AGENT_WAIT_EVERY_S`: if set, prints a heartbeat while waiting.
+        """
+        # Hard guardrail: never allow a single chat completion to block for too long.
+        # This is essential for RL data-gen stability; long hangs should be treated as failures (score=0).
+        timeout_s_raw = os.getenv("ATROPOS_AGENT_CHAT_TIMEOUT_S")
+        timeout_s_default = 240.0
+        timeout_s = float(timeout_s_raw) if timeout_s_raw else timeout_s_default
+        timeout_s = min(timeout_s, 240.0)
+
+        wait_every_raw = os.getenv("ATROPOS_DEBUG_AGENT_WAIT_EVERY_S")
+        wait_every_s = float(wait_every_raw) if wait_every_raw else None
+
+        async def _await_call() -> Any:
+            if not wait_every_s or wait_every_s <= 0:
+                return await managed.chat_completion(**chat_kwargs)
+
+            # Heartbeat mode: wait in chunks without cancelling the underlying request.
+            # NOTE: do NOT use `asyncio.wait_for(task, timeout=...)` here, because a timeout
+            # will cancel the task and surface as `CancelledError` on the next loop.
+            task = asyncio.create_task(managed.chat_completion(**chat_kwargs))
+            t0 = time.perf_counter()
+            try:
+                while True:
+                    done, _pending = await asyncio.wait({task}, timeout=wait_every_s)
+                    if task in done:
+                        return task.result()
+
+                    waited = time.perf_counter() - t0
+                    print(
+                        f"[AtroposAgent] step={step_num} still waiting for chat_completion... ({waited:.1f}s)",
+                        flush=True,
+                    )
+            except asyncio.CancelledError:
+                task.cancel()
+                raise
+
+        try:
+            return await asyncio.wait_for(_await_call(), timeout=timeout_s)
+        except asyncio.TimeoutError as e:
+            print("\n=== ATROPOS_DEBUG_AGENT_CHAT_TIMEOUT ===", flush=True)
+            print({"step": step_num, "timeout_s": timeout_s}, flush=True)
+            raise RuntimeError(f"chat_completion timed out after {timeout_s:.1f}s") from e
+        except asyncio.CancelledError:
+            # Treat cancellation as a hard failure rather than crashing the whole env run.
+            # (Atropos/BaseEnv may cancel tasks during shutdown or retries.)
+            raise RuntimeError("chat_completion cancelled") from None
+        except Exception as e:
+            detail: Dict[str, Any] = {
+                "step": step_num,
+                "exc_type": type(e).__name__,
+                "exc_str": str(e),
+            }
+            if isinstance(e, httpx.HTTPStatusError):
+                try:
+                    detail["status_code"] = e.response.status_code
+                    detail["response_text"] = e.response.text[:20_000]
+                except Exception:
+                    pass
+            elif isinstance(e, httpx.RequestError):
+                detail["request"] = repr(getattr(e, "request", None))
+
+            print("\n=== ATROPOS_DEBUG_AGENT_CHAT_FAILURE ===", flush=True)
+            print(detail, flush=True)
+            raise
+
+    async def run(
+        self,
+        task: str,
+        initial_messages: Optional[List[Dict[str, str]]] = None,
+    ) -> AgentResult:
+        """
+        Run the agent on a task using ManagedServer for token tracking.
+        
+        Args:
+            task: The task/prompt for the agent
+            initial_messages: Optional additional context messages
+            
+        Returns:
+            AgentResult with the trajectory, final response, and token data
+        """
+        messages = [
+            {"role": "system", "content": self._build_system_prompt()},
+        ]
+        
+        if initial_messages:
+            messages.extend(initial_messages)
+        
+        messages.append({"role": "user", "content": task})
+        
+        steps = []
+        final_response = ""
+        final_node = None
+        final_prompt_messages: Optional[List[Dict[str, str]]] = None
+        last_node = None
+        last_prompt_messages: Optional[List[Dict[str, str]]] = None
+        last_response_text: str = ""
+        
+        # Use ManagedServer for automatic token tracking
+        async with self._managed() as managed:
+            for step_num in range(self.config.max_steps):
+                # ReACT loop iteration here, just call -> tools -> observe until done (no tools called)
+                try:
+                    # Keep a copy of the prompt messages used for this completion.
+                    # Useful for reconstructing tokens/masks when state tracking is unavailable.
+                    prompt_messages = list(messages)
+                    chat_kwargs: Dict[str, Any] = {"messages": messages, "n": 1}
+                    if self.config.max_tokens is not None:
+                        chat_kwargs["max_tokens"] = self.config.max_tokens
+                    if self.config.temperature is not None:
+                        chat_kwargs["temperature"] = self.config.temperature
+
+                    t_req = time.perf_counter()
+                    print(
+                        f"[AtroposAgent] step={step_num+1} chat_completion start "
+                        f"(messages={len(messages)}, max_tokens={self.config.max_tokens}, temp={self.config.temperature})",
+                        flush=True,
+                    )
+                    self._debug_dump_request(step_num=step_num + 1, chat_kwargs=chat_kwargs)
+                    response = await self._chat_completion_with_debug(
+                        managed=managed, step_num=step_num + 1, chat_kwargs=chat_kwargs
+                    )
+                    self._debug_dump_response(step_num=step_num + 1, response=response)
+                    response_meta = self._extract_response_metadata(response)
+                    print(
+                        f"[AtroposAgent] step={step_num+1} chat_completion done in {time.perf_counter() - t_req:.2f}s",
+                        flush=True,
+                    )
+                    
+                    current_node = None
+                    if hasattr(managed, "get_state"):
+                        state = managed.get_state()
+                        nodes = state.get("nodes", [])
+                        current_node = nodes[-1] if nodes else None
+                    
+                except Exception as e:
+                    return AgentResult(
+                        success=False,
+                        final_response="",
+                        steps=steps,
+                        error=f"Generation error: {str(e)}",
+                    )
+                
+                msg = response.choices[0].message
+                # Some OpenAI-compatible servers populate `message.reasoning` and leave `content=""`.
+                response_text = (msg.content or "") or (getattr(msg, "reasoning", None) or "")
+                tool_calls = ToolCall.parse_from_text(response_text)
+                last_node = current_node
+                last_prompt_messages = prompt_messages
+                last_response_text = response_text
+
+                step_sequence_data = SequenceData.from_sequence_node(current_node) if current_node else None
+                if step_sequence_data is None:
+                    if response_meta:
+                        # We still want metadata for debugging even if token/logprob state tracking is unavailable.
+                        step_sequence_data = SequenceData(
+                            full_text=response_text,
+                            tokens=[],
+                            masked_tokens=[],
+                            logprobs=[],
+                            metadata=response_meta,
+                        )
+                else:
+                    merged = dict(response_meta)
+                    node_meta = step_sequence_data.metadata
+                    if isinstance(node_meta, dict):
+                        merged.update(node_meta)
+                    step_sequence_data.metadata = merged or step_sequence_data.metadata
+                
+                step = AgentStep(
+                    step_number=step_num + 1,
+                    assistant_message=response_text,
+                    tool_calls=tool_calls,
+                    sequence_data=step_sequence_data,
+                )
+                
+                if not tool_calls:
+                    steps.append(step)
+                    final_response = response_text
+                    final_node = current_node
+                    final_prompt_messages = prompt_messages
+                    break
+                
+                messages.append({"role": "assistant", "content": response_text})
+                
+                tool_responses = []
+                for call in tool_calls:
+                    result = await self.execute_tool(call)
+                    step.tool_results.append(result)
+                    tool_responses.append(result.to_xml())
+                    if self.config.tool_delay_s > 0:
+                        await asyncio.sleep(self.config.tool_delay_s)
+                
+                steps.append(step)
+            
+                responses_text = "\n".join(tool_responses)
+                # Tool observations are represented as user content with Hermes-style tags.
+                # This is compatible with most OpenAI-compatible chat APIs and ensures
+                # tokenizers/chat templates include tool outputs during training.
+                messages.append({"role": "user", "content": responses_text})
+            
+            else:
+                # Reached max steps without completing
+                # Return a failure result but include the last observed completion so callers can
+                # record the trajectory (score=0) without triggering retries.
+                final_response = last_response_text or final_response
+                final_node = last_node
+                final_prompt_messages = last_prompt_messages
+                trajectory_data = None
+                if final_node:
+                    trajectory_data = SequenceData.from_sequence_node(final_node)
+                elif final_prompt_messages is not None and self.tokenizer is not None:
+                    if hasattr(self.tokenizer, "apply_chat_template"):
+                        prompt_text = self.tokenizer.apply_chat_template(
+                            final_prompt_messages, tokenize=False, add_generation_prompt=True
+                        )
+                        prompt_tokens = self.tokenizer.encode(prompt_text, add_special_tokens=False)
+                    else:
+                        prompt_text = "\n".join([f"{m['role']}: {m['content']}" for m in final_prompt_messages])
+                        prompt_tokens = self.tokenizer.encode(prompt_text, add_special_tokens=True)
+                    output_tokens = self.tokenizer.encode(final_response, add_special_tokens=False)
+                    tokens = prompt_tokens + output_tokens
+                    masked_tokens = ([-100] * len(prompt_tokens)) + output_tokens
+                    logprobs = ([1.0] * len(prompt_tokens)) + ([0.0] * len(output_tokens))
+                    trajectory_data = SequenceData(
+                        full_text=f"{prompt_text}{final_response}",
+                        tokens=tokens,
+                        masked_tokens=masked_tokens,
+                        logprobs=logprobs,
+                    )
+                # Preserve response metadata (if any) even on failure trajectories.
+                try:
+                    if trajectory_data is not None and steps:
+                        last_step = steps[-1]
+                        if last_step.sequence_data and isinstance(last_step.sequence_data.metadata, dict):
+                            trajectory_data.metadata = dict(last_step.sequence_data.metadata)
+                except Exception:
+                    pass
+                return AgentResult(
+                    success=False,
+                    final_response=final_response,
+                    steps=steps,
+                    error=f"Reached maximum steps ({self.config.max_steps})",
+                    trajectory_data=trajectory_data,
+                )
+        
+        # Build result with trajectory data
+        trajectory_data = None
+        if final_node:
+            trajectory_data = SequenceData.from_sequence_node(final_node)
+        elif final_prompt_messages is not None and self.tokenizer is not None:
+            if hasattr(self.tokenizer, "apply_chat_template"):
+                prompt_text = self.tokenizer.apply_chat_template(
+                    final_prompt_messages, tokenize=False, add_generation_prompt=True
+                )
+                prompt_tokens = self.tokenizer.encode(prompt_text, add_special_tokens=False)
+            else:
+                prompt_text = "\n".join([f"{m['role']}: {m['content']}" for m in final_prompt_messages])
+                prompt_tokens = self.tokenizer.encode(prompt_text, add_special_tokens=True)
+            output_tokens = self.tokenizer.encode(final_response, add_special_tokens=False)
+            tokens = prompt_tokens + output_tokens
+            masked_tokens = ([-100] * len(prompt_tokens)) + output_tokens
+            logprobs = ([1.0] * len(prompt_tokens)) + ([0.0] * len(output_tokens))
+            trajectory_data = SequenceData(
+                full_text=f"{prompt_text}{final_response}",
+                tokens=tokens,
+                masked_tokens=masked_tokens,
+                logprobs=logprobs,
+            )
+
+        # Ensure trajectory_data carries the most recent metadata we observed (if any).
+        try:
+            if trajectory_data is not None and steps:
+                last_step = steps[-1]
+                if last_step.sequence_data and isinstance(last_step.sequence_data.metadata, dict):
+                    trajectory_data.metadata = dict(last_step.sequence_data.metadata)
+        except Exception:
+            pass
+        
+        return AgentResult(
+            success=True,
+            final_response=final_response,
+            steps=steps,
+            trajectory_data=trajectory_data,
+        )
+    
+    async def run_single_turn(
+        self,
+        messages: List[Dict[str, str]],
+        execute_tools: bool = True,
+    ) -> tuple[str, List[ToolResult], Optional[SequenceData]]:
+        """
+        Run a single turn of the agent (one LLM call + tool execution).
+        
+        This is useful for integration with BaseEnv where you want more
+        control over the loop.
+        
+        Args:
+            messages: The conversation history
+            execute_tools: Whether to execute parsed tool calls
+            
+        Returns:
+            Tuple of (response_text, tool_results, sequence_data)
+        """
+        async with self._managed() as managed:
+            chat_kwargs: Dict[str, Any] = {"messages": messages, "n": 1}
+            if self.config.max_tokens is not None:
+                chat_kwargs["max_tokens"] = self.config.max_tokens
+            if self.config.temperature is not None:
+                chat_kwargs["temperature"] = self.config.temperature
+
+            self._debug_dump_request(step_num=1, chat_kwargs=chat_kwargs)
+            response = await self._chat_completion_with_debug(managed=managed, step_num=1, chat_kwargs=chat_kwargs)
+            self._debug_dump_response(step_num=1, response=response)
+            
+            current_node = None
+            if hasattr(managed, "get_state"):
+                state = managed.get_state()
+                nodes = state.get("nodes", [])
+                current_node = nodes[-1] if nodes else None
+        
+        msg = response.choices[0].message
+        response_text = (msg.content or "") or (getattr(msg, "reasoning", None) or "")
+        tool_results = []
+        
+        if execute_tools:
+            tool_calls = ToolCall.parse_from_text(response_text)
+            for call in tool_calls:
+                result = await self.execute_tool(call)
+                tool_results.append(result)
+        
+        sequence_data = SequenceData.from_sequence_node(current_node) if current_node else None
+        
+        return response_text, tool_results, sequence_data
+
+
+class _DirectChatCompletionClient:
+    """
+    Minimal stand-in for ManagedServer that calls the OpenAI-compatible endpoint directly.
+
+    This is for isolating issues where `ManagedServer.chat_completion()` hangs or misbehaves.
+    It intentionally does NOT do token/logprob tracking.
+    """
+
+    def __init__(self, server: Any):
+        self._server = server
+
+    def _server_config(self) -> tuple[str, str, str]:
+        # ServerManager case: first configured server.
+        servers = getattr(self._server, "servers", None)
+        if isinstance(servers, list) and servers:
+            s0 = servers[0]
+            cfg = getattr(s0, "config", None)
+            base_url = getattr(cfg, "base_url", None) or getattr(s0, "base_url", None)
+            api_key = getattr(cfg, "api_key", None) or getattr(s0, "api_key", None)
+            model = getattr(cfg, "model_name", None) or getattr(s0, "model_name", None)
+            if isinstance(base_url, str) and isinstance(api_key, str) and isinstance(model, str):
+                return base_url.rstrip("/"), api_key, model
+
+        # APIServer-like fallback.
+        base_url = getattr(self._server, "base_url", None)
+        api_key = getattr(self._server, "api_key", None)
+        model = getattr(self._server, "model_name", None) or getattr(self._server, "model", None)
+        if isinstance(base_url, str) and isinstance(api_key, str) and isinstance(model, str):
+            return base_url.rstrip("/"), api_key, model
+
+        raise RuntimeError("Unable to resolve server base_url/api_key/model for direct chat completion")
+
+    async def chat_completion(self, *, messages: List[Dict[str, str]], n: int = 1, **kwargs: Any) -> Any:
+        base_url, api_key, model = self._server_config()
+        url = f"{base_url}/chat/completions"
+
+        payload: Dict[str, Any] = {
+            "model": model,
+            "messages": messages,
+            "n": n,
+        }
+        # Pass through common generation kwargs.
+        for k in ("max_tokens", "temperature", "top_p", "presence_penalty", "frequency_penalty", "stop"):
+            if k in kwargs and kwargs[k] is not None:
+                payload[k] = kwargs[k]
+
+        timeout_s = float(os.getenv("ATROPOS_DIRECT_REQUEST_TIMEOUT_S") or "120")
+        print(f"[AtroposAgent] DIRECT chat_completion POST {url} (timeout={timeout_s}s)", flush=True)
+        async with httpx.AsyncClient(timeout=timeout_s) as client:
+            resp = await client.post(
+                url,
+                headers={"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"},
+                json=payload,
+            )
+            resp.raise_for_status()
+            data = resp.json()
+
+        # Return a very small object compatible with the code paths that read
+        # `response.choices[0].message.content`.
+        class _Msg:
+            def __init__(self, d: Dict[str, Any]):
+                self.content = d.get("content")
+                self.reasoning = d.get("reasoning")
+
+        class _Choice:
+            def __init__(self, d: Dict[str, Any]):
+                self.message = _Msg(d.get("message") or {})
+
+        class _Resp:
+            def __init__(self, d: Dict[str, Any]):
+                self._d = d
+                self.choices = [_Choice(c) for c in (d.get("choices") or [])]
+
+            def model_dump(self) -> Dict[str, Any]:
+                return self._d
+
+        return _Resp(data)

atropos/api/__init__.py

@@ -0,0 +1,6 @@
+"""
+FastAPI services for atropos-agent.
+
+- tool_executor_server: queued/batched sandbox tool execution (Phase 4)
+"""
+

atropos/api/tool_executor_server.py

@@ -0,0 +1,254 @@
+"""
+Tool Executor API (Phase 4)
+
+This service provides a queued, batched execution layer on top of a ToolBackend.
+It mirrors the stateful FastAPI + app.state pattern used in:
+  atropos/atroposlib/api/server.py
+
+Run (dev):
+  uv run uvicorn atropos_agent.api.tool_executor_server:app --host 0.0.0.0 --port 9001
+"""
+
+from __future__ import annotations
+
+import os
+from typing import Any, Dict, Optional
+from pathlib import Path
+
+from fastapi import FastAPI, Header, HTTPException, status
+from pydantic import BaseModel, Field
+
+from ..backends.nomad_backend import NomadBackendConfig, NomadToolBackend
+from ..tools import ToolRegistry, build_tool_registry
+from ..tools.base import (
+    ArtifactArchiveRequestPayload,
+    ArtifactArchiveResponsePayload,
+    ArtifactListRequestPayload,
+    ArtifactListResponsePayload,
+    ArtifactReadRequestPayload,
+    ArtifactReadResponsePayload,
+    ToolExecutorExecuteRequest,
+    ToolExecutorReleaseRequest,
+    ToolResultPayload,
+)
+from ..tools.tool_executor import ToolExecutor, ToolExecutorConfig
+
+
+class ToolExecutorServerConfig(BaseModel):
+    nomad_address: str = Field(default="http://localhost:4646")
+    job_id: str = Field(default="atropos-sandbox-tool-executor")
+    image: str = Field(default="atropos-sandbox:local")
+    slots_per_container: int = Field(default=10)
+    min_containers: int = Field(default=1)
+    max_containers: int = Field(default=10)
+    privileged: bool = Field(default=False)
+    acquire_timeout_s: float = Field(default=30.0)
+
+    batch_window_ms: int = Field(default=20)
+    max_batch_size: int = Field(default=200)
+    allow_network: bool = Field(default=True)
+
+    tool_server_url: Optional[str] = Field(default=None)
+    tool_server_token: Optional[str] = Field(default=None)
+
+    token: Optional[str] = Field(default=None, description="Bearer token required for requests (optional in dev).")
+
+    purge_job_on_shutdown: bool = Field(default=True)
+
+    @classmethod
+    def from_env(cls) -> "ToolExecutorServerConfig":
+        # In dev, prefer loading secrets/config from the repo-local `.env` (not committed).
+        try:
+            from dotenv import load_dotenv  # type: ignore
+        except Exception:  # pragma: no cover
+            load_dotenv = None  # type: ignore[assignment]
+        if load_dotenv is not None:
+            env_path = Path(__file__).resolve().parents[2] / ".env"
+            if env_path.exists():
+                load_dotenv(dotenv_path=env_path)
+
+        def _get_bool(name: str, default: bool) -> bool:
+            raw = os.getenv(name)
+            if raw is None:
+                return default
+            return raw.strip().lower() in {"1", "true", "yes", "y", "on"}
+
+        return cls(
+            nomad_address=os.getenv("TOOL_EXECUTOR_NOMAD_ADDRESS", "http://localhost:4646"),
+            job_id=os.getenv("TOOL_EXECUTOR_JOB_ID", "atropos-sandbox-tool-executor"),
+            image=os.getenv("TOOL_EXECUTOR_IMAGE", "atropos-sandbox:local"),
+            slots_per_container=int(os.getenv("TOOL_EXECUTOR_SLOTS", "10")),
+            min_containers=int(os.getenv("TOOL_EXECUTOR_MIN_CONTAINERS", "1")),
+            max_containers=int(os.getenv("TOOL_EXECUTOR_MAX_CONTAINERS", "10")),
+            privileged=_get_bool("TOOL_EXECUTOR_PRIVILEGED", False),
+            acquire_timeout_s=float(os.getenv("TOOL_EXECUTOR_ACQUIRE_TIMEOUT_S", "30.0")),
+            batch_window_ms=int(os.getenv("TOOL_EXECUTOR_BATCH_WINDOW_MS", "20")),
+            max_batch_size=int(os.getenv("TOOL_EXECUTOR_MAX_BATCH_SIZE", "200")),
+            allow_network=_get_bool("TOOL_EXECUTOR_ALLOW_NETWORK", True),
+            tool_server_url=os.getenv("TOOL_EXECUTOR_TOOL_SERVER_URL") or None,
+            tool_server_token=os.getenv("TOOL_EXECUTOR_TOOL_SERVER_TOKEN") or None,
+            token=os.getenv("TOOL_EXECUTOR_TOKEN") or None,
+            purge_job_on_shutdown=_get_bool("TOOL_EXECUTOR_PURGE_JOB_ON_SHUTDOWN", True),
+        )
+
+
+app = FastAPI(title="Atropos-Agent Tool Executor")
+
+
+@app.get("/")
+async def root() -> Dict[str, str]:
+    return {"message": "Atropos-Agent Tool Executor"}
+
+
+def _check_auth(cfg: ToolExecutorServerConfig, authorization: Optional[str]) -> None:
+    if not cfg.token:
+        return
+    if not authorization:
+        raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail="Missing Authorization header")
+    if not authorization.lower().startswith("bearer "):
+        raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail="Invalid Authorization header")
+    token = authorization.split(" ", 1)[1].strip()
+    if token != cfg.token:
+        raise HTTPException(status_code=status.HTTP_403_FORBIDDEN, detail="Invalid token")
+
+
+@app.on_event("startup")
+async def _startup() -> None:
+    cfg = ToolExecutorServerConfig.from_env()
+
+    # Default to Atropos "full" tool surface: sandbox + external (if tool_server_url provided).
+    tools: ToolRegistry = build_tool_registry(
+        enabled_toolsets=["full"],
+        disabled_toolsets=None,
+        tool_server_url=cfg.tool_server_url,
+    )
+
+    backend = NomadToolBackend(
+        NomadBackendConfig(
+            nomad_address=cfg.nomad_address,
+            sandbox_job_id=cfg.job_id,
+            sandbox_image=cfg.image,
+            slots_per_container=cfg.slots_per_container,
+            min_containers=cfg.min_containers,
+            max_containers=cfg.max_containers,
+            privileged=cfg.privileged,
+            acquire_timeout_s=cfg.acquire_timeout_s,
+            purge_job_on_start=False,
+        )
+    )
+    await backend.start()
+
+    executor = ToolExecutor(
+        backend=backend,
+        tools=tools,
+        config=ToolExecutorConfig(
+            batch_window_ms=cfg.batch_window_ms,
+            max_batch_size=cfg.max_batch_size,
+            allow_network=cfg.allow_network,
+            tool_server_url=cfg.tool_server_url,
+            tool_server_token=cfg.tool_server_token,
+        ),
+    )
+    await executor.start()
+
+    app.state.cfg = cfg
+    app.state.backend = backend
+    app.state.executor = executor
+
+
+@app.on_event("shutdown")
+async def _shutdown() -> None:
+    executor: Optional[ToolExecutor] = getattr(app.state, "executor", None)
+    backend: Optional[NomadToolBackend] = getattr(app.state, "backend", None)
+    cfg: Optional[ToolExecutorServerConfig] = getattr(app.state, "cfg", None)
+
+    if executor is not None:
+        await executor.close()
+
+    if backend is not None:
+        await backend.stop(purge=bool(cfg.purge_job_on_shutdown) if cfg else False)
+
+
+@app.get("/health")
+async def health() -> Dict[str, Any]:
+    return {"status": "ok"}
+
+
+@app.get("/status")
+async def status_endpoint() -> Dict[str, Any]:
+    executor: ToolExecutor = app.state.executor
+    backend: NomadToolBackend = app.state.backend
+
+    return {
+        "queue_size": executor.queue_size(),
+        "total_requests": executor.total_requests,
+        "total_errors": executor.total_errors,
+        "pool": backend.get_stats(),
+    }
+
+
+@app.post("/execute", response_model=ToolResultPayload)
+async def execute_tool(
+    req: ToolExecutorExecuteRequest,
+    authorization: Optional[str] = Header(default=None),
+    status_code: int = status.HTTP_200_OK,  # noqa: B008
+) -> ToolResultPayload:
+    cfg: ToolExecutorServerConfig = app.state.cfg
+    _check_auth(cfg, authorization)
+
+    executor: ToolExecutor = app.state.executor
+    result = await executor.execute(
+        trajectory_id=req.trajectory_id,
+        call=req.tool.to_tool_call(),
+        timeout_s=req.timeout_s,
+    )
+    return ToolResultPayload.from_tool_result(result)
+
+
+@app.post("/release")
+async def release_trajectory(
+    req: ToolExecutorReleaseRequest,
+    authorization: Optional[str] = Header(default=None),
+) -> Dict[str, Any]:
+    cfg: ToolExecutorServerConfig = app.state.cfg
+    _check_auth(cfg, authorization)
+
+    executor: ToolExecutor = app.state.executor
+    await executor.release_trajectory(req.trajectory_id, reset_workspace=req.reset_workspace)
+    return {"status": "ok"}
+
+
+@app.post("/artifacts/read", response_model=ArtifactReadResponsePayload)
+async def artifacts_read(
+    req: ArtifactReadRequestPayload,
+    authorization: Optional[str] = Header(default=None),
+) -> ArtifactReadResponsePayload:
+    cfg: ToolExecutorServerConfig = app.state.cfg
+    _check_auth(cfg, authorization)
+
+    executor: ToolExecutor = app.state.executor
+    return await executor.read_artifact(req)
+
+
+@app.post("/artifacts/list", response_model=ArtifactListResponsePayload)
+async def artifacts_list(
+    req: ArtifactListRequestPayload,
+    authorization: Optional[str] = Header(default=None),
+) -> ArtifactListResponsePayload:
+    cfg: ToolExecutorServerConfig = app.state.cfg
+    _check_auth(cfg, authorization)
+
+    executor: ToolExecutor = app.state.executor
+    return await executor.list_artifacts(req)
+
+
+@app.post("/artifacts/archive", response_model=ArtifactArchiveResponsePayload)
+async def artifacts_archive(
+    req: ArtifactArchiveRequestPayload,
+    authorization: Optional[str] = Header(default=None),
+) -> ArtifactArchiveResponsePayload:
+    cfg: ToolExecutorServerConfig = app.state.cfg
+    _check_auth(cfg, authorization)
+
+    executor: ToolExecutor = app.state.executor
+    return await executor.archive_artifacts(req)

atropos/api/tool_server.py

@@ -0,0 +1,140 @@
+"""
+External ToolServer (Phase 4.5+).
+
+This server executes tools that must NOT run inside the sandbox, typically
+because they require credentials or access to external services.
+
+Run (dev):
+  uv run uvicorn atropos_agent.api.tool_server:app --host 0.0.0.0 --port 9002
+"""
+
+from __future__ import annotations
+
+import asyncio
+import os
+import inspect
+from typing import Any, Dict, List, Optional
+from pathlib import Path
+
+from fastapi import FastAPI, Header, HTTPException, status
+from pydantic import BaseModel, Field
+
+from ..tools import ToolRegistry, build_tool_registry
+from ..tools.base import ToolResultPayload, ToolServerExecuteRequest
+
+
+class ToolServerConfig(BaseModel):
+    token: Optional[str] = Field(
+        default=None,
+        description="Bearer token required for requests (optional in dev).",
+    )
+    max_concurrency: int = Field(default=16, ge=1, description="Max concurrent tool executions.")
+
+    @classmethod
+    def from_env(cls) -> "ToolServerConfig":
+        # In dev, prefer loading secrets from the repo-local `.env` (not committed).
+        try:
+            from dotenv import load_dotenv  # type: ignore
+        except Exception:  # pragma: no cover
+            load_dotenv = None  # type: ignore[assignment]
+        if load_dotenv is not None:
+            env_path = Path(__file__).resolve().parents[2] / ".env"
+            if env_path.exists():
+                load_dotenv(dotenv_path=env_path)
+
+        token = os.getenv("TOOL_SERVER_TOKEN") or None
+        max_concurrency = int(os.getenv("TOOL_SERVER_MAX_CONCURRENCY", "16"))
+        return cls(token=token, max_concurrency=max_concurrency)
+
+
+app = FastAPI(title="Atropos-Agent Tool Server")
+
+
+@app.get("/")
+async def root() -> Dict[str, str]:
+    return {"message": "Atropos-Agent Tool Server"}
+
+
+@app.on_event("startup")
+async def _startup() -> None:
+    cfg = ToolServerConfig.from_env()
+
+    # External-only registry. It will only include tools that are enabled by toolsets and
+    # whose Hermes requirements/keys are satisfied in this process.
+    tools: ToolRegistry = build_tool_registry(
+        enabled_toolsets=["all"],
+        disabled_toolsets=["terminal", "sandbox", "filesystem", "terminal_stateful", "default"],
+        tool_server_url="enabled",
+    )
+
+    app.state.cfg = cfg
+    app.state.tools = tools
+    app.state.semaphore = asyncio.Semaphore(cfg.max_concurrency)
+
+
+@app.get("/health")
+async def health() -> Dict[str, Any]:
+    return {"status": "ok"}
+
+
+@app.get("/tools")
+async def list_tools() -> Dict[str, Any]:
+    tools: ToolRegistry = app.state.tools
+    return {"tools": [s.to_dict() for s in tools.get_schemas()]}
+
+
+def _check_auth(cfg: ToolServerConfig, authorization: Optional[str]) -> None:
+    if not cfg.token:
+        return
+    if not authorization:
+        raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail="Missing Authorization header")
+    if not authorization.lower().startswith("bearer "):
+        raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail="Invalid Authorization header")
+    token = authorization.split(" ", 1)[1].strip()
+    if token != cfg.token:
+        raise HTTPException(status_code=status.HTTP_403_FORBIDDEN, detail="Invalid token")
+
+
+@app.post("/execute", response_model=ToolResultPayload)
+async def execute_tool(
+    req: ToolServerExecuteRequest,
+    authorization: Optional[str] = Header(default=None),
+) -> ToolResultPayload:
+    cfg: ToolServerConfig = app.state.cfg
+    _check_auth(cfg, authorization)
+
+    tools: ToolRegistry = app.state.tools
+    sem: asyncio.Semaphore = app.state.semaphore
+
+    tool = tools.get(req.tool.name)
+    if tool is None:
+        return ToolResultPayload(
+            success=False,
+            error=f"Unknown tool: {req.tool.name}",
+            uniq_id=req.tool.uniq_id,
+        )
+
+    async with sem:
+        try:
+            kwargs = dict(req.tool.arguments)
+            sig = inspect.signature(tool.execute).parameters
+            # Some tools can benefit from extra context.
+            if req.trajectory_id and "trajectory_id" in sig:
+                kwargs["trajectory_id"] = req.trajectory_id
+            if req.slot_id and "slot_id" in sig:
+                kwargs["slot_id"] = req.slot_id
+            if req.container_addr and "container_addr" in sig:
+                kwargs["container_addr"] = req.container_addr
+            if "task_id" in sig:
+                kwargs["task_id"] = req.trajectory_id
+            result = await tool.execute(**kwargs)
+        except Exception as e:
+            return ToolResultPayload(
+                success=False,
+                error=f"Tool execution error: {e}",
+                uniq_id=req.tool.uniq_id,
+            )
+
+    if result.uniq_id is None:
+        result.uniq_id = req.tool.uniq_id
+    return ToolResultPayload.from_tool_result(result)

atropos/backends/__init__.py

@@ -0,0 +1,27 @@
+from __future__ import annotations
+
+from typing import Any
+
+from .base import ToolBackend
+from .modal_backend import ModalSandboxConfig, ModalToolBackend
+from .nomad_backend import NomadBackendConfig, NomadToolBackend
+
+
+def create_tool_backend(cfg: Any) -> ToolBackend:
+    mode = str(getattr(cfg, "tool_pool_mode", "nomad")).strip().lower()
+    if mode == "nomad":
+        return NomadToolBackend(NomadBackendConfig.from_agent_env_config(cfg))
+    if mode == "modal":
+        return ModalToolBackend(ModalSandboxConfig.from_agent_env_config(cfg))
+    raise ValueError(f"Unknown tool_pool_mode: {mode}")
+
+
+__all__ = [
+    "ToolBackend",
+    "create_tool_backend",
+    "NomadBackendConfig",
+    "NomadToolBackend",
+    "ModalSandboxConfig",
+    "ModalToolBackend",
+]
+

atropos/backends/base.py

@@ -0,0 +1,89 @@
+"""
+Backend interfaces for AgentEnv tool execution.
+
+The goal of this module is to decouple ToolExecutor / AgentEnv from any single
+execution backend (Nomad/Docker today; Modal later).
+"""
+
+from __future__ import annotations
+
+from typing import Any, Dict, List, Optional, Protocol, Tuple
+
+from ..slots.executor import ExecutionResult
+from ..slots.slot import Slot
+
+
+class ToolBackend(Protocol):
+    """
+    Minimal interface required by ToolExecutor.
+
+    Backends provide:
+    - lifecycle (start/stop)
+    - slot acquisition/release (workspace affinity)
+    - batched tool execution across slots
+    - optional artifact helpers (for env verification / demos)
+    """
+
+    @property
+    def default_timeout_s(self) -> Optional[float]:
+        """Default sandbox execution timeout in seconds (if any)."""
+
+    async def start(self) -> None:
+        """Start the backend (provision workers/containers, health checks, etc)."""
+
+    async def stop(self, *, purge: bool = False) -> None:
+        """Stop the backend and optionally purge remote resources."""
+
+    async def acquire(self, trajectory_id: Optional[str] = None) -> Slot:
+        """Acquire a slot for a trajectory (workspace affinity)."""
+
+    async def release(self, slot: Slot, *, reset_workspace: bool = False) -> None:
+        """Release a slot back to the pool."""
+
+    async def execute_batch(
+        self,
+        requests: List[Tuple[Slot, str, Dict[str, Any]]],
+        *,
+        timeout_s: Optional[float] = None,
+    ) -> List[ExecutionResult]:
+        """Execute a batch of sandbox tool calls and return results in order."""
+
+    # ---------------------------------------------------------------------
+    # Optional artifact helpers (supported by the Nomad sandbox-server today)
+    # ---------------------------------------------------------------------
+
+    async def read_artifact(
+        self,
+        slot: Slot,
+        path: str,
+        *,
+        encoding: str = "text",
+        max_bytes: Optional[int] = None,
+        include_sha256: bool = False,
+        timeout_s: Optional[float] = None,
+    ) -> Dict[str, Any]:
+        raise NotImplementedError
+
+    async def list_artifacts(
+        self,
+        slot: Slot,
+        path: str = ".",
+        *,
+        recursive: bool = False,
+        max_entries: Optional[int] = None,
+        timeout_s: Optional[float] = None,
+    ) -> Dict[str, Any]:
+        raise NotImplementedError
+
+    async def archive_artifacts(
+        self,
+        slot: Slot,
+        path: str = ".",
+        *,
+        archive_format: str = "tar.gz",
+        max_bytes: Optional[int] = None,
+        max_entries: Optional[int] = None,
+        timeout_s: Optional[float] = None,
+    ) -> Dict[str, Any]:
+        raise NotImplementedError
+

atropos/backends/nomad_backend.py

@@ -0,0 +1,156 @@
+"""
+Nomad/Docker tool backend.
+
+This backend is the current default for AgentEnv: it provisions a Nomad job
+running `sandbox_server.py` and multiplexes stateless slots inside each container.
+"""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Any, Dict, List, Optional, Tuple
+
+from ..slots import Slot, SlotPool, SlotPoolConfig
+from ..slots.executor import ExecutionResult
+from .base import ToolBackend
+
+
+@dataclass(frozen=True)
+class NomadBackendConfig:
+    nomad_address: str
+    sandbox_job_id: str
+    sandbox_image: str
+    slots_per_container: int
+    min_containers: int
+    max_containers: int
+    privileged: bool
+    acquire_timeout_s: float
+    purge_job_on_start: bool
+    # Driver selection: "docker" or "singularity"
+    driver: str = "docker"
+    # Path to .sif file for singularity driver (required if driver="singularity")
+    singularity_image: Optional[str] = None
+
+    @classmethod
+    def from_agent_env_config(cls, cfg: Any) -> "NomadBackendConfig":
+        return cls(
+            nomad_address=str(getattr(cfg, "nomad_address")),
+            sandbox_job_id=str(getattr(cfg, "sandbox_job_id")),
+            sandbox_image=str(getattr(cfg, "sandbox_image")),
+            slots_per_container=int(getattr(cfg, "slots_per_container")),
+            min_containers=int(getattr(cfg, "min_containers")),
+            max_containers=int(getattr(cfg, "max_containers")),
+            privileged=bool(getattr(cfg, "privileged")),
+            acquire_timeout_s=float(getattr(cfg, "acquire_timeout_s")),
+            purge_job_on_start=bool(getattr(cfg, "purge_job_on_start", False)),
+            driver=str(getattr(cfg, "driver", "docker")),
+            singularity_image=getattr(cfg, "singularity_image", None),
+        )
+
+
+class NomadToolBackend(ToolBackend):
+    def __init__(self, config: NomadBackendConfig):
+        self.config = config
+        self.pool = SlotPool(
+            SlotPoolConfig(
+                nomad_address=config.nomad_address,
+                job_id=config.sandbox_job_id,
+                image=config.sandbox_image,
+                slots_per_container=config.slots_per_container,
+                min_containers=config.min_containers,
+                max_containers=config.max_containers,
+                privileged=config.privileged,
+                acquire_timeout=config.acquire_timeout_s,
+                purge_job_on_start=bool(config.purge_job_on_start),
+                driver=config.driver,
+                singularity_image=config.singularity_image,
+            )
+        )
+
+    @property
+    def default_timeout_s(self) -> Optional[float]:
+        t = getattr(self.pool.executor, "timeout", None)
+        total = getattr(t, "total", None)
+        try:
+            return float(total) if total is not None else None
+        except Exception:
+            return None
+
+    async def start(self) -> None:
+        await self.pool.start()
+
+    async def stop(self, *, purge: bool = False) -> None:
+        await self.pool.stop(purge_job=purge)
+
+    async def acquire(self, trajectory_id: Optional[str] = None) -> Slot:
+        return await self.pool.acquire(trajectory_id)
+
+    async def release(self, slot: Slot, *, reset_workspace: bool = False) -> None:
+        await self.pool.release(slot, reset_workspace=reset_workspace)
+
+    async def execute_batch(
+        self,
+        requests: List[Tuple[Slot, str, Dict[str, Any]]],
+        *,
+        timeout_s: Optional[float] = None,
+    ) -> List[ExecutionResult]:
+        return await self.pool.execute_batch(requests, timeout=timeout_s)
+
+    async def read_artifact(
+        self,
+        slot: Slot,
+        path: str,
+        *,
+        encoding: str = "text",
+        max_bytes: Optional[int] = None,
+        include_sha256: bool = False,
+        timeout_s: Optional[float] = None,
+    ) -> Dict[str, Any]:
+        return await self.pool.executor.read_artifact(
+            slot,
+            path,
+            encoding=encoding,
+            max_bytes=max_bytes,
+            include_sha256=include_sha256,
+            timeout=timeout_s,
+        )
+
+    async def list_artifacts(
+        self,
+        slot: Slot,
+        path: str = ".",
+        *,
+        recursive: bool = False,
+        max_entries: Optional[int] = None,
+        timeout_s: Optional[float] = None,
+    ) -> Dict[str, Any]:
+        return await self.pool.executor.list_artifacts(
+            slot,
+            path,
+            recursive=recursive,
+            max_entries=max_entries,
+            timeout=timeout_s,
+        )
+
+    async def archive_artifacts(
+        self,
+        slot: Slot,
+        path: str = ".",
+        *,
+        archive_format: str = "tar.gz",
+        max_bytes: Optional[int] = None,
+        max_entries: Optional[int] = None,
+        timeout_s: Optional[float] = None,
+    ) -> Dict[str, Any]:
+        return await self.pool.executor.archive_artifacts(
+            slot,
+            path,
+            archive_format=archive_format,
+            max_bytes=max_bytes,
+            max_entries=max_entries,
+            timeout=timeout_s,
+        )
+
+    def get_stats(self) -> Dict[str, Any]:
+        return self.pool.get_stats()
+

atropos/envs/__init__.py

@@ -0,0 +1,10 @@
+"""
+Environment implementations for atropos-agent.
+"""
+
+from .agent_env import AgentEnv, AgentEnvConfig
+
+# NOTE: Additional example envs exist as modules (e.g. `test_env`, `swe_smith_oracle_env`),
+# but are intentionally not imported here to avoid pulling heavy optional deps at import time.
+
+__all__ = ["AgentEnv", "AgentEnvConfig"]

atropos/envs/agent_env.py

@@ -0,0 +1,526 @@
+"""
+AgentEnv - Atropos BaseEnv extension for agent/tool-call workloads.
+
+AgentEnv is responsible for starting the sandbox tool execution backend and
+providing helpers for running agent trajectories with queued/batched tool calls.
+"""
+
+from __future__ import annotations
+import os
+import asyncio
+import time
+import uuid
+from abc import ABC, abstractmethod
+from typing import Any, Awaitable, Callable, Dict, Generic, List, Optional, Tuple, TypeVar
+
+from pydantic import Field
+
+from atroposlib.envs.base import APIServerConfig, BaseEnv, BaseEnvConfig, Item, ScoredDataGroup, ScoredDataItem
+from atroposlib.envs.server_handling.server_baseline import AsyncSemWithAdaptiveWeight
+
+from ..agent import AgentConfig, AgentResult, AtroposAgent
+from ..backends import ToolBackend, create_tool_backend
+from ..tools import ToolRegistry, build_tool_registry
+from ..tools.tool_executor import ToolExecutor, ToolExecutorConfig
+
+# Main BaseEnv child classes. Child class THESE to get agent+tooling functionality easily.
+
+class AgentEnvConfig(BaseEnvConfig):
+    tool_pool_mode: str = Field(default="nomad", description="Tool execution backend ('nomad' or 'modal')")
+
+    allow_network: bool = Field(
+        default=True,
+        description="Whether sandbox bash commands may access the network (env policy).",
+    )
+    require_sandbox: bool = Field(
+        default=False,
+        description="Fail closed if bubblewrap sandboxing is unavailable/unusable for stateless sandbox tools.",
+    )
+    require_stateful_sandbox: bool = Field(
+        default=False,
+        description="Fail closed if bubblewrap/PID isolation is unavailable for stateful terminal tools (tmux).",
+    )
+    tool_batch_window_ms: int = Field(default=20, description="ToolExecutor batching window (ms)")
+    tool_max_batch_size: int = Field(default=200, description="ToolExecutor maximum batch size")
+
+    # nomad mode settings. TODO: Add Modal support, split this into own config
+    nomad_address: str = Field(default="http://localhost:4646", description="Nomad API address")
+    sandbox_job_id: str = Field(default="atropos-sandbox-agent-env", description="Nomad job id for sandbox containers")
+    sandbox_image: str = Field(default="atropos-sandbox:local", description="Docker image for sandbox containers")
+    slots_per_container: int = Field(default=10, description="Nomad mode: slots per container")
+    min_containers: int = Field(default=1, description="Nomad mode: minimum containers")
+    max_containers: int = Field(default=10, description="Nomad mode: maximum containers")
+    privileged: bool = Field(default=False, description="Nomad mode: run container privileged")
+    acquire_timeout_s: float = Field(default=30.0, description="Slot acquisition timeout (seconds)")
+    purge_job_on_start: bool = Field(
+        default=False,
+        description=(
+            "Nomad mode: stop/purge the sandbox job on startup. This is helpful in local dev and training runs "
+            "to recover from previous crashes that leave the job in a restart backoff state."
+        ),
+    )
+    purge_job_on_shutdown: bool = Field(default=True, description="Nomad mode: stop/purge job on shutdown")
+    
+    # Nomad driver selection (docker or singularity)
+    driver: str = Field(
+        default="docker",
+        description="Nomad task driver: 'docker' (default) or 'singularity' (for HPC without sudo Docker)",
+    )
+    singularity_image: Optional[str] = Field(
+        default=None,
+        description="Path to .sif file for Singularity driver (required if driver='singularity')",
+    )
+
+    # modal mode settings (stub; implementation pending)
+    modal_app_name: str = Field(default="atropos-sandbox", description="Modal app name (stub)")
+    modal_function_name: str = Field(default="sandbox_server", description="Modal function/actor name (stub)")
+    modal_volume_name: Optional[str] = Field(default=None, description="Modal Volume name for persistent storage (stub)")
+    modal_volume_mount_path: str = Field(default="/data", description="Modal Volume mount path (stub)")
+
+    # basic agent defaults
+    agent_max_steps: int = Field(default=50, description="Max ReACT steps per trajectory")
+    agent_temperature: float = Field(default=0.7, description="Sampling temperature")
+    agent_max_tokens: Optional[int] = Field(
+        default=None,
+        description="Max tokens per model response (default: let backend decide)",
+    )
+    agent_tool_delay_s: float = Field(default=0.0, description="Delay between tool calls (seconds)")
+
+    # tool selection
+    enabled_toolsets: List[str] = Field(
+        default_factory=lambda: ["default"],
+        description="Toolsets to enable (Hermes-style grouping).",
+    )
+    disabled_toolsets: List[str] = Field(
+        default_factory=list,
+        description="Toolsets to disable (applied after enabled_toolsets).",
+    )
+
+    # external ToolServer routing (Phase 4.5+)
+    tool_server_url: Optional[str] = Field(
+        default=None,
+        description="Base URL for external ToolServer (enables external tools).",
+    )
+    tool_server_token: Optional[str] = Field(
+        default=None,
+        description="Bearer token for ToolServer auth (optional in dev).",
+    )
+
+AgentEnvConfigT = TypeVar("AgentEnvConfigT", bound="AgentEnvConfig")
+
+
+class AgentEnv(BaseEnv, ABC, Generic[AgentEnvConfigT]):
+    env_config_cls = AgentEnvConfig
+
+    def __init__(
+        self,
+        config: AgentEnvConfigT,
+        server_configs: List[APIServerConfig],
+        slurm: bool = False,
+        testing: bool = False,
+    ):
+        super().__init__(config, server_configs, slurm, testing)
+        self.config: AgentEnvConfigT = config
+
+        self.tools: ToolRegistry = self.build_tools()
+
+        self._backend: Optional[ToolBackend] = None
+        self._tool_executor: Optional[ToolExecutor] = None
+        self._tool_server_inprocess: bool = False
+        self._trajectory_workspace_meta: Dict[str, Dict[str, Any]] = {}
+
+    def build_tools(self) -> ToolRegistry:
+        """Wraps original Hermes-Agent ToolRegistry for atropos AgentEnv use.
+        See Hermes-Agent docs for toolsets and available tools etc.
+        """
+        return build_tool_registry(
+            enabled_toolsets=self.config.enabled_toolsets or ["default"],
+            disabled_toolsets=self.config.disabled_toolsets or None,
+            tool_server_url=self.config.tool_server_url,
+        )
+
+    @abstractmethod
+    def build_task(self, item: Item) -> str:
+        """Return the user-facing task string for the agent."""
+
+    @abstractmethod
+    async def score_trajectory(self, item: Item, final_response: str) -> float:
+        """Return a scalar score for this trajectory."""
+
+    async def setup_trajectory_workspace(
+        self,
+        item: Item,
+        *,
+        trajectory_id: str,
+        exec_tool: Callable[["ToolCall"], Awaitable["ToolResult"]],
+    ) -> Dict[str, Any]:
+        """
+        Optional hook: prepare the sandbox workspace before the agent starts.
+
+        Examples:
+        - clone a repo and checkout a commit
+        - write fixture files (e.g. images) for external-tool demos
+        - pre-install dependencies
+
+        Default: no-op.
+        """
+        _ = (item, trajectory_id, exec_tool)
+        return {}
+
+    async def verify_and_score_trajectory(
+        self,
+        item: Item,
+        final_response: str,
+        *,
+        trajectory_id: str,
+        exec_tool: Callable[["ToolCall"], Awaitable["ToolResult"]],
+        agent_result: Optional[AgentResult] = None,
+        workspace_meta: Optional[Dict[str, Any]] = None,
+    ) -> tuple[float, Dict[str, Any]]:
+        """
+        Optional hook: run in-sandbox verification before scoring.
+
+        Many agent envs need to execute verification inside the same trajectory
+        workspace (e.g. pytest) before releasing/resetting the slot.
+
+        Default: calls `score_trajectory()` and returns empty metadata.
+        """
+        _ = (trajectory_id, exec_tool, agent_result, workspace_meta)  # default ignores in-workspace verification
+        score = await self.score_trajectory(item, final_response)
+        return score, {}
+
+    def build_agent_config(self, item: Item) -> AgentConfig:  # noqa: ARG002
+        return AgentConfig(
+            max_steps=self.config.agent_max_steps,
+            temperature=self.config.agent_temperature,
+            max_tokens=self.config.agent_max_tokens,
+            tool_delay_s=self.config.agent_tool_delay_s,
+        )
+
+    async def setup(self) -> None:
+        print(f"[AgentEnv] setup(): starting tool backend ({self.config.tool_pool_mode})", flush=True)
+        await self._start_tool_backend()
+        print("[AgentEnv] setup(): configuring server concurrency", flush=True)
+        self._configure_server_concurrency()
+        print("[AgentEnv] setup(): running env-specific setup_agent_env()", flush=True)
+        await self.setup_agent_env()
+        print("[AgentEnv] setup(): done", flush=True)
+
+    def _configure_server_concurrency(self) -> None:
+        """
+        Ensure the LLM server concurrency isn't accidentally capped below `group_size`.
+
+        In `BaseEnv process` mode, groups are collected concurrently and if the underlying
+        ServerManager/OpenAIServer semaphore is left at 1, we serialize inference even
+        when `--env.group_size` is > 1.
+        """
+        desired = int(getattr(self.config, "group_size", 1) or 1)
+        if desired <= 1:
+            return
+
+        servers = getattr(self.server, "servers", None)
+        if not isinstance(servers, list) or not servers:
+            return
+
+        for s in servers:
+            sem = getattr(s, "sem", None)
+            eval_sem = getattr(s, "eval_sem", None)
+            # Only increase; never shrink.
+            if sem is not None and getattr(sem, "max_val", 0) < desired:
+                s.sem = AsyncSemWithAdaptiveWeight(desired)
+                if hasattr(s, "config") and hasattr(s.config, "num_max_requests_at_once"):
+                    s.config.num_max_requests_at_once = desired
+            if eval_sem is not None and getattr(eval_sem, "max_val", 0) < desired:
+                s.eval_sem = AsyncSemWithAdaptiveWeight(desired)
+                if hasattr(s, "config") and hasattr(s.config, "num_requests_for_eval"):
+                    s.config.num_requests_for_eval = desired
+
+    @abstractmethod
+    async def setup_agent_env(self) -> None:
+        """Subclass hook for env-specific setup."""
+
+    async def evaluate(self, *args, **kwargs):  # noqa: ARG002
+        """
+        Default eval hook (no-op).
+
+        Atropos BaseEnv requires an `evaluate()` implementation. Many agent envs
+        won't have a meaningful evaluation path during early PoC work; they can
+        override this when needed.
+        """
+        return {}
+
+    async def env_manager(self):
+        try:
+            return await super().env_manager()
+        finally:
+            await self.shutdown_tool_backend()
+
+    async def process_manager(self):
+        try:
+            return await super().process_manager()
+        finally:
+            await self.shutdown_tool_backend()
+
+    async def _start_tool_backend(self) -> None:
+        if self._tool_executor is not None:
+            return
+
+        tool_server_url = self.config.tool_server_url
+        tool_server_client = None
+        if tool_server_url == "inprocess":
+            import httpx
+            from ..api.tool_server import app as tool_server_app
+
+            await tool_server_app.router.startup()
+            tool_server_client = httpx.AsyncClient(
+                transport=httpx.ASGITransport(app=tool_server_app),
+                base_url="http://toolserver",
+            )
+            tool_server_url = "http://toolserver"
+            self._tool_server_inprocess = True
+
+        backend = create_tool_backend(self.config)
+        await backend.start()
+
+        executor = ToolExecutor(
+            backend=backend,
+            tools=self.tools,
+            config=ToolExecutorConfig(
+                batch_window_ms=self.config.tool_batch_window_ms,
+                max_batch_size=self.config.tool_max_batch_size,
+                allow_network=self.config.allow_network,
+                require_sandbox=self.config.require_sandbox,
+                require_stateful_sandbox=self.config.require_stateful_sandbox,
+                tool_server_url=tool_server_url,
+                tool_server_token=self.config.tool_server_token,
+            ),
+        )
+        await executor.start()
+        if tool_server_client is not None:
+            executor._tool_server_client = tool_server_client  # type: ignore[attr-defined]
+
+        self._backend = backend
+        self._tool_executor = executor
+
+    async def shutdown_tool_backend(self) -> None:
+        executor = self._tool_executor
+        backend = self._backend
+        inprocess_tool_server = self._tool_server_inprocess
+        self._tool_executor = None
+        self._backend = None
+        self._tool_server_inprocess = False
+
+        if executor is not None:
+            await executor.close()
+        if backend is not None:
+            await backend.stop(purge=bool(self.config.purge_job_on_shutdown))
+        if inprocess_tool_server:
+            from ..api.tool_server import app as tool_server_app
+
+            await tool_server_app.router.shutdown()
+
+    async def collect_trajectory(
+        self, item: Item
+    ) -> Tuple[Optional[ScoredDataItem], List[Item]]:
+        if self._tool_executor is None:
+            raise RuntimeError("Tool backend not started")
+
+        trajectory_id = str(uuid.uuid4())
+        t0 = time.perf_counter()
+        print(f"[AgentEnv] collect_trajectory(): tid={trajectory_id} start", flush=True)
+        task = self.build_task(item)
+        agent_config = self.build_agent_config(item)
+        if os.getenv("ATROPOS_DEBUG_PRINT_TASK") == "1":
+            print(f"Starting trajectory {trajectory_id} with task: {task}", flush=True)
+        else:
+            # Avoid printing the full task prompt by default (can be huge/noisy).
+            one_line = " ".join(str(task).splitlines()).strip()
+            preview = one_line[:240] + ("…" if len(one_line) > 240 else "")
+            print(f"Starting trajectory {trajectory_id} (task preview): {preview}", flush=True)
+
+        async def _exec(call):
+            return await self._tool_executor.execute(trajectory_id, call)
+
+        agent = AtroposAgent(
+            server=self.server,
+            tokenizer=self.tokenizer,
+            tools=self.tools,
+            config=agent_config,
+            execute_tool=_exec,
+        )
+
+        try:
+            print(f"[AgentEnv] tid={trajectory_id} setup_trajectory_workspace() start", flush=True)
+            workspace_meta = await self.setup_trajectory_workspace(item, trajectory_id=trajectory_id, exec_tool=_exec)
+            if not isinstance(workspace_meta, dict):
+                workspace_meta = {}
+            self._trajectory_workspace_meta[trajectory_id] = workspace_meta
+            print(
+                f"[AgentEnv] tid={trajectory_id} setup_trajectory_workspace() done in {time.perf_counter() - t0:.2f}s",
+                flush=True,
+            )
+
+            print(f"[AgentEnv] tid={trajectory_id} agent.run() start", flush=True)
+            result = await agent.run(task)
+            print(
+                f"[AgentEnv] tid={trajectory_id} agent.run() done in {time.perf_counter() - t0:.2f}s "
+                f"success={result.success} tool_calls={result.total_tool_calls}",
+                flush=True,
+            )
+            if not result.success or result.trajectory_data is None:
+                # Do not trigger BaseEnv retries for agent failures.
+                # Record the trajectory with score 0.0 so training/eval can see the failure mode.
+                messages = [{"role": "system", "content": agent._build_system_prompt()}]  # noqa: SLF001
+                messages.append({"role": "user", "content": task})
+                for step in result.steps:
+                    messages.append({"role": "assistant", "content": step.assistant_message})
+                    if step.tool_results:
+                        tool_text = "\n".join(r.to_xml() for r in step.tool_results)
+                        messages.append({"role": "user", "content": tool_text})
+
+                scored: ScoredDataItem = {
+                    "tokens": (result.trajectory_data.tokens if result.trajectory_data else []),
+                    "masks": (result.trajectory_data.masked_tokens if result.trajectory_data else []),
+                    "scores": 0.0,
+                }
+                if result.trajectory_data is not None:
+                    scored["inference_logprobs"] = result.trajectory_data.logprobs  # type: ignore[typeddict-unknown-key]
+                    if getattr(result.trajectory_data, "metadata", None):
+                        scored["overrides"] = {"managed_metadata": result.trajectory_data.metadata}
+                if self.config.include_messages:
+                    # Record a final failure marker as a user-side tool_response-like block so it survives templates.
+                    import json
+
+                    err = result.error or "agent_failed"
+                    messages.append(
+                        {
+                            "role": "user",
+                            "content": f"<tool_response>{json.dumps({'success': False, 'error': err})}</tool_response>",
+                        }
+                    )
+                    scored["messages"] = messages
+                return scored, []
+
+            print(f"[AgentEnv] tid={trajectory_id} verify_and_score_trajectory() start", flush=True)
+            score, score_metadata = await self.verify_and_score_trajectory(
+                item,
+                result.final_response,
+                trajectory_id=trajectory_id,
+                exec_tool=_exec,
+                agent_result=result,
+                workspace_meta=workspace_meta,
+            )
+            print(
+                f"[AgentEnv] tid={trajectory_id} verify_and_score_trajectory() done in {time.perf_counter() - t0:.2f}s "
+                f"score={score}",
+                flush=True,
+            )
+
+            messages = [{"role": "system", "content": agent._build_system_prompt()}]  # noqa: SLF001
+            messages.append({"role": "user", "content": task})
+            for step in result.steps:
+                messages.append({"role": "assistant", "content": step.assistant_message})
+                if step.tool_results:
+                    tool_text = "\n".join(r.to_xml() for r in step.tool_results)
+                    messages.append({"role": "user", "content": tool_text})
+
+            # Optional: allow env verification to attach additional messages (e.g. install logs).
+            if self.config.include_messages and isinstance(score_metadata, dict):
+                extra = score_metadata.get("verification_messages")
+                if isinstance(extra, list):
+                    for m in extra:
+                        if isinstance(m, dict) and isinstance(m.get("role"), str) and isinstance(m.get("content"), str):
+                            messages.append({"role": m["role"], "content": m["content"]})
+
+            scored: ScoredDataItem = {
+                "tokens": result.trajectory_data.tokens,
+                "masks": result.trajectory_data.masked_tokens,
+                "scores": score,
+            }
+            # Atroposlib expects policy logprobs at the *group* level under `inference_logprobs`.
+            # We stash per-item values here and lift them into the group in `collect_trajectories()`.
+            scored["inference_logprobs"] = result.trajectory_data.logprobs  # type: ignore[typeddict-unknown-key]
+            if getattr(result.trajectory_data, "metadata", None):
+                scored["overrides"] = {"managed_metadata": result.trajectory_data.metadata}
+            if self.config.include_messages:
+                scored["messages"] = messages
+
+            return scored, []
+        finally:
+            self._trajectory_workspace_meta.pop(trajectory_id, None)
+            print(f"[AgentEnv] tid={trajectory_id} release_trajectory(reset_workspace=True)", flush=True)
+            await self._tool_executor.release_trajectory(trajectory_id, reset_workspace=True)
+            print(f"[AgentEnv] collect_trajectory(): tid={trajectory_id} done in {time.perf_counter() - t0:.2f}s", flush=True)
+
+    async def collect_trajectories(
+        self, item: Item
+    ) -> Tuple[Optional[ScoredDataGroup], List[Item]]:
+        tasks = [self.collect_trajectory(item) for _ in range(self.config.group_size)]
+        results = await asyncio.gather(*tasks)
+
+        backlog: List[Item] = []
+        items: List[ScoredDataItem] = []
+        for scored, b in results:
+            backlog.extend(b)
+            if scored is not None:
+                items.append(scored)
+
+        if len(items) != self.config.group_size:
+            return None, backlog
+
+        group: ScoredDataGroup = ScoredDataGroup(
+            tokens=[],
+            masks=[],
+            scores=[],
+            advantages=[],
+            ref_logprobs=[],
+            messages=[] if self.config.include_messages else None,
+            inference_logprobs=[],
+            group_overrides={},
+            overrides=[],
+            images=[],
+            generation_params=None,
+        )
+
+        for it in items:
+            group["tokens"].append(it["tokens"])
+            group["masks"].append(it["masks"])
+            group["scores"].append(it["scores"])
+            # policy logprobs (for PPO/GRPO training) if present
+            lp = it.get("inference_logprobs")  # type: ignore[typeddict-item]
+            if lp is not None:
+                group["inference_logprobs"].append(lp)
+            group["overrides"].append(it.get("overrides") or {})  # type: ignore[typeddict-item]
+            if group.get("messages") is not None and it.get("messages") is not None:
+                group["messages"].append(it["messages"])
+
+        return group, backlog
+
+    async def run_agent(self, task: str, *, trajectory_id: Optional[str] = None) -> Tuple[str, Dict[str, Any]]:
+        """
+        Run the AtroposAgent on a single task and return (final_response, debug).
+
+        This is a helper intended for simple environments and tests.
+        """
+        if self._tool_executor is None:
+            raise RuntimeError("Tool backend not started")
+
+        tid = trajectory_id or str(uuid.uuid4())
+
+        async def _exec(call):
+            return await self._tool_executor.execute(tid, call)
+
+        agent = AtroposAgent(
+            server=self.server,
+            tokenizer=self.tokenizer,
+            tools=self.tools,
+            config=AgentConfig(
+                max_steps=self.config.agent_max_steps,
+                temperature=self.config.agent_temperature,
+                max_tokens=self.config.agent_max_tokens,
+            ),
+            execute_tool=_exec,
+        )
+        result = await agent.run(task)
+        await self._tool_executor.release_trajectory(tid, reset_workspace=True)
+        return result.final_response, {"success": result.success, "error": result.error, "tool_calls": result.total_tool_calls}

atropos/envs/hermes_compat_test_env.py

@@ -0,0 +1,171 @@
+"""
+Hermes-Agent + Atropos (Nomad sandbox) compatibility smoke environment.
+
+This environment is intended to validate, end-to-end:
+  BaseEnv.process -> AgentEnv -> ToolExecutor (batched) -> Nomad SlotPool -> sandbox_server
+
+It forces the model to use a sandbox tool by asking it to run a command that
+generates a high-entropy token inside the sandbox, then repeat it exactly.
+
+Run (process mode):
+  uv run python -m atropos.envs.hermes_compat_test_env process --env.use_wandb false --env.total_steps 2 --env.group_size 1
+"""
+
+from __future__ import annotations
+
+import os
+from typing import Any, Dict, List, Tuple
+
+from dotenv import load_dotenv
+from pydantic import Field
+
+from atroposlib.envs.base import APIServerConfig, Item
+
+from ..agent import AgentConfig, AgentResult
+from ..tools import ToolCall
+from .agent_env import AgentEnv, AgentEnvConfig
+
+load_dotenv()
+
+
+def _forced_tool_item() -> Item:
+    # Use double quotes in the shell command and show JSON escaping explicitly.
+    # This avoids invalid JSON escapes like `\\'` (not valid JSON) that some models produce.
+    cmd = 'python -c "import secrets; print(secrets.token_hex(16))"'
+    return {
+        "command": cmd,
+        "prompt": (
+            "You are acting as an agent inside a sandboxed environment.\n"
+            "You MUST use the terminal tool to execute commands.\n"
+            "Run this exact command:\n"
+            f"{cmd}\n"
+            "When you call the tool, use valid JSON inside <tool_call>. Example:\n"
+            '<tool_call>{"name": "terminal", "arguments": {"command": '
+            '"python -c \\\\"import secrets; print(secrets.token_hex(16))\\\\""}}'
+            "</tool_call>\n"
+            "Then respond with EXACTLY what it printed (the hex token) and nothing else.\n"
+            "Do not guess. Do not explain."
+        ),
+    }
+
+
+class HermesCompatTestEnvConfig(AgentEnvConfig):
+    server_base_url: str = Field(
+        default="http://127.0.0.1:8080",
+        description="Base URL for an OpenAI-compatible chat server (without /v1).",
+    )
+    server_model: str = Field(default="hermes-4-36b", description="Model name")
+    tokenizer_name: str = Field(default="NousResearch/Hermes-4.3-36B", description="Tokenizer name for RL tokenization")
+
+
+class HermesCompatTestEnv(AgentEnv[HermesCompatTestEnvConfig]):
+    name = "hermes_compat_test_env"
+    env_config_cls = HermesCompatTestEnvConfig
+
+    def __init__(
+        self,
+        config: HermesCompatTestEnvConfig,
+        server_configs: List[APIServerConfig],
+        slurm: bool = False,
+        testing: bool = False,
+    ):
+        super().__init__(config, server_configs, slurm, testing)
+        self._iter = 0
+
+    @classmethod
+    def config_init(cls) -> Tuple[HermesCompatTestEnvConfig, List[APIServerConfig]]:
+        base_url = (
+            os.getenv("ATROPOS_SERVER_BASE_URL")
+            or os.getenv("OPENAI_BASE_URL")
+            or os.getenv("LLM_BASE_URL")
+            or "http://127.0.0.1:8080"
+        )
+        model = os.getenv("ATROPOS_SERVER_MODEL") or os.getenv("LLM_MODEL") or "hermes-4-36b"
+        api_key = os.getenv("ATROPOS_SERVER_API_KEY") or os.getenv("NOUS_API_KEY") or os.getenv("OPENAI_API_KEY") or "local"
+
+        env_config = HermesCompatTestEnvConfig(
+            tokenizer_name=os.getenv("ATROPOS_TOKENIZER_NAME") or "NousResearch/Hermes-4.3-36B",
+            group_size=1,
+            use_wandb=False,
+            include_messages=True,
+            ensure_scores_are_not_same=False,
+            total_steps=2,
+            batch_size=1,
+            server_base_url=base_url,
+            server_model=model,
+            # Tooling: sandbox-only terminal.
+            enabled_toolsets=["terminal"],
+            disabled_toolsets=[],
+            # Default to Nomad sandboxing; users can override via --env.* args.
+            sandbox_image=os.getenv("ATROPOS_SANDBOX_IMAGE") or "atropos-sandbox:local",
+            # In local dev it's common for a previous crash to leave the job in backoff.
+            purge_job_on_start=True,
+            purge_job_on_shutdown=True,
+        )
+
+        server_configs = [
+            APIServerConfig(
+                model_name=model,
+                base_url=f"{base_url.rstrip('/')}/v1",
+                api_key=api_key,
+                num_max_requests_at_once=1,
+                num_requests_for_eval=1,
+                timeout=120,
+            )
+        ]
+        return env_config, server_configs
+
+    async def setup_agent_env(self) -> None:
+        return None
+
+    async def get_next_item(self) -> Item:
+        self._iter += 1
+        return _forced_tool_item()
+
+    def build_task(self, item: Item) -> str:
+        return str(item.get("prompt") or "")
+
+    def build_agent_config(self, item: Item) -> AgentConfig:  # noqa: ARG002
+        # Avoid imposing max_tokens by default; tool-tag responses can be long for some models.
+        return AgentConfig(
+            max_steps=min(8, int(self.config.agent_max_steps)),
+            temperature=0.2,
+            max_tokens=None,
+        )
+
+    async def score_trajectory(self, item: Item, final_response: str) -> float:
+        # Scoring happens in verify_and_score_trajectory so we can inspect tool results.
+        _ = (item, final_response)
+        return 0.0
+
+    async def verify_and_score_trajectory(
+        self,
+        item: Item,
+        final_response: str,
+        *,
+        trajectory_id: str,  # noqa: ARG002
+        exec_tool,  # noqa: ARG002
+        agent_result: AgentResult | None = None,
+        workspace_meta: Dict[str, Any] | None = None,  # noqa: ARG002
+    ) -> tuple[float, Dict[str, Any]]:
+        if agent_result is None:
+            return 0.0, {"error": "Missing agent_result"}
+
+        observed: str = ""
+        tool_ok = False
+        for step in agent_result.steps:
+            for res in step.tool_results:
+                if not res.success:
+                    return 0.0, {"error": res.error, "output": res.output}
+                out = (res.output or "").strip()
+                if out:
+                    observed = out.splitlines()[-1].strip()
+                    tool_ok = True
+
+        final = (final_response or "").strip()
+        score = 1.0 if tool_ok and agent_result.total_tool_calls > 0 and observed and final == observed else 0.0
+        return score, {"observed": observed, "tool_calls": agent_result.total_tool_calls, "command": item.get("command")}
+
+
+if __name__ == "__main__":
+    HermesCompatTestEnv.cli()

atropos/envs/sandbox_terminal_smoke_env.py

@@ -0,0 +1,172 @@
+"""
+Nomad sandbox terminal smoke environment (training-oriented).
+
+Validates, end-to-end:
+  BaseEnv.process -> AgentEnv -> ToolExecutor (batched) -> Nomad SlotPool -> sandbox_server
+
+It forces the model to use a sandbox tool by asking it to run a command that
+generates a high-entropy token inside the sandbox, then repeat it exactly.
+
+Run (process mode):
+  uv run python -m atropos.envs.sandbox_terminal_smoke_env process --env.use_wandb false --env.total_steps 2 --env.group_size 1
+"""
+
+from __future__ import annotations
+
+import os
+from typing import Any, Dict, List, Tuple
+
+from dotenv import load_dotenv
+from pydantic import Field
+
+from atroposlib.envs.base import APIServerConfig, Item
+
+from ..agent import AgentConfig, AgentResult
+from ..tools import ToolCall
+from .agent_env import AgentEnv, AgentEnvConfig
+
+load_dotenv()
+
+STRICT_TOOLCALL_SYSTEM_PROMPT = None
+
+
+def _forced_tool_item() -> Item:
+    # Use double quotes in the shell command and show JSON escaping explicitly.
+    # This avoids invalid JSON escapes like `\\'` (not valid JSON) that some models produce.
+    cmd = 'python -c "import secrets; print(secrets.token_hex(16))"'
+    return {
+        "command": cmd,
+        "prompt": (
+            "You MUST use the terminal tool.\n"
+            "Run this exact command:\n"
+            f"{cmd}\n"
+            "When you call the tool, use valid JSON inside <tool_call>. Example:\n"
+            '<tool_call>{"name": "terminal", "arguments": {"command": '
+            '"python -c \\\\"import secrets; print(secrets.token_hex(16))\\\\""}}'
+            "</tool_call>\n"
+            "Then respond with EXACTLY what it printed (the hex token) and nothing else.\n"
+            "Do not guess. Do not explain."
+        ),
+    }
+
+
+class SandboxTerminalSmokeEnvConfig(AgentEnvConfig):
+    server_base_url: str = Field(
+        default="http://127.0.0.1:8080",
+        description="Base URL for an OpenAI-compatible chat server (without /v1).",
+    )
+    server_model: str = Field(default="hermes-4-36b", description="Model name")
+    tokenizer_name: str = Field(default="NousResearch/Hermes-4.3-36B", description="Tokenizer name for RL tokenization")
+
+
+class SandboxTerminalSmokeEnv(AgentEnv[SandboxTerminalSmokeEnvConfig]):
+    name = "sandbox_terminal_smoke_env"
+    env_config_cls = SandboxTerminalSmokeEnvConfig
+
+    def __init__(
+        self,
+        config: SandboxTerminalSmokeEnvConfig,
+        server_configs: List[APIServerConfig],
+        slurm: bool = False,
+        testing: bool = False,
+    ):
+        super().__init__(config, server_configs, slurm, testing)
+        self._iter = 0
+
+    @classmethod
+    def config_init(cls) -> Tuple[SandboxTerminalSmokeEnvConfig, List[APIServerConfig]]:
+        base_url = (
+            os.getenv("ATROPOS_SERVER_BASE_URL")
+            or os.getenv("OPENAI_BASE_URL")
+            or os.getenv("LLM_BASE_URL")
+            or "http://127.0.0.1:8080"
+        )
+        model = os.getenv("ATROPOS_SERVER_MODEL") or os.getenv("LLM_MODEL") or "hermes-4-36b"
+        api_key = os.getenv("ATROPOS_SERVER_API_KEY") or os.getenv("NOUS_API_KEY") or os.getenv("OPENAI_API_KEY") or "local"
+
+        env_config = SandboxTerminalSmokeEnvConfig(
+            tokenizer_name=os.getenv("ATROPOS_TOKENIZER_NAME") or "NousResearch/Hermes-4.3-36B",
+            group_size=1,
+            use_wandb=False,
+            include_messages=True,
+            ensure_scores_are_not_same=False,
+            total_steps=2,
+            batch_size=1,
+            server_base_url=base_url,
+            server_model=model,
+            # Tooling: sandbox-only terminal.
+            enabled_toolsets=["terminal"],
+            disabled_toolsets=[],
+            # Default to Nomad sandboxing; users can override via --env.* args.
+            sandbox_image=os.getenv("ATROPOS_SANDBOX_IMAGE") or "atropos-sandbox:local",
+            purge_job_on_start=True,
+            purge_job_on_shutdown=True,
+        )
+
+        server_configs = [
+            APIServerConfig(
+                model_name=model,
+                base_url=f"{base_url.rstrip('/')}/v1",
+                api_key=api_key,
+                num_max_requests_at_once=1,
+                num_requests_for_eval=1,
+                timeout=120,
+            )
+        ]
+        return env_config, server_configs
+
+    async def setup_agent_env(self) -> None:
+        return None
+
+    async def get_next_item(self) -> Item:
+        self._iter += 1
+        return _forced_tool_item()
+
+    def build_task(self, item: Item) -> str:
+        return str(item.get("prompt") or "")
+
+    def build_agent_config(self, item: Item) -> AgentConfig:  # noqa: ARG002
+        # Avoid imposing max_tokens by default; tool-tag responses can be long for some models.
+        return AgentConfig(
+            max_steps=min(8, int(self.config.agent_max_steps)),
+            temperature=0.2,
+            max_tokens=None,
+            system_prompt=STRICT_TOOLCALL_SYSTEM_PROMPT,
+        )
+
+    async def score_trajectory(self, item: Item, final_response: str) -> float:
+        # Scoring happens in verify_and_score_trajectory so we can inspect tool results.
+        _ = (item, final_response)
+        return 0.0
+
+    async def verify_and_score_trajectory(
+        self,
+        item: Item,
+        final_response: str,
+        *,
+        trajectory_id: str,  # noqa: ARG002
+        exec_tool,  # noqa: ARG002
+        agent_result: AgentResult | None = None,
+        workspace_meta: Dict[str, Any] | None = None,  # noqa: ARG002
+    ) -> tuple[float, Dict[str, Any]]:
+        if agent_result is None:
+            return 0.0, {"error": "Missing agent_result"}
+
+        observed: str = ""
+        tool_ok = False
+        for step in agent_result.steps:
+            for res in step.tool_results:
+                if not res.success:
+                    return 0.0, {"error": res.error, "output": res.output}
+                out = (res.output or "").strip()
+                if out:
+                    observed = out.splitlines()[-1].strip()
+                    tool_ok = True
+
+        final = (final_response or "").strip()
+        score = 1.0 if tool_ok and agent_result.total_tool_calls > 0 and observed and final == observed else 0.0
+        return score, {"observed": observed, "tool_calls": agent_result.total_tool_calls, "command": item.get("command")}
+
+
+if __name__ == "__main__":
+    SandboxTerminalSmokeEnv.cli()

atropos/envs/swe_smith_oracle_env.py

@@ -0,0 +1,418 @@
+"""
+SWE-smith-oracle environment.
+
+This environment is intentionally minimal:
+- prepares a sandbox workspace by cloning a public GitHub repo at `base_commit`
+- runs an AtroposAgent tool loop to apply a fix
+- verifies by running pytest nodeids from the dataset (reward = pass/fail)
+- Python only (no multi-language support currently, need to properly bauild & add to dropbox)
+- TODO: Get the other nonpython sandboxes up and running, then add a config knob to switch between them per row
+- oh and add to dockerhub
+
+Dataset: NousResearch/SWE-smith-oracle (train; does NOT use SWE-bench eval set).
+"""
+
+from __future__ import annotations
+
+import os
+import random
+import time
+from typing import Any, Dict, List, Optional, Tuple
+
+from pydantic import Field
+
+from atroposlib.envs.base import APIServerConfig, Item
+
+from ..agent import AgentConfig
+from ..tools import ToolCall
+from .agent_env import AgentEnv, AgentEnvConfig
+
+
+class SweSmithOracleEnvConfig(AgentEnvConfig):
+    dataset_name: str = Field(default="NousResearch/SWE-smith-oracle")
+    dataset_split: str = Field(default="train")
+    max_items: int = Field(default=0, description="0 = no limit")
+    shuffle: bool = Field(default=True)
+    seed: int = Field(default=0)
+
+    python_only: bool = Field(default=True, description="Filter to Python-evaluable rows")
+    score_include_fail_to_pass: bool = Field(
+        default=True,
+        description=(
+            "If true (default), score tests on PASS_TO_PASS ∪ FAIL_TO_PASS. "
+            "Disable to only run PASS_TO_PASS (faster but weaker signal)."
+        ),
+    )
+
+    prompt_mode: str = Field(
+        default="problem_statement",
+        description="Task prompt content: 'problem_statement' (fast) or 'problem_statement+text' (slower, includes dataset 'text').",
+    )
+
+    repo_base_url: str = Field(default="https://github.com", description="Base URL for repo cloning")
+    install_timeout_s: float = Field(default=600.0)
+    test_timeout_s: float = Field(default=600.0)
+
+    tokenizer_name: str = Field(default="NousResearch/Hermes-4.3-36B", description="Tokenizer name for RL tokenization")
+
+
+class SweSmithOracleEnv(AgentEnv[SweSmithOracleEnvConfig]):
+    """
+    SWE-smith-oracle AgentEnv.
+
+    This is designed for benchmarking multiplexed slot execution vs naive container-per-trajectory.
+    """
+
+    name = "swe_smith_oracle_env"
+    env_config_cls = SweSmithOracleEnvConfig
+
+    def __init__(
+        self,
+        config: SweSmithOracleEnvConfig,
+        server_configs: List[APIServerConfig],
+        slurm: bool = False,
+        testing: bool = False,
+    ):
+        super().__init__(config, server_configs, slurm, testing)
+        self._dataset = None
+        self._indices: List[int] = []
+        self._cursor = 0
+
+    @classmethod
+    def config_init(cls) -> Tuple[SweSmithOracleEnvConfig, List[APIServerConfig]]:
+        # Defaults for running the env via CLI in offline `process` mode.
+        # Override via env vars or `--env.*` flags as needed.
+        base_url_raw = (
+            os.getenv("ATROPOS_SERVER_BASE_URL")
+            or os.getenv("OPENAI_BASE_URL")
+            or os.getenv("LLM_BASE_URL")
+            or "http://127.0.0.1:8080"
+        )
+        base_url = base_url_raw.rstrip("/")
+        if not base_url.endswith("/v1"):
+            base_url = f"{base_url}/v1"
+        model = os.getenv("ATROPOS_SERVER_MODEL") or os.getenv("LLM_MODEL") or "hermes-4-36b"
+        api_key = os.getenv("ATROPOS_SERVER_API_KEY") or os.getenv("NOUS_API_KEY") or os.getenv("OPENAI_API_KEY") or "local"
+
+        env_config = SweSmithOracleEnvConfig(
+            tokenizer_name=os.getenv("ATROPOS_TOKENIZER_NAME") or "NousResearch/Hermes-4.3-36B",
+            group_size=1,
+            use_wandb=False,
+            rollout_server_url="http://localhost:8000",
+            total_steps=1,
+            batch_size=1,
+            steps_per_eval=1,
+            max_token_length=8192,
+            inference_weight=1.0,
+            wandb_name="swe_smith_oracle",
+            enabled_toolsets=["terminal"],
+            disabled_toolsets=[],
+            sandbox_image=os.getenv("ATROPOS_SANDBOX_IMAGE") or "atropos-sandbox:local",
+            purge_job_on_start=True,
+            purge_job_on_shutdown=True,
+        )
+
+        server_configs = [
+            APIServerConfig(
+                model_name=model,
+                base_url=base_url,
+                api_key=api_key,
+                num_max_requests_at_once=1,
+                num_requests_for_eval=1,
+                timeout=int(os.getenv("ATROPOS_SERVER_TIMEOUT_S") or "300"),
+            ),
+        ]
+
+        return env_config, server_configs
+
+    async def setup_agent_env(self) -> None:
+        from datasets import load_dataset
+
+        t0 = time.perf_counter()
+        print(
+            f"[SweSmithOracleEnv] loading dataset {self.config.dataset_name}:{self.config.dataset_split} "
+            f"(python_only={self.config.python_only}, max_items={self.config.max_items or 'all'})",
+            flush=True,
+        )
+        ds = load_dataset(self.config.dataset_name, split=self.config.dataset_split)
+        self._dataset = ds
+
+        indices: List[int] = []
+        for idx in range(len(ds)):
+            row = ds[idx]
+            if self.config.python_only and not self._is_python_row(row):
+                continue
+            indices.append(idx)
+
+        if self.config.shuffle:
+            rnd = random.Random(self.config.seed)
+            rnd.shuffle(indices)
+
+        if self.config.max_items and self.config.max_items > 0:
+            indices = indices[: self.config.max_items]
+
+        self._indices = indices
+        self._cursor = 0
+
+        print(
+            f"[SweSmithOracleEnv] loaded {len(self._indices)} items from {self.config.dataset_name}:{self.config.dataset_split} "
+            f"in {time.perf_counter() - t0:.2f}s",
+            flush=True,
+        )
+
+    def _is_python_row(self, row: Dict[str, Any]) -> bool:
+        nodeids = row.get("PASS_TO_PASS")
+        if not isinstance(nodeids, list) or not nodeids:
+            return False
+        for nid in nodeids:
+            if not isinstance(nid, str) or ".py::" not in nid:
+                return False
+        return True
+
+    async def get_next_item(self) -> Item:
+        print(f"[SweSmithOracleEnv] get_next_item() cursor={self._cursor}/{len(self._indices)}", flush=True)
+        if not self._dataset or not self._indices:
+            raise RuntimeError("Dataset not initialized (did setup() run?)")
+        if self._cursor >= len(self._indices):
+            self._cursor = 0
+        idx = self._indices[self._cursor]
+        self._cursor += 1
+        return dict(self._dataset[idx])
+
+    def _repo_name(self, item: Item) -> str:
+        repo = item.get("repo") or ""
+        if isinstance(repo, str) and "/" in repo:
+            return repo.split("/")[-1]
+        return "repo"
+
+    def build_task(self, item: Item) -> str:
+        repo = item.get("repo") or ""
+        base_commit = item.get("base_commit") or ""
+        problem = str(item.get("problem_statement") or "")
+        context = str(item.get("text") or "")
+
+        nodeids = self._tests_for_item(item)
+        tests_list = "\n".join(f"- {t}" for t in nodeids)
+
+        repo_dir = self._repo_name(item)
+
+        tests_block = (
+            "Run these tests to verify:\n"
+            f"{tests_list}\n\n"
+            "When done, briefly describe what you changed and confirm tests pass."
+        )
+
+        prompt_mode = (self.config.prompt_mode or "problem_statement").strip().lower()
+        if prompt_mode not in {"problem_statement", "problem_statement+text"}:
+            raise ValueError(
+                f"Invalid prompt_mode={self.config.prompt_mode!r}. "
+                "Expected 'problem_statement' or 'problem_statement+text'."
+            )
+
+        context_block = ""
+        if prompt_mode == "problem_statement+text" and context:
+            # Note: We intentionally do NOT truncate/cap here. This mode is for debugging / richer prompts and can be slow.
+            context_block = f"\nAdditional context:\n{context}\n"
+
+        return (
+            "You are a senior software engineer. Fix the repository so the specified tests pass.\n\n"
+            f"Repository: {repo} (checked out at base_commit={base_commit})\n"
+            f"Workspace path: ./{repo_dir}\n\n"
+            "Constraints:\n"
+            "- You MUST use the terminal tool to inspect, edit, and verify the repository. Do not respond with a patch file.\n"
+            f"- Start by inspecting the repo (e.g. `ls`, `cd ./{repo_dir}`, `git status`).\n"
+            "- Use a workspace-local virtualenv (e.g. inside the repo at ./.venv) to avoid cross-run contamination.\n"
+            "- Use non-interactive commands only.\n\n"
+            "- Terminal commands run under POSIX /bin/sh and each tool call runs in a fresh shell (no persisted env vars).\n"
+            "  Avoid bash-only `source`; prefer `. .venv/bin/activate` or `.venv/bin/python ...`.\n\n"
+            "Problem statement:\n"
+            f"{problem}\n\n"
+            f"{context_block}\n"
+            f"{tests_block}"
+        )
+
+    def build_agent_config(self, item: Item) -> AgentConfig:  # noqa: ARG002
+        # SWE tasks are longer than the simple test env.
+        return AgentConfig(
+            max_steps=self.config.agent_max_steps,
+            temperature=self.config.agent_temperature,
+            max_tokens=self.config.agent_max_tokens,
+            tool_delay_s=self.config.agent_tool_delay_s,
+        )
+
+    async def setup_trajectory_workspace(self, item: Item, *, trajectory_id: str, exec_tool) -> Dict[str, Any]:
+        t0 = time.perf_counter()
+        repo = item.get("repo")
+        base_commit = item.get("base_commit")
+        instance_id = item.get("instance_id") or item.get("id") or item.get("problem_id")
+        if not isinstance(repo, str) or not isinstance(base_commit, str):
+            raise RuntimeError("Invalid dataset row: missing repo/base_commit")
+
+        repo_dir = self._repo_name(item)
+        clone_url = f"{self.config.repo_base_url.rstrip('/')}/{repo}.git"
+        print(
+            f"[SweSmithOracleEnv] tid={trajectory_id} setup_trajectory_workspace(): "
+            f"repo={repo} base_commit={base_commit} instance_id={instance_id} dir=./{repo_dir}",
+            flush=True,
+        )
+
+        # Repo setup strategy:
+        # - Maintain a shared, per-container bare repo cache under /data/repo_cache
+        # - For each trajectory, create an isolated git worktree under the slot workspace
+        # This avoids cloning/fetching full repos per trajectory and is crucial for multiplexing.
+
+        def _repo_cache_slug(repo_name: str) -> str:
+            return repo_name.replace("/", "__")
+
+        repo_slug = _repo_cache_slug(repo)
+        cache_root = "/data/repo_cache"
+        bare_repo = f"{cache_root}/{repo_slug}.git"
+        lock_file = f"{cache_root}/.locks/{repo_slug}.lock"
+
+        # Use flock to serialize operations that mutate the shared bare repo (fetch/worktree).
+        # util-linux (flock) is included in the sandbox image.
+        worktree_cmd = (
+            "set -e; "
+            f"rm -rf {repo_dir}; "
+            f"mkdir -p {cache_root}/.locks; "
+            f": > {lock_file}; "
+            f"flock -x {lock_file} sh -lc '"
+            f"set -e; "
+            "export GIT_TERMINAL_PROMPT=0; "
+            "export GIT_LFS_SKIP_SMUDGE=1; "
+            f"if [ ! -d \"{bare_repo}\" ]; then "
+            f"  git init --bare \"{bare_repo}\"; "
+            f"  git -C \"{bare_repo}\" remote add origin \"{clone_url}\"; "
+            "fi; "
+            f"git -C \"{bare_repo}\" remote set-url origin \"{clone_url}\"; "
+            f"git -C \"{bare_repo}\" worktree prune || true; "
+            f"if ! git -C \"{bare_repo}\" cat-file -e \"{base_commit}^{{commit}}\" 2>/dev/null; then "
+            f"  git -C \"{bare_repo}\" fetch --depth 1 origin \"{base_commit}\" || true; "
+            "fi; "
+            f"if ! git -C \"{bare_repo}\" cat-file -e \"{base_commit}^{{commit}}\" 2>/dev/null; then "
+            f"  git -C \"{bare_repo}\" fetch --prune origin; "
+            "fi; "
+            f"git --git-dir=\"{bare_repo}\" worktree add --detach \"{repo_dir}\" \"{base_commit}\"; "
+            "'"
+        )
+
+        print(f"[SweSmithOracleEnv] tid={trajectory_id} preparing worktree from repo cache", flush=True)
+        res = await exec_tool(
+            ToolCall(
+                name="terminal",
+                arguments={"command": worktree_cmd, "timeout": self.config.install_timeout_s},
+            )
+        )
+        if not res.success:
+            raise RuntimeError(
+                "git worktree setup failed "
+                f"(repo={repo}, base_commit={base_commit}, instance_id={instance_id}): {res.error}\n{res.output}"
+            )
+
+        print(
+            f"[SweSmithOracleEnv] tid={trajectory_id} setup_trajectory_workspace(): worktree ready in {time.perf_counter() - t0:.2f}s",
+            flush=True,
+        )
+        return {"repo_dir": repo_dir, "base_commit": base_commit}
+
+    def _tests_for_item(self, item: Item) -> List[str]:
+        tests: List[str] = []
+        if self.config.score_include_fail_to_pass:
+            for key in ("PASS_TO_PASS", "FAIL_TO_PASS"):
+                nodeids = item.get(key)
+                if isinstance(nodeids, list):
+                    tests.extend([n for n in nodeids if isinstance(n, str)])
+        else:
+            nodeids = item.get("PASS_TO_PASS")
+            if isinstance(nodeids, list):
+                tests.extend([n for n in nodeids if isinstance(n, str)])
+        # Stable order for reproducibility.
+        return sorted(dict.fromkeys(tests))
+
+    def _chunk_nodeids(self, nodeids: List[str], max_per_chunk: int = 50) -> List[List[str]]:
+        chunks: List[List[str]] = []
+        for i in range(0, len(nodeids), max_per_chunk):
+            chunks.append(nodeids[i : i + max_per_chunk])
+        return chunks
+
+    async def verify_and_score_trajectory(
+        self,
+        item: Item,
+        final_response: str,  # noqa: ARG002
+        *,
+        trajectory_id: str,
+        exec_tool,
+        agent_result=None,
+        workspace_meta: Optional[Dict[str, Any]] = None,
+    ) -> tuple[float, Dict[str, Any]]:
+        _ = trajectory_id
+        repo_dir = self._repo_name(item)
+
+        # Training correctness: do not reward trajectories that never actually used tools.
+        if agent_result is not None and getattr(agent_result, "total_tool_calls", 0) <= 0:
+            print(
+                f"[SweSmithOracleEnv] tid={trajectory_id} verify (dataset_tests): no tool calls; score=0.0",
+                flush=True,
+            )
+            return 0.0, {
+                "verification_mode": "dataset_tests",
+                "error": "No tool calls were made by the agent",
+            }
+
+        nodeids = self._tests_for_item(item)
+        if not nodeids:
+            return 0.0, {"error": "No tests provided"}
+
+        print(f"[SweSmithOracleEnv] tid={trajectory_id} verify (dataset_tests): ensuring venv + deps", flush=True)
+        setup_cmd = (
+            f"cd {repo_dir} && "
+            "python -m venv .venv && "
+            ". .venv/bin/activate && "
+            "python -m pip install -U pip setuptools wheel && "
+            "python -m pip install -e . && "
+            "python -m pip install pytest"
+        )
+        setup_res = await exec_tool(
+            ToolCall(name="terminal", arguments={"command": setup_cmd, "timeout": self.config.install_timeout_s})
+        )
+        verification_messages = [{"role": "user", "content": setup_res.to_xml()}]
+        if not setup_res.success:
+            return 0.0, {
+                "verification_mode": "dataset_tests",
+                "phase": "install",
+                "error": setup_res.error,
+                "output": setup_res.output,
+                "verification_messages": verification_messages,
+            }
+
+        chunks = self._chunk_nodeids(nodeids, max_per_chunk=50)
+        for chunk_idx, chunk in enumerate(chunks):
+            joined = " ".join(chunk)
+            cmd = f"cd {repo_dir} && . .venv/bin/activate && python -m pytest -q {joined}"
+            res = await exec_tool(
+                ToolCall(
+                    name="terminal",
+                    arguments={"command": cmd, "timeout": self.config.test_timeout_s},
+                )
+            )
+            verification_messages.append({"role": "user", "content": res.to_xml()})
+            if not res.success:
+                return 0.0, {
+                    "verification_mode": "dataset_tests",
+                    "phase": "pytest",
+                    "failed_chunk": chunk_idx,
+                    "error": res.error,
+                    "output": res.output,
+                    "verification_messages": verification_messages,
+                }
+
+        return 1.0, {"verification_mode": "dataset_tests", "passed": True, "verification_messages": verification_messages}
+
+    async def score_trajectory(self, item: Item, final_response: str) -> float:
+        # Not used; scoring happens in verify_and_score_trajectory.
+        _ = (item, final_response)
+        return 0.0
+
+
+if __name__ == "__main__":
+    SweSmithOracleEnv.cli()

atropos/envs/test_env.py

@@ -0,0 +1,217 @@
+"""
+Simple test environment for validating the atropos-agent setup.
+
+This environment uses a local OpenAI-compatible server for LLM testing to verify:
+- BaseEnv extension works correctly
+- API communication via OpenAI-compatible endpoint
+- Basic trajectory collection
+
+This is a minimal environment for testing, not production use.
+"""
+
+import os
+from typing import Dict, List, Optional, Tuple
+
+from dotenv import load_dotenv
+from pydantic import Field
+
+from atroposlib.envs.base import (
+    APIServerConfig,
+    Item,
+)
+
+from ..agent import AgentConfig
+from .agent_env import AgentEnv, AgentEnvConfig
+
+# Load environment variables from .env file
+load_dotenv()
+
+
+# Simple test prompts for validation
+TEST_PROMPTS = [
+    {
+        "prompt": "What is 2 + 2? Answer with just the number.",
+        "expected": "4",
+    },
+    {
+        "prompt": "What is the capital of France? Answer with just the city name.",
+        "expected": "Paris",
+    },
+    {
+        "prompt": "What color is the sky on a clear day? Answer with just the color.",
+        "expected": "Blue",
+    },
+    {
+        "prompt": "How many days are in a week? Answer with just the number.",
+        "expected": "7",
+    },
+    {
+        "prompt": "What is 10 * 5? Answer with just the number.",
+        "expected": "50",
+    },
+]
+
+SYSTEM_PROMPT = (
+    "You are a helpful assistant. Answer questions concisely and directly. "
+    "When asked for a simple answer, provide just that answer without explanation."
+)
+
+
+class SimpleTestEnvConfig(AgentEnvConfig):
+    """Configuration for the simple test environment."""
+
+    server_base_url: str = Field(
+        default="http://127.0.0.1:8080",
+        description="Base URL for an OpenAI-compatible server (without /v1)",
+    )
+    server_model: str = Field(
+        default="hermes-4-36b",
+        description="Model name",
+    )
+    tokenizer_name: str = Field(default="NousResearch/Hermes-4.3-36B", description="Tokenizer name for RL tokenization")
+
+
+class SimpleTestEnv(AgentEnv[SimpleTestEnvConfig]):
+    """
+    A simple test environment to validate the atropos-agent setup.
+    
+    Uses a local OpenAI-compatible LLM endpoint with basic question-answering tasks.
+    Scoring is based on whether the response contains the expected answer.
+    """
+
+    name = "simple_test_env"
+    env_config_cls = SimpleTestEnvConfig
+
+    def __init__(
+        self,
+        config: SimpleTestEnvConfig,
+        server_configs: List[APIServerConfig],
+        slurm: bool = False,
+        testing: bool = False,
+    ):
+        super().__init__(config, server_configs, slurm, testing)
+        self.iter = 0
+        self.test_prompts = TEST_PROMPTS
+        self.percent_correct_buffer: List[float] = []
+
+    @classmethod
+    def config_init(cls) -> Tuple[SimpleTestEnvConfig, List[APIServerConfig]]:
+        """
+        Initialize configuration with local server settings from environment variables.
+        """
+        base_url = (
+            os.getenv("ATROPOS_SERVER_BASE_URL")
+            or os.getenv("OPENAI_BASE_URL")
+            or os.getenv("LLM_BASE_URL")
+            or "http://127.0.0.1:8080"
+        )
+        model = os.getenv("ATROPOS_SERVER_MODEL") or os.getenv("LLM_MODEL") or "hermes-4-36b"
+        api_key = os.getenv("ATROPOS_SERVER_API_KEY") or os.getenv("NOUS_API_KEY") or os.getenv("OPENAI_API_KEY") or "local"
+
+        env_config = SimpleTestEnvConfig(
+            tokenizer_name=os.getenv("ATROPOS_TOKENIZER_NAME") or "NousResearch/Hermes-4.3-36B",
+            group_size=4,
+            use_wandb=False,  # Disable wandb for simple testing
+            rollout_server_url="http://localhost:8000",
+            total_steps=10,
+            batch_size=16,
+            steps_per_eval=5,
+            max_token_length=2048,
+            inference_weight=1.0,
+            wandb_name="simple_test",
+            server_base_url=base_url,
+            server_model=model,
+        )
+
+        # OpenAI-compatible servers typically expose chat completions at /v1.
+        server_configs = [
+            APIServerConfig(
+                model_name=model,
+                base_url=f"{base_url}/v1",
+                api_key=api_key,
+                num_max_requests_at_once=4,
+                num_requests_for_eval=8,
+                timeout=120,  # Local models may be slower
+            ),
+        ]
+
+        return env_config, server_configs
+
+    async def setup_agent_env(self):
+        """Setup the environment - load test data."""
+        print(f"SimpleTestEnv setup complete. {len(self.test_prompts)} test prompts loaded.")
+        print(f"Using server at: {self.config.server_base_url}")
+        print(f"Model: {self.config.server_model}")
+
+    async def get_next_item(self) -> Item:
+        """Get the next test prompt."""
+        item = self.test_prompts[self.iter % len(self.test_prompts)]
+        self.iter += 1
+        return item
+
+    def build_task(self, item: Item) -> str:
+        return item["prompt"]
+
+    def build_agent_config(self, item: Item) -> AgentConfig:  # noqa: ARG002
+        return AgentConfig(
+            max_steps=5,
+            temperature=0.7,
+            max_tokens=256,
+            system_prompt=SYSTEM_PROMPT,
+        )
+
+    async def score_trajectory(self, item: Item, final_response: str) -> float:
+        expected = item["expected"].lower()
+        response_lower = (final_response or "").lower()
+        score = 1.0 if expected in response_lower else 0.0
+        self.percent_correct_buffer.append(score)
+        return score
+
+    async def evaluate(self, *args, **kwargs):
+        """
+        Simple evaluation - run through all test prompts once.
+        """
+        correct = 0
+        total = len(self.test_prompts)
+
+        for item in self.test_prompts:
+            messages = [
+                {"role": "system", "content": SYSTEM_PROMPT},
+                {"role": "user", "content": item["prompt"]},
+            ]
+
+            response = await self.server.chat_completion(
+                messages=messages,
+                n=1,
+                max_tokens=256,
+                temperature=0.0,  # Greedy for eval
+                split="eval",
+            )
+
+            response_text = response.choices[0].message.content or ""
+            expected = item["expected"].lower()
+
+            if expected in response_text.lower():
+                correct += 1
+
+        accuracy = correct / total
+        print(f"Evaluation: {correct}/{total} = {accuracy:.2%} accuracy")
+        return {"eval_accuracy": accuracy}
+
+    async def wandb_log(self, wandb_metrics: Optional[Dict] = None):
+        """Log metrics (simplified for testing)."""
+        if wandb_metrics is None:
+            wandb_metrics = {}
+
+        if self.percent_correct_buffer:
+            avg_correct = sum(self.percent_correct_buffer) / len(self.percent_correct_buffer)
+            wandb_metrics["train/percent_correct"] = avg_correct
+            print(f"Train accuracy: {avg_correct:.2%}")
+            self.percent_correct_buffer = []
+
+        await super().wandb_log(wandb_metrics)
+
+
+if __name__ == "__main__":
+    # Allow running as CLI
+    SimpleTestEnv.cli()

atropos/envs/toolserver_smoke_env.py

@@ -0,0 +1,165 @@
+"""
+ToolServer routing smoke environment.
+
+Validates that:
+  - sandbox tools run through Nomad SlotPool (terminal -> bash in sandbox)
+  - external tools run through ToolServer (skills_list)
+
+This env uses ToolServer in-process by default (`tool_server_url="inprocess"`),
+so it is self-contained for local testing.
+
+Run:
+  uv run python -m atropos.envs.toolserver_smoke_env process --env.use_wandb false --env.total_steps 1 --env.group_size 1
+"""
+
+from __future__ import annotations
+
+import os
+from typing import Any, Dict, List, Tuple
+
+from dotenv import load_dotenv
+from pydantic import Field
+
+from atroposlib.envs.base import APIServerConfig, Item
+
+from ..agent import AgentConfig, AgentResult
+from .agent_env import AgentEnv, AgentEnvConfig
+
+load_dotenv()
+
+
+class ToolServerSmokeEnvConfig(AgentEnvConfig):
+    server_base_url: str = Field(
+        default="http://127.0.0.1:8080",
+        description="Base URL for an OpenAI-compatible chat server (without /v1).",
+    )
+    server_model: str = Field(default="hermes-4-36b", description="Model name")
+    tokenizer_name: str = Field(default="NousResearch/Hermes-4.3-36B", description="Tokenizer name for RL tokenization")
+
+
+class ToolServerSmokeEnv(AgentEnv[ToolServerSmokeEnvConfig]):
+    name = "toolserver_smoke_env"
+    env_config_cls = ToolServerSmokeEnvConfig
+
+    def __init__(
+        self,
+        config: ToolServerSmokeEnvConfig,
+        server_configs: List[APIServerConfig],
+        slurm: bool = False,
+        testing: bool = False,
+    ):
+        super().__init__(config, server_configs, slurm, testing)
+        self._iter = 0
+
+    @classmethod
+    def config_init(cls) -> Tuple[ToolServerSmokeEnvConfig, List[APIServerConfig]]:
+        base_url = (
+            os.getenv("ATROPOS_SERVER_BASE_URL")
+            or os.getenv("OPENAI_BASE_URL")
+            or os.getenv("LLM_BASE_URL")
+            or "http://127.0.0.1:8080"
+        )
+        model = os.getenv("ATROPOS_SERVER_MODEL") or os.getenv("LLM_MODEL") or "hermes-4-36b"
+        api_key = os.getenv("ATROPOS_SERVER_API_KEY") or os.getenv("NOUS_API_KEY") or os.getenv("OPENAI_API_KEY") or "local"
+
+        env_config = ToolServerSmokeEnvConfig(
+            tokenizer_name=os.getenv("ATROPOS_TOKENIZER_NAME") or "NousResearch/Hermes-4.3-36B",
+            group_size=1,
+            use_wandb=False,
+            include_messages=True,
+            ensure_scores_are_not_same=False,
+            total_steps=1,
+            batch_size=1,
+            server_base_url=base_url,
+            server_model=model,
+            enabled_toolsets=["terminal", "skills"],
+            disabled_toolsets=[],
+            # Self-contained ToolServer for local smoke.
+            tool_server_url="inprocess",
+            sandbox_image=os.getenv("ATROPOS_SANDBOX_IMAGE") or "atropos-sandbox:local",
+            purge_job_on_start=True,
+            purge_job_on_shutdown=True,
+        )
+
+        server_configs = [
+            APIServerConfig(
+                model_name=model,
+                base_url=f"{base_url.rstrip('/')}/v1",
+                api_key=api_key,
+                num_max_requests_at_once=1,
+                num_requests_for_eval=1,
+                timeout=120,
+            )
+        ]
+        return env_config, server_configs
+
+    async def setup_agent_env(self) -> None:
+        return None
+
+    async def get_next_item(self) -> Item:
+        self._iter += 1
+        return {
+            "prompt": (
+                "You MUST call exactly one tool per assistant message.\n"
+                "\n"
+                "Step 1) Call the skills_list tool (no arguments), then stop.\n"
+                "Step 2) After you receive the tool response, call the terminal tool to run:\n"
+                "python -c \"print('ok')\"\n"
+                "Step 3) After you receive the terminal tool response, answer with just: ok\n"
+                "\n"
+                "Tool call format requirements:\n"
+                "- Every tool call MUST be a complete XML block with a closing tag.\n"
+                "- Do NOT emit a second <tool_call> in the same assistant message.\n"
+                "\n"
+                "Example:\n"
+                "<tool_call>{\"name\": \"skills_list\", \"arguments\": {}}</tool_call>\n"
+                "Do not include anything else in your final answer."
+            )
+        }
+
+    def build_task(self, item: Item) -> str:
+        return str(item.get("prompt") or "")
+
+    def build_agent_config(self, item: Item) -> AgentConfig:  # noqa: ARG002
+        return AgentConfig(
+            max_steps=min(10, int(self.config.agent_max_steps)),
+            temperature=0.2,
+            max_tokens=None,
+        )
+
+    async def score_trajectory(self, item: Item, final_response: str) -> float:
+        _ = (item, final_response)
+        return 0.0
+
+    async def verify_and_score_trajectory(
+        self,
+        item: Item,
+        final_response: str,
+        *,
+        trajectory_id: str,  # noqa: ARG002
+        exec_tool,  # noqa: ARG002
+        agent_result: AgentResult | None = None,
+        workspace_meta: Dict[str, Any] | None = None,  # noqa: ARG002
+    ) -> tuple[float, Dict[str, Any]]:
+        if agent_result is None:
+            return 0.0, {"error": "Missing agent_result"}
+
+        called = {c.name for s in agent_result.steps for c in s.tool_calls}
+        need = {"skills_list", "terminal"}
+        if not need.issubset(called):
+            return 0.0, {"error": f"Missing tool calls: {sorted(need - called)}", "called": sorted(called)}
+
+        terminal_ok = False
+        for step in agent_result.steps:
+            for call, res in zip(step.tool_calls, step.tool_results):
+                if call.name != "terminal":
+                    continue
+                if res.success and (res.output or "").strip().splitlines()[-1].strip() == "ok":
+                    terminal_ok = True
+
+        score = 1.0 if terminal_ok and (final_response or "").strip() == "ok" else 0.0
+        return score, {"called": sorted(called), "final": (final_response or "").strip()}
+
+
+if __name__ == "__main__":
+    ToolServerSmokeEnv.cli()

atropos/nomad/__init__.py

@@ -0,0 +1,11 @@
+"""
+Nomad integration for atropos-agent.
+
+Provides:
+- NomadClient: Client for Nomad HTTP API
+- Job templates for sandbox containers
+"""
+
+from .client import NomadClient
+
+__all__ = ["NomadClient"]

atropos/nomad/client.py

@@ -0,0 +1,500 @@
+"""
+Nomad API Client for atropos-agent.
+
+Provides a simple async client for interacting with the Nomad HTTP API:
+- Submit/stop jobs
+- Query allocations
+- Get allocation addresses
+- Scale jobs up/down
+"""
+
+import asyncio
+import json
+import os
+from dataclasses import dataclass, field
+from enum import Enum
+from pathlib import Path
+from typing import Any, Dict, List, Optional
+
+import aiohttp
+
+
+class AllocationStatus(Enum):
+    """Nomad allocation status."""
+    PENDING = "pending"
+    RUNNING = "running"
+    COMPLETE = "complete"
+    FAILED = "failed"
+    LOST = "lost"
+
+
+@dataclass
+class Allocation:
+    """Information about a Nomad allocation."""
+    id: str
+    job_id: str
+    task_group: str
+    node_id: str
+    status: AllocationStatus
+    # Network info for reaching the allocation
+    address: Optional[str] = None
+    port: Optional[int] = None
+    
+    @property
+    def http_address(self) -> Optional[str]:
+        """Get full HTTP address for the allocation."""
+        if self.address and self.port:
+            return f"http://{self.address}:{self.port}"
+        return None
+
+
+@dataclass
+class JobStatus:
+    """Status of a Nomad job."""
+    id: str
+    name: str
+    status: str
+    allocations: List[Allocation] = field(default_factory=list)
+    count: int = 0  # Number of task groups
+
+
+class NomadClient:
+    """
+    Async client for Nomad HTTP API.
+    
+    Usage:
+        client = NomadClient(address="http://localhost:4646")
+        
+        # Submit a job
+        await client.submit_job(job_spec)
+        
+        # Get allocations
+        allocs = await client.get_job_allocations("sandbox-python")
+        
+        # Scale job
+        await client.scale_job("sandbox-python", count=5)
+    """
+    
+    def __init__(
+        self,
+        address: str = "http://localhost:4646",
+        token: Optional[str] = None,
+        timeout: float = 30.0,
+    ):
+        self.address = address.rstrip("/")
+        self.token = token or os.environ.get("NOMAD_TOKEN")
+        self.timeout = aiohttp.ClientTimeout(total=timeout)
+        self._session: Optional[aiohttp.ClientSession] = None
+    
+    async def _get_session(self) -> aiohttp.ClientSession:
+        """Get or create HTTP session."""
+        if self._session is None or self._session.closed:
+            headers = {}
+            if self.token:
+                headers["X-Nomad-Token"] = self.token
+            self._session = aiohttp.ClientSession(
+                timeout=self.timeout,
+                headers=headers,
+            )
+        return self._session
+    
+    async def close(self):
+        """Close the HTTP session."""
+        if self._session and not self._session.closed:
+            await self._session.close()
+    
+    async def __aenter__(self):
+        return self
+    
+    async def __aexit__(self, exc_type, exc_val, exc_tb):
+        await self.close()
+    
+    async def _request(
+        self,
+        method: str,
+        path: str,
+        data: Optional[Dict[str, Any]] = None,
+    ) -> Dict[str, Any]:
+        """Make an HTTP request to Nomad API."""
+        session = await self._get_session()
+        url = f"{self.address}{path}"
+        
+        try:
+            async with session.request(method, url, json=data) as response:
+                if response.status == 404:
+                    return {"error": "not_found", "status": 404}
+                
+                text = await response.text()
+                if not text:
+                    return {"status": response.status}
+                
+                try:
+                    result = json.loads(text)
+                except json.JSONDecodeError:
+                    return {"text": text, "status": response.status}
+                
+                if response.status >= 400:
+                    return {"error": result, "status": response.status}
+                
+                return result if isinstance(result, dict) else {"data": result, "status": response.status}
+                
+        except aiohttp.ClientError as e:
+            return {"error": str(e), "status": 0}
+    
+    # Job Operations
+    
+    async def submit_job(self, job_spec: Dict[str, Any]) -> Dict[str, Any]:
+        """
+        Submit a job to Nomad.
+        
+        Args:
+            job_spec: Job specification dict (HCL converted to JSON)
+            
+        Returns:
+            Response with EvalID if successful
+        """
+        return await self._request("POST", "/v1/jobs", {"Job": job_spec})
+    
+    async def stop_job(self, job_id: str, purge: bool = False) -> Dict[str, Any]:
+        """
+        Stop (and optionally purge) a job.
+        
+        Args:
+            job_id: Job identifier
+            purge: If True, completely remove the job
+        """
+        path = f"/v1/job/{job_id}"
+        if purge:
+            path += "?purge=true"
+        return await self._request("DELETE", path)
+    
+    async def get_job(self, job_id: str) -> Optional[Dict[str, Any]]:
+        """Get job details."""
+        result = await self._request("GET", f"/v1/job/{job_id}")
+        if "error" in result and result.get("status") == 404:
+            return None
+        return result
+    
+    async def get_job_status(self, job_id: str) -> Optional[JobStatus]:
+        """Get job status with allocations."""
+        job = await self.get_job(job_id)
+        if not job:
+            return None
+        
+        allocs = await self.get_job_allocations(job_id)
+        
+        # Get count from task groups
+        count = 0
+        task_groups = job.get("TaskGroups", [])
+        for tg in task_groups:
+            count += tg.get("Count", 1)
+        
+        return JobStatus(
+            id=job_id,
+            name=job.get("Name", job_id),
+            status=job.get("Status", "unknown"),
+            allocations=allocs,
+            count=count,
+        )
+    
+    # Allocation Operations
+    
+    async def get_job_allocations(self, job_id: str) -> List[Allocation]:
+        """Get all allocations for a job."""
+        result = await self._request("GET", f"/v1/job/{job_id}/allocations")
+        
+        if "error" in result:
+            return []
+        
+        allocs_data = result.get("data", result) if isinstance(result, dict) else result
+        if not isinstance(allocs_data, list):
+            return []
+        
+        allocations = []
+        for alloc_data in allocs_data:
+            # Parse allocation info
+            alloc_id = alloc_data.get("ID", "")
+            status_str = alloc_data.get("ClientStatus", "unknown")
+            
+            try:
+                status = AllocationStatus(status_str)
+            except ValueError:
+                status = AllocationStatus.PENDING
+            
+            # Get network info - need to fetch detailed allocation for this
+            address = None
+            port = None
+            
+            # First try the summary data
+            resources = alloc_data.get("AllocatedResources") or {}
+            shared = resources.get("Shared") or {}
+            networks = shared.get("Networks") or []
+            
+            # If no networks in summary, fetch detailed allocation
+            if not networks and alloc_id:
+                detailed = await self.get_allocation(alloc_id)
+                if detailed:
+                    resources = detailed.get("AllocatedResources") or {}
+                    shared = resources.get("Shared") or {}
+                    networks = shared.get("Networks") or []
+            
+            if networks:
+                network = networks[0]
+                address = network.get("IP")
+                # Look for dynamic ports OR reserved ports (Singularity/raw_exec uses reserved)
+                dyn_ports = network.get("DynamicPorts") or []
+                reserved_ports = network.get("ReservedPorts") or []
+                for dp in dyn_ports + reserved_ports:
+                    if dp.get("Label") == "http":
+                        port = dp.get("Value")
+                        break
+            
+            allocations.append(Allocation(
+                id=alloc_id,
+                job_id=job_id,
+                task_group=alloc_data.get("TaskGroup", ""),
+                node_id=alloc_data.get("NodeID", ""),
+                status=status,
+                address=address,
+                port=port,
+            ))
+        
+        return allocations
+    
+    async def get_allocation(self, alloc_id: str) -> Optional[Dict[str, Any]]:
+        """Get detailed allocation info."""
+        result = await self._request("GET", f"/v1/allocation/{alloc_id}")
+        if "error" in result and result.get("status") == 404:
+            return None
+        return result
+    
+    # Scaling Operations
+    
+    async def scale_job(self, job_id: str, count: int, task_group: str = "sandbox") -> Dict[str, Any]:
+        """
+        Scale a job's task group to specified count.
+        
+        Args:
+            job_id: Job identifier
+            count: Desired number of allocations
+            task_group: Name of task group to scale
+        """
+        payload = {
+            "Count": count,
+            "Target": {
+                "Group": task_group,
+            },
+        }
+        return await self._request("POST", f"/v1/job/{job_id}/scale", payload)
+    
+    async def get_job_scale_status(self, job_id: str) -> Dict[str, int]:
+        """
+        Get current scale status for a job.
+        
+        Returns:
+            Dict mapping task group name to count
+        """
+        result = await self._request("GET", f"/v1/job/{job_id}/scale")
+        
+        if "error" in result:
+            return {}
+        
+        task_groups = result.get("TaskGroups", {})
+        return {
+            name: info.get("Running", 0)
+            for name, info in task_groups.items()
+        }
+    
+    # Health Check
+    
+    async def is_healthy(self) -> bool:
+        """Check if Nomad is reachable and healthy."""
+        try:
+            result = await self._request("GET", "/v1/status/leader")
+            return "error" not in result
+        except Exception:
+            return False
+    
+    async def get_leader(self) -> Optional[str]:
+        """Get current Nomad leader address."""
+        result = await self._request("GET", "/v1/status/leader")
+        if isinstance(result, dict) and "data" in result:
+            return result["data"]
+        return None
+
+
+def load_job_template(
+    template_name: str = "sandbox",
+    **kwargs,
+) -> Dict[str, Any]:
+    """
+    Load and configure a job template.
+    
+    Args:
+        template_name: Name of template (e.g., "sandbox")
+        **kwargs: Template variables to substitute
+        
+    Returns:
+        Job specification dict ready for Nomad API
+    """
+    # Default job template for sandbox container
+    if template_name == "sandbox":
+        return create_sandbox_job(**kwargs)
+    else:
+        raise ValueError(f"Unknown template: {template_name}")
+
+
+def create_sandbox_job(
+    job_id: str = "atropos-sandbox",
+    image: str = "atropos-sandbox:local",  # Use :local tag to avoid registry pull
+    count: int = 1,
+    slots_per_container: int = 10,
+    privileged: bool = False,
+    cpu: int = 500,
+    memory: int = 512,
+    port: int = 8080,
+    datacenter: str = "dc1",
+    driver: str = "docker",  # "docker" or "singularity"
+    singularity_image: str = None,  # Path to .sif file for singularity driver
+) -> Dict[str, Any]:
+    """
+    Create a sandbox job specification.
+    
+    This job runs the sandbox_server.py inside a container,
+    with the specified number of slots for agent workspaces.
+    
+    Args:
+        job_id: Unique job identifier
+        image: Docker image to use (for docker driver)
+        count: Number of container instances
+        slots_per_container: Number of slots per container
+        privileged: Run container in privileged mode (recommended for bubblewrap)
+        cpu: CPU allocation in MHz
+        memory: Memory allocation in MB
+        port: HTTP port for sandbox server
+        datacenter: Nomad datacenter
+        driver: Container driver - "docker" or "singularity"
+        singularity_image: Path to .sif file (required if driver="singularity")
+        
+    Returns:
+        Job specification dict
+    """
+    # Build task config based on driver
+    if driver == "singularity":
+        if not singularity_image:
+            raise ValueError("singularity_image path required when driver='singularity'")
+        
+        # Use raw_exec driver to run apptainer via shell for variable expansion
+        # The container binds the allocation directory for workspace persistence
+        # For raw_exec, we use static port since Nomad's dynamic port mapping doesn't
+        # work the same as Docker - the process runs directly on the host.
+        shell_cmd = (
+            f'apptainer run '
+            f'--bind "$NOMAD_ALLOC_DIR/data:/data" '
+            f'--pwd /app '
+            f'--env PYTHONUNBUFFERED=1 '
+            f'{singularity_image} '
+            f'python sandbox_server.py '
+            f'--port {port} '
+            f'--slots {slots_per_container} '
+            f'--data-dir /data'
+        )
+        task_config = {
+            "command": "/bin/sh",
+            "args": ["-c", shell_cmd],
+        }
+        task_driver = "raw_exec"
+    else:
+        # Docker driver (default)
+        task_config = {
+            "image": image,
+            "force_pull": False,  # Use local image, don't try to pull
+            "ports": ["http"],
+            "privileged": privileged,
+            "command": "python",
+            "args": [
+                "sandbox_server.py",
+                "--port", str(port),
+                "--slots", str(slots_per_container),
+                "--data-dir", "/data",
+            ],
+            # Note: On Linux, you can mount persistent storage:
+            # "volumes": ["${NOMAD_ALLOC_DIR}/data:/data"],
+            # On macOS/Docker Desktop, skip volumes for PoC
+            # (container /data is ephemeral but works for testing)
+        }
+        task_driver = "docker"
+    
+    # For Singularity/raw_exec, use static ports since the process runs directly on host.
+    # For Docker, use dynamic ports with port mapping.
+    if driver == "singularity":
+        network_config = {
+            "Mode": "host",
+            "ReservedPorts": [
+                {
+                    "Label": "http",
+                    "Value": port,
+                }
+            ],
+        }
+    else:
+        network_config = {
+            "Mode": "host",
+            "DynamicPorts": [
+                {
+                    "Label": "http",
+                    "To": port,
+                }
+            ],
+        }
+    
+    return {
+        "ID": job_id,
+        "Name": job_id,
+        "Type": "service",
+        "Datacenters": [datacenter],
+        "TaskGroups": [
+            {
+                "Name": "sandbox",
+                "Count": count,
+                # Speed up deployments and avoid Consul checks. Without this, Nomad may
+                # keep an "active deployment" around for the default MinHealthyTime,
+                # which blocks immediate scaling under load.
+                "Update": {
+                    "HealthCheck": "task_states",
+                    "MinHealthyTime": 0,
+                },
+                "Networks": [network_config],
+                "Tasks": [
+                    {
+                        "Name": "sandbox-server",
+                        "Driver": task_driver,
+                        "Config": task_config,
+                        "Env": {
+                            "PYTHONUNBUFFERED": "1",
+                            "NOMAD_ALLOC_DIR": "${NOMAD_ALLOC_DIR}",
+                        },
+                        "Resources": {
+                            "CPU": cpu,
+                            "MemoryMB": memory,
+                        },
+                        # Note: Services with Checks require Consul, which we skip for the PoC
+                    }
+                ],
+                "RestartPolicy": {
+                    "Attempts": 3,
+                    "Interval": 300_000_000_000,  # 5 minutes
+                    "Delay": 10_000_000_000,     # 10 seconds
+                    "Mode": "delay",
+                },
+                "ReschedulePolicy": {
+                    "Attempts": 5,
+                    "Interval": 3600_000_000_000,  # 1 hour
+                    "Delay": 30_000_000_000,      # 30 seconds
+                    "DelayFunction": "exponential",
+                    "MaxDelay": 300_000_000_000,  # 5 minutes
+                    "Unlimited": False,
+                },
+            }
+        ],
+    }

atropos/slots/__init__.py

@@ -0,0 +1,20 @@
+"""
+Slot-based multiplexing for atropos-agent.
+
+Provides:
+- Slot: Isolated workspace for a single trajectory
+- SlotPool: Manages slots across Nomad allocations  
+- SandboxExecutor: Executes tools in sandbox containers
+"""
+
+from .executor import SandboxExecutor
+from .pool import SlotPool, SlotPoolConfig
+from .slot import Slot, SlotState
+
+__all__ = [
+    "Slot",
+    "SlotState",
+    "SlotPool",
+    "SlotPoolConfig",
+    "SandboxExecutor",
+]

atropos/slots/executor.py

@@ -0,0 +1,457 @@
+"""
+SandboxExecutor - HTTP client for sandbox container communication.
+
+Sends tool execution requests to sandbox_server.py running inside Nomad containers.
+Supports single and batch execution for efficiency.
+"""
+
+import asyncio
+import uuid
+from dataclasses import dataclass, field
+from typing import Any, Dict, List, Optional, Tuple
+
+import aiohttp
+
+from .slot import Slot, SlotState
+from ..tools.base import ToolCall, ToolResult
+
+
+@dataclass
+class ExecutionRequest:
+    """Request to execute a tool in a slot."""
+    slot: Slot
+    tool_name: str
+    args: Dict[str, Any]
+    execution_id: str = field(default_factory=lambda: str(uuid.uuid4()))
+    timeout: float = 30.0
+
+
+@dataclass
+class ExecutionResult:
+    """Result from sandbox execution."""
+    success: bool
+    output: str = ""
+    error: str = ""
+    execution_id: str = ""
+    slot_id: str = ""
+    metadata: Dict[str, Any] = field(default_factory=dict)
+    
+    def to_tool_result(self) -> ToolResult:
+        """Convert to ToolResult for agent consumption."""
+        return ToolResult(
+            success=self.success,
+            output=self.output,
+            error=self.error,
+            metadata=self.metadata,
+            uniq_id=self.execution_id,
+        )
+
+
+class SandboxExecutor:
+    """
+    HTTP client for executing tools in sandbox containers.
+    
+    Communicates with sandbox_server.py running inside Nomad allocations.
+    Supports both single execution and batched parallel execution.
+    
+    Usage:
+        executor = SandboxExecutor()
+        
+        # Single execution
+        result = await executor.execute(slot, "bash", {"command": "ls"})
+        
+        # Batch execution
+        results = await executor.execute_batch([
+            (slot1, "bash", {"command": "ls"}),
+            (slot2, "write_file", {"path": "test.txt", "content": "hello"}),
+        ])
+    """
+    
+    def __init__(
+        self,
+        timeout: float = 30.0,
+        max_retries: int = 3,
+        retry_delay: float = 1.0,
+    ):
+        self.timeout = aiohttp.ClientTimeout(total=timeout)
+        self.max_retries = max_retries
+        self.retry_delay = retry_delay
+        self._session: Optional[aiohttp.ClientSession] = None
+    
+    async def _get_session(self) -> aiohttp.ClientSession:
+        """Get or create HTTP session."""
+        if self._session is None or self._session.closed:
+            self._session = aiohttp.ClientSession(timeout=self.timeout)
+        return self._session
+    
+    async def close(self):
+        """Close HTTP session."""
+        if self._session and not self._session.closed:
+            await self._session.close()
+    
+    async def __aenter__(self):
+        return self
+    
+    async def __aexit__(self, exc_type, exc_val, exc_tb):
+        await self.close()
+    
+    async def execute(
+        self,
+        slot: Slot,
+        tool_name: str,
+        args: Dict[str, Any],
+        timeout: Optional[float] = None,
+    ) -> ExecutionResult:
+        """
+        Execute a tool in a slot's workspace.
+        
+        Args:
+            slot: Slot to execute in
+            tool_name: Name of tool (bash, read_file, write_file)
+            args: Tool arguments
+            timeout: Optional timeout override
+            
+        Returns:
+            ExecutionResult with output or error
+        """
+        execution_id = str(uuid.uuid4())
+        exec_timeout = timeout or self.timeout.total or 30.0
+        
+        # Mark slot as executing
+        original_state = slot.state
+        try:
+            if slot.state == SlotState.ACQUIRED:
+                slot.start_execution(execution_id)
+            
+            result = await self._send_execute_request(
+                container_addr=slot.container_addr,
+                slot_id=slot.slot_id,
+                tool_name=tool_name,
+                args=args,
+                execution_id=execution_id,
+                timeout=exec_timeout,
+            )
+            result.slot_id = slot.slot_id
+            return result
+            
+        finally:
+            # Restore slot state
+            if slot.state == SlotState.EXECUTING:
+                slot.end_execution()
+    
+    async def _send_execute_request(
+        self,
+        container_addr: str,
+        slot_id: str,
+        tool_name: str,
+        args: Dict[str, Any],
+        execution_id: str,
+        timeout: float,
+    ) -> ExecutionResult:
+        """Send execution request to sandbox server with retry logic."""
+        session = await self._get_session()
+        url = f"{container_addr}/execute"
+        
+        payload = {
+            "slot_id": slot_id,
+            "tool": tool_name,
+            "args": args,
+            "execution_id": execution_id,
+            "timeout": timeout,
+        }
+        
+        last_error = None
+        for attempt in range(self.max_retries):
+            try:
+                async with session.post(url, json=payload) as response:
+                    data = await response.json()
+                    
+                    return ExecutionResult(
+                        success=data.get("success", False),
+                        output=data.get("output", ""),
+                        error=data.get("error", ""),
+                        execution_id=data.get("execution_id", execution_id),
+                        metadata=data.get("metadata", {}),
+                    )
+                    
+            except aiohttp.ClientError as e:
+                last_error = str(e)
+                if attempt < self.max_retries - 1:
+                    await asyncio.sleep(self.retry_delay * (attempt + 1))
+                continue
+            except asyncio.TimeoutError:
+                last_error = f"Request timed out after {timeout}s"
+                break
+            except Exception as e:
+                last_error = str(e)
+                break
+        
+        return ExecutionResult(
+            success=False,
+            error=f"Failed after {self.max_retries} attempts: {last_error}",
+            execution_id=execution_id,
+        )
+    
+    async def execute_batch(
+        self,
+        requests: List[Tuple[Slot, str, Dict[str, Any]]],
+        timeout: Optional[float] = None,
+    ) -> List[ExecutionResult]:
+        """
+        Execute multiple tools in parallel across slots.
+        
+        This is the key optimization - we batch tool calls to maximize
+        container utilization while agents are waiting for LLM responses.
+        
+        Args:
+            requests: List of (slot, tool_name, args) tuples
+            timeout: Optional timeout override
+            
+        Returns:
+            List of ExecutionResults in same order as requests
+        """
+        if not requests:
+            return []
+        
+        # Group requests by container address for batch API
+        by_container: Dict[str, List[Tuple[int, Slot, str, Dict[str, Any], str]]] = {}
+        
+        for idx, (slot, tool_name, args) in enumerate(requests):
+            execution_id = str(uuid.uuid4())
+            container = slot.container_addr
+            
+            if container not in by_container:
+                by_container[container] = []
+            by_container[container].append((idx, slot, tool_name, args, execution_id))
+            
+            # Mark slots as executing
+            if slot.state == SlotState.ACQUIRED:
+                slot.start_execution(execution_id)
+        
+        # Execute batches in parallel
+        exec_timeout = timeout or self.timeout.total or 30.0
+        batch_tasks = []
+        
+        for container_addr, batch_requests in by_container.items():
+            task = self._send_batch_request(
+                container_addr=container_addr,
+                batch_requests=batch_requests,
+                timeout=exec_timeout,
+            )
+            batch_tasks.append(task)
+        
+        # Gather all batch results
+        batch_results = await asyncio.gather(*batch_tasks, return_exceptions=True)
+        
+        # Collect results in original order
+        results: List[Optional[ExecutionResult]] = [None] * len(requests)
+        
+        for batch_result in batch_results:
+            if isinstance(batch_result, Exception):
+                # Mark all in this batch as failed
+                continue
+            
+            for idx, result in batch_result:
+                results[idx] = result
+        
+        # Fill in any missing results
+        for idx, result in enumerate(results):
+            if result is None:
+                slot, tool_name, args = requests[idx]
+                results[idx] = ExecutionResult(
+                    success=False,
+                    error="Batch execution failed",
+                    slot_id=slot.slot_id,
+                )
+        
+        # End execution on all slots
+        for slot, _, _ in requests:
+            if slot.state == SlotState.EXECUTING:
+                slot.end_execution()
+        
+        return results  # type: ignore
+    
+    async def _send_batch_request(
+        self,
+        container_addr: str,
+        batch_requests: List[Tuple[int, Slot, str, Dict[str, Any], str]],
+        timeout: float,
+    ) -> List[Tuple[int, ExecutionResult]]:
+        """Send batch execution request to a single container."""
+        session = await self._get_session()
+        url = f"{container_addr}/batch"
+        
+        # Build batch payload
+        payload = [
+            {
+                "slot_id": slot.slot_id,
+                "tool": tool_name,
+                "args": args,
+                "execution_id": execution_id,
+                "timeout": timeout,
+            }
+            for _, slot, tool_name, args, execution_id in batch_requests
+        ]
+        
+        try:
+            async with session.post(url, json=payload) as response:
+                data = await response.json()
+                
+                if not isinstance(data, list):
+                    raise ValueError(f"Expected list response, got {type(data)}")
+                
+                results = []
+                for i, (idx, slot, _, _, execution_id) in enumerate(batch_requests):
+                    if i < len(data):
+                        item = data[i]
+                        result = ExecutionResult(
+                            success=item.get("success", False),
+                            output=item.get("output", ""),
+                            error=item.get("error", ""),
+                            execution_id=item.get("execution_id", execution_id),
+                            slot_id=slot.slot_id,
+                            metadata=item.get("metadata", {}),
+                        )
+                    else:
+                        result = ExecutionResult(
+                            success=False,
+                            error="Missing result in batch response",
+                            execution_id=execution_id,
+                            slot_id=slot.slot_id,
+                        )
+                    results.append((idx, result))
+                
+                return results
+                
+        except Exception as e:
+            # Return error for all requests in batch
+            return [
+                (idx, ExecutionResult(
+                    success=False,
+                    error=str(e),
+                    execution_id=execution_id,
+                    slot_id=slot.slot_id,
+                ))
+                for idx, slot, _, _, execution_id in batch_requests
+            ]
+    
+    async def reset_slot(self, slot: Slot) -> ExecutionResult:
+        """
+        Reset a slot's workspace (delete all files).
+        
+        Useful when reusing a slot for a new trajectory.
+        """
+        session = await self._get_session()
+        url = f"{slot.container_addr}/reset"
+        
+        try:
+            async with session.post(url, json={"slot_id": slot.slot_id}) as response:
+                data = await response.json()
+                return ExecutionResult(
+                    success=data.get("success", False),
+                    output=data.get("output", ""),
+                    error=data.get("error", ""),
+                    slot_id=slot.slot_id,
+                )
+        except Exception as e:
+            return ExecutionResult(
+                success=False,
+                error=str(e),
+                slot_id=slot.slot_id,
+            )
+    
+    async def health_check(self, container_addr: str) -> bool:
+        """Check if a sandbox container is healthy."""
+        session = await self._get_session()
+        url = f"{container_addr}/health"
+        
+        try:
+            async with session.get(url) as response:
+                data = await response.json()
+                return data.get("status") == "ok"
+        except Exception:
+            return False
+    
+    async def get_container_status(
+        self, 
+        container_addr: str
+    ) -> Optional[Dict[str, Any]]:
+        """Get status info from a sandbox container."""
+        session = await self._get_session()
+        url = f"{container_addr}/health"
+        
+        try:
+            async with session.get(url) as response:
+                return await response.json()
+        except Exception:
+            return None
+
+    # -------------------------------------------------------------------------
+    # Artifact helpers (optional)
+    # -------------------------------------------------------------------------
+
+    async def _post_json(
+        self,
+        url: str,
+        payload: Dict[str, Any],
+        timeout: Optional[float] = None,
+    ) -> Dict[str, Any]:
+        session = await self._get_session()
+        try:
+            async with session.post(url, json=payload, timeout=timeout) as response:
+                data = await response.json()
+                if isinstance(data, dict):
+                    data.setdefault("http_status", response.status)
+                    return data
+                return {"success": False, "error": f"Unexpected response type: {type(data)}", "http_status": response.status}
+        except Exception as e:
+            return {"success": False, "error": str(e)}
+
+    async def read_artifact(
+        self,
+        slot: Slot,
+        path: str,
+        *,
+        encoding: str = "text",
+        max_bytes: Optional[int] = None,
+        include_sha256: bool = False,
+        timeout: Optional[float] = None,
+    ) -> Dict[str, Any]:
+        url = f"{slot.container_addr}/artifacts/read"
+        payload: Dict[str, Any] = {"slot_id": slot.slot_id, "path": path, "encoding": encoding, "include_sha256": include_sha256}
+        if max_bytes is not None:
+            payload["max_bytes"] = max_bytes
+        return await self._post_json(url, payload, timeout=timeout)
+
+    async def list_artifacts(
+        self,
+        slot: Slot,
+        path: str = ".",
+        *,
+        recursive: bool = False,
+        max_entries: Optional[int] = None,
+        timeout: Optional[float] = None,
+    ) -> Dict[str, Any]:
+        url = f"{slot.container_addr}/artifacts/list"
+        payload: Dict[str, Any] = {"slot_id": slot.slot_id, "path": path, "recursive": recursive}
+        if max_entries is not None:
+            payload["max_entries"] = max_entries
+        return await self._post_json(url, payload, timeout=timeout)
+
+    async def archive_artifacts(
+        self,
+        slot: Slot,
+        path: str = ".",
+        *,
+        archive_format: str = "tar.gz",
+        max_bytes: Optional[int] = None,
+        max_entries: Optional[int] = None,
+        timeout: Optional[float] = None,
+    ) -> Dict[str, Any]:
+        url = f"{slot.container_addr}/artifacts/archive"
+        payload: Dict[str, Any] = {"slot_id": slot.slot_id, "path": path, "format": archive_format}
+        if max_bytes is not None:
+            payload["max_bytes"] = max_bytes
+        if max_entries is not None:
+            payload["max_entries"] = max_entries
+        return await self._post_json(url, payload, timeout=timeout)

atropos/slots/pool.py

@@ -0,0 +1,659 @@
+"""
+SlotPool - Manages slots across Nomad allocations.
+
+The SlotPool is the core abstraction for slot-based multiplexing:
+- Tracks available/acquired slots across containers
+- Handles slot acquisition and release
+- Auto-scales Nomad job count based on demand
+- Provides batched tool execution
+"""
+
+import asyncio
+import logging
+import os
+import subprocess
+from dataclasses import dataclass, field
+from pathlib import Path
+from typing import Any, Dict, List, Optional, Tuple
+
+from ..nomad.client import (
+    Allocation,
+    AllocationStatus,
+    NomadClient,
+    create_sandbox_job,
+)
+from .executor import ExecutionResult, SandboxExecutor
+from .slot import Slot, SlotState, create_slots_for_allocation
+
+logger = logging.getLogger(__name__)
+
+
+@dataclass
+class SlotPoolConfig:
+    """Configuration for SlotPool."""
+    
+    # Nomad settings
+    nomad_address: str = "http://localhost:4646"
+    job_id: str = "atropos-sandbox"
+    datacenter: str = "dc1"
+    
+    # Container settings
+    image: str = "atropos-sandbox:local"  # Use :local tag to avoid registry pull
+    slots_per_container: int = 10
+    privileged: bool = False
+    cpu: int = 500  # MHz
+    memory: int = 512  # MB
+    
+    # Driver selection: "docker" or "singularity"
+    driver: str = "docker"
+    # Path to .sif file for singularity driver (required if driver="singularity")
+    singularity_image: Optional[str] = None
+    
+    # Scaling settings
+    min_containers: int = 1
+    max_containers: int = 10
+    
+    # Timeouts
+    acquire_timeout: float = 30.0  # Seconds between acquire polls (also triggers scale-up attempts)
+    health_check_interval: float = 30.0  # Seconds between health checks
+    scale_cooldown: float = 60.0  # Seconds between scale operations
+
+    # Job lifecycle
+    purge_job_on_start: bool = False  # Purge any pre-existing job before starting (local dev/training friendly)
+
+    # Local Docker image convenience (macOS/Nomad dev mode)
+    auto_build_local_image: bool = True  # If image endswith :local and is missing, build it from the bundled Dockerfile.
+    dockerfile_path: Optional[str] = None  # Override Dockerfile path (default: Hermes-Agent/atropos/Dockerfile).
+    docker_build_context: Optional[str] = None  # Override build context (default: Hermes-Agent/atropos).
+
+
+class SlotPool:
+    """
+    Manages a pool of slots across Nomad allocations.
+    
+    The SlotPool:
+    - Deploys sandbox containers to Nomad
+    - Tracks slots across all running containers
+    - Handles slot acquisition/release
+    - Auto-scales based on demand
+    - Provides batched execution via SandboxExecutor
+    
+    Usage:
+        config = SlotPoolConfig(
+            nomad_address="http://localhost:4646",
+            job_id="my-sandbox",
+            slots_per_container=10,
+        )
+        
+        pool = SlotPool(config)
+        await pool.start()
+        
+        # Acquire a slot
+        slot = await pool.acquire()
+        
+        # Execute tool
+        result = await pool.execute(slot, "bash", {"command": "ls"})
+        
+        # Release slot
+        await pool.release(slot)
+        
+        # Shutdown
+        await pool.stop()
+    """
+    
+    def __init__(self, config: Optional[SlotPoolConfig] = None):
+        self.config = config or SlotPoolConfig()
+        
+        # Nomad client
+        self.nomad = NomadClient(address=self.config.nomad_address)
+        
+        # Sandbox executor for tool execution
+        self.executor = SandboxExecutor()
+        
+        # Slot tracking
+        self._slots: Dict[str, Slot] = {}  # slot_key -> Slot
+        self._available_queue: asyncio.Queue[str] = asyncio.Queue()
+        self._lock = asyncio.Lock()
+        self._scale_lock = asyncio.Lock()
+        
+        # State
+        self._started = False
+        self._health_task: Optional[asyncio.Task] = None
+        self._scale_task: Optional[asyncio.Task] = None
+        self._last_scale_time = 0.0
+
+    def _default_dockerfile_path(self) -> Path:
+        # Hermes-Agent/atropos/Dockerfile lives next to this module in source checkouts.
+        return Path(__file__).resolve().parents[1] / "Dockerfile"
+
+    def _default_build_context(self) -> Path:
+        return Path(__file__).resolve().parents[1]
+
+    def _docker_image_exists(self, image: str) -> bool:
+        try:
+            proc = subprocess.run(
+                ["docker", "image", "inspect", image],
+                stdout=subprocess.DEVNULL,
+                stderr=subprocess.DEVNULL,
+                check=False,
+                env={**os.environ, "DOCKER_CLI_HINTS": "false"},
+            )
+            return proc.returncode == 0
+        except FileNotFoundError:
+            return False
+
+    def _try_build_local_image(self, image: str) -> None:
+        dockerfile = Path(self.config.dockerfile_path) if self.config.dockerfile_path else self._default_dockerfile_path()
+        context = Path(self.config.docker_build_context) if self.config.docker_build_context else self._default_build_context()
+
+        if not dockerfile.exists():
+            raise RuntimeError(
+                f"Sandbox Dockerfile not found at {dockerfile}. "
+                "Build the sandbox image manually or set --env.purge_job_on_start false and provide a non-local image."
+            )
+        if not context.exists():
+            raise RuntimeError(f"Docker build context not found at {context}")
+
+        # Prefer buildx+--load to ensure the image ends up in the local daemon (required by Nomad's docker driver).
+        buildx_cmd = [
+            "docker",
+            "buildx",
+            "build",
+            "--load",
+            "-t",
+            image,
+            "-f",
+            str(dockerfile),
+            str(context),
+        ]
+        proc = subprocess.run(buildx_cmd, check=False, env={**os.environ, "DOCKER_CLI_HINTS": "false"})
+        if proc.returncode == 0:
+            return
+
+        # Fallback to classic docker build if buildx isn't available.
+        build_cmd = ["docker", "build", "-t", image, "-f", str(dockerfile), str(context)]
+        proc2 = subprocess.run(build_cmd, check=False, env={**os.environ, "DOCKER_CLI_HINTS": "false"})
+        if proc2.returncode != 0:
+            raise RuntimeError(
+                f"Failed to build local sandbox image {image}. "
+                f"Tried: {' '.join(buildx_cmd)} and {' '.join(build_cmd)}"
+            )
+
+    def _ensure_local_image(self) -> None:
+        image = (self.config.image or "").strip()
+        if not image.endswith(":local"):
+            return
+        if not self.config.auto_build_local_image:
+            return
+
+        if self._docker_image_exists(image):
+            return
+
+        logger.info(f"Local sandbox image {image} not found; building it now...")
+        self._try_build_local_image(image)
+
+    def _slot_key(self, alloc_id: str, slot_id: str) -> str:
+        """Generate unique key for a slot."""
+        return f"{alloc_id}:{slot_id}"
+    
+    @property
+    def total_slots(self) -> int:
+        """Total number of slots in pool."""
+        return len(self._slots)
+    
+    @property
+    def available_slots(self) -> int:
+        """Number of available slots."""
+        return sum(1 for s in self._slots.values() if s.is_available)
+    
+    @property
+    def acquired_slots(self) -> int:
+        """Number of acquired slots."""
+        return sum(1 for s in self._slots.values() if s.is_acquired)
+    
+    async def start(self) -> None:
+        """
+        Start the slot pool.
+        
+        - Checks if Nomad is healthy
+        - Deploys sandbox job if not running
+        - Discovers existing allocations
+        - Starts health check background task
+        """
+        if self._started:
+            return
+        
+        logger.info(f"Starting SlotPool (job_id={self.config.job_id})")
+
+        try:
+            # Make sure local sandbox images exist before Nomad tries to pull them.
+            # This is a common footgun in macOS dev mode with :local tags.
+            self._ensure_local_image()
+
+            # Check Nomad health
+            if not await self.nomad.is_healthy():
+                raise RuntimeError(f"Nomad is not reachable at {self.config.nomad_address}")
+
+            if self.config.purge_job_on_start:
+                logger.info(f"Purging any existing Nomad job: {self.config.job_id}")
+                await self.nomad.stop_job(self.config.job_id, purge=True)
+
+            # Check if job exists (after optional purge)
+            job = await self.nomad.get_job(self.config.job_id)
+
+            if job is None:
+                # Deploy new job
+                logger.info(f"Deploying sandbox job: {self.config.job_id} (driver={self.config.driver})")
+                job_spec = create_sandbox_job(
+                    job_id=self.config.job_id,
+                    image=self.config.image,
+                    count=self.config.min_containers,
+                    slots_per_container=self.config.slots_per_container,
+                    privileged=self.config.privileged,
+                    cpu=self.config.cpu,
+                    memory=self.config.memory,
+                    datacenter=self.config.datacenter,
+                    driver=self.config.driver,
+                    singularity_image=self.config.singularity_image,
+                )
+                result = await self.nomad.submit_job(job_spec)
+                if "error" in result:
+                    raise RuntimeError(f"Failed to submit job: {result}")
+
+            # Wait for allocations to be running (even if the job already existed).
+            await self._wait_for_healthy_allocations(self.config.min_containers)
+
+            # Discover existing allocations and slots
+            await self._refresh_slots()
+
+            # Start health check task
+            self._health_task = asyncio.create_task(self._health_check_loop())
+
+            self._started = True
+            logger.info(f"SlotPool started: {self.total_slots} slots available")
+        except Exception:
+            # Ensure aiohttp sessions are not leaked if we fail to start.
+            await self.stop(purge_job=False)
+            raise
+    
+    async def stop(self, purge_job: bool = False) -> None:
+        """
+        Stop the slot pool.
+        
+        Args:
+            purge_job: If True, also stop the Nomad job
+        """
+        logger.info("Stopping SlotPool")
+
+        # Cancel health check task
+        if self._health_task:
+            self._health_task.cancel()
+            try:
+                await self._health_task
+            except asyncio.CancelledError:
+                pass
+            finally:
+                self._health_task = None
+
+        if self._scale_task:
+            self._scale_task.cancel()
+            try:
+                await self._scale_task
+            except asyncio.CancelledError:
+                pass
+            finally:
+                self._scale_task = None
+
+        # Optionally stop the job (do this even if start() never completed).
+        if purge_job:
+            logger.info(f"Stopping Nomad job: {self.config.job_id}")
+            await self.nomad.stop_job(self.config.job_id, purge=True)
+
+        # Close connections
+        await self.executor.close()
+        await self.nomad.close()
+
+        self._started = False
+        self._slots.clear()
+
+        # Clear the queue
+        while not self._available_queue.empty():
+            try:
+                self._available_queue.get_nowait()
+            except asyncio.QueueEmpty:
+                break
+    
+    async def acquire(self, trajectory_id: Optional[str] = None) -> Slot:
+        """
+        Acquire an available slot.
+        
+        If no slots are available, waits up to acquire_timeout seconds.
+        If still no slots, attempts to scale up.
+        
+        Args:
+            trajectory_id: Optional ID of trajectory acquiring the slot
+            
+        Returns:
+            Acquired Slot
+            
+        Raises:
+            asyncio.TimeoutError: If no slot becomes available
+        """
+        if not self._started:
+            raise RuntimeError("SlotPool not started")
+
+        while True:
+            try:
+                # Try to get an available slot
+                slot_key = await asyncio.wait_for(
+                    self._available_queue.get(),
+                    timeout=self.config.acquire_timeout,
+                )
+            except asyncio.TimeoutError:
+                # Try to scale up, but keep waiting even if scaling isn't possible.
+                # In practice, slots may become available shortly (e.g. contention),
+                # and scaling may be temporarily blocked by Nomad deployments.
+                await self._try_scale_up()
+                continue
+
+            slot = self._slots.get(slot_key)
+            if slot is None:
+                # Slot was removed; discard stale queue entry and retry.
+                continue
+
+            try:
+                slot.acquire(trajectory_id)
+            except RuntimeError:
+                # Slot isn't actually available (e.g. duplicate queue entry); retry.
+                continue
+
+            logger.debug(f"Acquired slot {slot.slot_id} (alloc={slot.alloc_id[:8]})")
+            return slot
+    
+    async def release(self, slot: Slot, reset_workspace: bool = False) -> None:
+        """
+        Release a slot back to the pool.
+        
+        Args:
+            slot: Slot to release
+            reset_workspace: If True, clear the workspace files
+        """
+        slot_key = self._slot_key(slot.alloc_id, slot.slot_id)
+        
+        if slot_key not in self._slots:
+            logger.warning(f"Releasing unknown slot: {slot_key}")
+            return
+        
+        # Optionally reset workspace
+        if reset_workspace:
+            await self.executor.reset_slot(slot)
+        
+        slot.release()
+        await self._available_queue.put(slot_key)
+        
+        logger.debug(f"Released slot {slot.slot_id}")
+    
+    async def execute(
+        self,
+        slot: Slot,
+        tool_name: str,
+        args: Dict[str, Any],
+        timeout: Optional[float] = None,
+    ) -> ExecutionResult:
+        """
+        Execute a tool in a slot's workspace.
+        
+        Args:
+            slot: Slot to execute in
+            tool_name: Name of tool (bash, read_file, write_file)
+            args: Tool arguments
+            timeout: Optional timeout override
+            
+        Returns:
+            ExecutionResult
+        """
+        return await self.executor.execute(slot, tool_name, args, timeout)
+    
+    async def execute_batch(
+        self,
+        requests: List[Tuple[Slot, str, Dict[str, Any]]],
+        timeout: Optional[float] = None,
+    ) -> List[ExecutionResult]:
+        """
+        Execute multiple tools in parallel.
+        
+        This is the key optimization - batch execution across multiple slots
+        maximizes container utilization.
+        
+        Args:
+            requests: List of (slot, tool_name, args) tuples
+            timeout: Optional timeout override
+            
+        Returns:
+            List of ExecutionResults in same order
+        """
+        return await self.executor.execute_batch(requests, timeout)
+    
+    async def _refresh_slots(self) -> None:
+        """Refresh slot inventory from Nomad allocations."""
+        async with self._lock:
+            allocs = await self.nomad.get_job_allocations(self.config.job_id)
+            
+            # Track which slots we've seen
+            seen_keys = set()
+            
+            for alloc in allocs:
+                if alloc.status != AllocationStatus.RUNNING:
+                    continue
+                
+                if not alloc.http_address:
+                    continue
+                
+                # Check container health
+                healthy = await self.executor.health_check(alloc.http_address)
+                if not healthy:
+                    continue
+                
+                # Create slots for this allocation
+                for i in range(self.config.slots_per_container):
+                    slot_id = f"slot_{i}"
+                    slot_key = self._slot_key(alloc.id, slot_id)
+                    seen_keys.add(slot_key)
+                    
+                    if slot_key not in self._slots:
+                        # New slot
+                        slot = Slot(
+                            slot_id=slot_id,
+                            alloc_id=alloc.id,
+                            container_addr=alloc.http_address,
+                        )
+                        self._slots[slot_key] = slot
+                        await self._available_queue.put(slot_key)
+                        logger.debug(f"Added slot: {slot_key}")
+            
+            # Remove slots from dead allocations
+            for slot_key in list(self._slots.keys()):
+                if slot_key not in seen_keys:
+                    slot = self._slots.pop(slot_key)
+                    logger.debug(f"Removed slot: {slot_key}")
+    
+    async def _wait_for_healthy_allocations(
+        self, 
+        min_count: int, 
+        timeout: float = 120.0
+    ) -> None:
+        """Wait for allocations to become healthy."""
+        import time
+        start = time.time()
+
+        def _summarize_alloc_detail(detail: Dict[str, Any]) -> str:
+            task_states = detail.get("TaskStates") or {}
+            parts: List[str] = []
+            if isinstance(task_states, dict):
+                for task_name, st in task_states.items():
+                    events = (st or {}).get("Events") or []
+                    if isinstance(events, list) and events:
+                        # Include a few recent events; the latest can be a generic restart message
+                        # while the true root cause is slightly earlier (e.g. image pull failure).
+                        recent = events[-3:]
+                        msgs: List[str] = []
+                        for ev in recent:
+                            desc = ev.get("DisplayMessage") or ev.get("Message") or ev.get("Type") or ""
+                            if desc:
+                                msgs.append(desc)
+                        if msgs:
+                            parts.append(f"{task_name}: " + " | ".join(msgs))
+            return "; ".join(parts)
+
+        def _alloc_events_lower(detail: Dict[str, Any]) -> str:
+            task_states = detail.get("TaskStates") or {}
+            texts: List[str] = []
+            if isinstance(task_states, dict):
+                for _task_name, st in task_states.items():
+                    events = (st or {}).get("Events") or []
+                    if isinstance(events, list):
+                        for ev in events[-10:]:
+                            desc = ev.get("DisplayMessage") or ev.get("Message") or ev.get("Type") or ""
+                            if desc:
+                                texts.append(desc)
+            return " ".join(texts).lower()
+        
+        while time.time() - start < timeout:
+            allocs = await self.nomad.get_job_allocations(self.config.job_id)
+            
+            healthy_count = 0
+            for alloc in allocs:
+                if alloc.status == AllocationStatus.RUNNING and alloc.http_address:
+                    if await self.executor.health_check(alloc.http_address):
+                        healthy_count += 1
+
+                # Fast-fail on obvious driver/image errors to avoid waiting out the full timeout.
+                if alloc.id:
+                    detail = await self.nomad.get_allocation(alloc.id)
+                    if isinstance(detail, dict):
+                        summary = _summarize_alloc_detail(detail)
+                        lowered = _alloc_events_lower(detail) or summary.lower()
+                        if "failed to pull" in lowered or "pull access denied" in lowered:
+                            raise RuntimeError(
+                                "Nomad allocation failed to start due to a Docker image pull error. "
+                                f"Allocation {alloc.id[:8]}: {summary}\n"
+                                "If you're using a local image tag (e.g. `atropos-sandbox:local`) on macOS, "
+                                "make sure the image is loaded into Docker, e.g.:\n"
+                                "  docker buildx build --load -t atropos-sandbox:local -f Hermes-Agent/atropos/Dockerfile Hermes-Agent/atropos"
+                            )
+                        if "exceeded allowed attempts" in lowered:
+                            raise RuntimeError(
+                                "Nomad allocation is crash-looping and has entered restart backoff. "
+                                f"Allocation {alloc.id[:8]}: {summary}\n"
+                                "Inspect logs with:\n"
+                                f"  nomad alloc logs -stderr -task sandbox-server {alloc.id}\n"
+                                "Common causes include: missing local Docker image tag, container entrypoint error, "
+                                "or sandbox-server startup failure."
+                            )
+            
+            if healthy_count >= min_count:
+                return
+            
+            await asyncio.sleep(2.0)
+
+        # Timed out: include allocation status detail to help debugging.
+        allocs = await self.nomad.get_job_allocations(self.config.job_id)
+        alloc_lines: List[str] = []
+        for alloc in allocs[:10]:
+            addr = alloc.http_address or "-"
+            line = f"{alloc.id[:8]} status={alloc.status.value} http={addr}"
+            detail = await self.nomad.get_allocation(alloc.id)
+            if isinstance(detail, dict):
+                summary = _summarize_alloc_detail(detail)
+                if summary:
+                    line += f" detail={summary}"
+            alloc_lines.append(line)
+
+        hint = (
+            "Timed out waiting for healthy sandbox allocations.\n"
+            f"Job: {self.config.job_id}, desired_healthy: {min_count}\n"
+            "Allocations:\n  - " + "\n  - ".join(alloc_lines)
+        )
+        raise RuntimeError(hint)
+    
+    async def _try_scale_up(self) -> bool:
+        """Attempt to scale up the job."""
+        import time
+
+        async with self._scale_lock:
+            # Check cooldown
+            if time.time() - self._last_scale_time < self.config.scale_cooldown:
+                return False
+
+            # Check max containers
+            status = await self.nomad.get_job_status(self.config.job_id)
+            if status is None:
+                return False
+
+            current_count = status.count
+            if current_count >= self.config.max_containers:
+                logger.warning(f"Cannot scale up: already at max ({self.config.max_containers})")
+                return False
+
+            # Scale up
+            new_count = min(current_count + 1, self.config.max_containers)
+            logger.info(f"Scaling up from {current_count} to {new_count} containers")
+
+            scale_resp = await self.nomad.scale_job(
+                self.config.job_id,
+                count=new_count,
+                task_group="sandbox",
+            )
+
+            # Nomad may return non-JSON errors (e.g. plain text) with a status field.
+            if isinstance(scale_resp, dict) and scale_resp.get("status", 200) >= 400:
+                logger.warning(f"Scale request rejected: {scale_resp}")
+                self._last_scale_time = time.time()
+                return False
+
+            self._last_scale_time = time.time()
+
+            # Wait for new allocation in the background so contended acquires can still
+            # make progress (e.g. by grabbing slots released by other trajectories).
+            if self._scale_task is None or self._scale_task.done():
+                self._scale_task = asyncio.create_task(self._wait_for_scale(new_count))
+
+            return True
+
+    async def _wait_for_scale(self, desired_count: int) -> None:
+        try:
+            await self._wait_for_healthy_allocations(desired_count, timeout=60.0)
+            await self._refresh_slots()
+        except asyncio.CancelledError:
+            raise
+        except Exception as e:
+            logger.error(f"Failed to scale up: {e}")
+    
+    async def _health_check_loop(self) -> None:
+        """Background task to monitor container health."""
+        while True:
+            try:
+                await asyncio.sleep(self.config.health_check_interval)
+                await self._refresh_slots()
+            except asyncio.CancelledError:
+                break
+            except Exception as e:
+                logger.error(f"Health check error: {e}")
+    
+    def get_stats(self) -> Dict[str, Any]:
+        """Get pool statistics."""
+        slots_by_state = {}
+        for slot in self._slots.values():
+            state = slot.state.value
+            slots_by_state[state] = slots_by_state.get(state, 0) + 1
+
+        container_count = len({s.alloc_id for s in self._slots.values()}) if self._slots else 0
+        
+        return {
+            "total_slots": self.total_slots,
+            "available_slots": self.available_slots,
+            "acquired_slots": self.acquired_slots,
+            "containers": container_count,
+            "slots_by_state": slots_by_state,
+            "started": self._started,
+        }

atropos/slots/slot.py

@@ -0,0 +1,159 @@
+"""
+Slot abstraction for atropos-agent.
+
+A Slot represents an isolated workspace for a single agent trajectory.
+Slots are hosted on Nomad allocations and provide workspace isolation
+via filesystem directories.
+"""
+
+from dataclasses import dataclass, field
+from enum import Enum
+from typing import Any, Dict, Optional
+import uuid
+
+
+class SlotState(Enum):
+    """State of a slot in the pool."""
+    AVAILABLE = "available"      # Ready to be acquired
+    ACQUIRED = "acquired"        # Assigned to a trajectory
+    EXECUTING = "executing"      # Currently executing a tool
+    RELEASING = "releasing"      # Being released back to pool
+    ERROR = "error"              # In error state
+
+
+@dataclass
+class Slot:
+    """
+    An isolated workspace for a single agent trajectory.
+    
+    Slots are the unit of scheduling - each trajectory runs in its own slot,
+    with an isolated workspace directory. Multiple slots share a container.
+    
+    Attributes:
+        slot_id: Unique identifier for this slot (e.g., "slot_0")
+        alloc_id: Nomad allocation ID hosting this slot
+        container_addr: HTTP address of the sandbox server (e.g., "http://10.0.0.1:8080")
+        workspace_dir: Path to workspace in container (e.g., "/data/slot_0")
+        state: Current state of the slot
+        trajectory_id: ID of trajectory currently using this slot (if acquired)
+        metadata: Additional metadata
+    """
+    slot_id: str
+    alloc_id: str
+    container_addr: str
+    workspace_dir: str = ""
+    state: SlotState = SlotState.AVAILABLE
+    trajectory_id: Optional[str] = None
+    metadata: Dict[str, Any] = field(default_factory=dict)
+    
+    def __post_init__(self):
+        """Set default workspace_dir if not provided."""
+        if not self.workspace_dir:
+            self.workspace_dir = f"/data/{self.slot_id}"
+    
+    @property
+    def is_available(self) -> bool:
+        """Check if slot is available for acquisition."""
+        return self.state == SlotState.AVAILABLE
+    
+    @property
+    def is_acquired(self) -> bool:
+        """Check if slot is currently acquired."""
+        return self.state in (SlotState.ACQUIRED, SlotState.EXECUTING)
+    
+    def acquire(self, trajectory_id: Optional[str] = None) -> None:
+        """
+        Mark slot as acquired by a trajectory.
+        
+        Args:
+            trajectory_id: Optional ID of acquiring trajectory
+        """
+        if not self.is_available:
+            raise RuntimeError(f"Cannot acquire slot {self.slot_id}: state is {self.state}")
+        
+        self.state = SlotState.ACQUIRED
+        self.trajectory_id = trajectory_id or str(uuid.uuid4())
+    
+    def start_execution(self, execution_id: Optional[str] = None) -> None:
+        """Mark slot as executing."""
+        if self.state != SlotState.ACQUIRED:
+            raise RuntimeError(f"Cannot start execution on slot {self.slot_id}: state is {self.state}")
+        
+        self.state = SlotState.EXECUTING
+        if execution_id:
+            self.metadata["current_execution_id"] = execution_id
+    
+    def end_execution(self) -> None:
+        """Mark execution as complete, return to acquired state."""
+        if self.state != SlotState.EXECUTING:
+            raise RuntimeError(f"Cannot end execution on slot {self.slot_id}: state is {self.state}")
+        
+        self.state = SlotState.ACQUIRED
+        self.metadata.pop("current_execution_id", None)
+    
+    def release(self) -> None:
+        """Release slot back to available state."""
+        self.state = SlotState.AVAILABLE
+        self.trajectory_id = None
+        self.metadata.pop("current_execution_id", None)
+    
+    def mark_error(self, error: str) -> None:
+        """Mark slot as in error state."""
+        self.state = SlotState.ERROR
+        self.metadata["error"] = error
+    
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert to dictionary for serialization."""
+        return {
+            "slot_id": self.slot_id,
+            "alloc_id": self.alloc_id,
+            "container_addr": self.container_addr,
+            "workspace_dir": self.workspace_dir,
+            "state": self.state.value,
+            "trajectory_id": self.trajectory_id,
+            "metadata": self.metadata,
+        }
+    
+    @classmethod
+    def from_dict(cls, data: Dict[str, Any]) -> "Slot":
+        """Create from dictionary."""
+        return cls(
+            slot_id=data["slot_id"],
+            alloc_id=data["alloc_id"],
+            container_addr=data["container_addr"],
+            workspace_dir=data.get("workspace_dir", ""),
+            state=SlotState(data.get("state", "available")),
+            trajectory_id=data.get("trajectory_id"),
+            metadata=data.get("metadata", {}),
+        )
+    
+    def __repr__(self) -> str:
+        return f"Slot({self.slot_id}, state={self.state.value}, alloc={self.alloc_id[:8]}...)"
+
+
+def create_slots_for_allocation(
+    alloc_id: str,
+    container_addr: str,
+    num_slots: int = 10,
+) -> list["Slot"]:
+    """
+    Create slots for a Nomad allocation.
+    
+    Args:
+        alloc_id: Nomad allocation ID
+        container_addr: HTTP address of sandbox server
+        num_slots: Number of slots to create
+        
+    Returns:
+        List of Slot objects
+    """
+    slots = []
+    for i in range(num_slots):
+        slot_id = f"slot_{i}"
+        slots.append(Slot(
+            slot_id=slot_id,
+            alloc_id=alloc_id,
+            container_addr=container_addr,
+            workspace_dir=f"/data/{slot_id}",
+        ))
+    return slots

atropos/terminal/__init__.py

@@ -0,0 +1,2 @@
+"""Terminal helpers for stateful sandbox interactions."""
+

atropos/terminal/asciinema_stream.py

@@ -0,0 +1,115 @@
+from __future__ import annotations
+
+import json
+from typing import Any
+
+import pyte
+
+
+class AsciinemaStreamDecoder:
+    def __init__(self, *, default_width: int = 80, default_height: int = 24) -> None:
+        self._default_width = max(1, int(default_width))
+        self._default_height = max(1, int(default_height))
+        self._buffer = ""
+        self._has_header = False
+        self.width = self._default_width
+        self.height = self._default_height
+        self._screen = pyte.Screen(self.width, self.height)
+        self._stream = pyte.Stream(self._screen)
+
+    def reset(self) -> None:
+        self._buffer = ""
+        self._has_header = False
+        self.width = self._default_width
+        self.height = self._default_height
+        self._screen = pyte.Screen(self.width, self.height)
+        self._stream = pyte.Stream(self._screen)
+
+    def feed(self, chunk: str | bytes) -> None:
+        if not chunk:
+            return
+        if isinstance(chunk, bytes):
+            chunk = chunk.decode("utf-8", errors="replace")
+        self._buffer += chunk
+        while True:
+            line, sep, rest = self._buffer.partition("\n")
+            if not sep:
+                break
+            self._buffer = rest
+            line = line.strip()
+            if not line:
+                continue
+            parsed = self._parse_json_line(line)
+            if parsed is None:
+                continue
+            if not self._has_header:
+                if isinstance(parsed, dict):
+                    self._init_from_header(parsed)
+                    continue
+                if isinstance(parsed, list):
+                    self._has_header = True
+                    self._apply_event(parsed)
+                    continue
+                continue
+            if isinstance(parsed, list):
+                self._apply_event(parsed)
+
+    def render(self) -> str:
+        return "\n".join(self._screen.display)
+
+    def _parse_json_line(self, line: str) -> Any | None:
+        try:
+            return json.loads(line)
+        except json.JSONDecodeError:
+            return None
+
+    def _init_from_header(self, header: dict[str, Any]) -> None:
+        width = _coerce_int(
+            header.get("width") or header.get("columns") or header.get("cols"),
+            self._default_width,
+        )
+        height = _coerce_int(
+            header.get("height") or header.get("rows") or header.get("lines"),
+            self._default_height,
+        )
+        self.width = max(1, width)
+        self.height = max(1, height)
+        self._screen = pyte.Screen(self.width, self.height)
+        self._stream = pyte.Stream(self._screen)
+        self._has_header = True
+
+    def _apply_event(self, event: list[Any]) -> None:
+        if len(event) < 2:
+            return
+        event_type = event[1]
+        payload = event[2] if len(event) > 2 else ""
+        if event_type == "o":
+            if isinstance(payload, str):
+                self._stream.feed(payload)
+        elif event_type == "r":
+            width, height = _parse_resize(payload)
+            if width and height:
+                self.width = width
+                self.height = height
+                self._screen.resize(width, height)
+
+
+def _coerce_int(value: Any, default: int) -> int:
+    try:
+        return int(value)
+    except (TypeError, ValueError):
+        return int(default)
+
+
+def _parse_resize(payload: Any) -> tuple[int, int]:
+    if isinstance(payload, str) and "x" in payload:
+        left, right = payload.lower().split("x", 1)
+        return _coerce_int(left, 0), _coerce_int(right, 0)
+    if isinstance(payload, dict):
+        width = _coerce_int(payload.get("width") or payload.get("columns") or payload.get("cols"), 0)
+        height = _coerce_int(payload.get("height") or payload.get("rows") or payload.get("lines"), 0)
+        return width, height
+    if isinstance(payload, list) and len(payload) >= 2:
+        return _coerce_int(payload[0], 0), _coerce_int(payload[1], 0)
+    return 0, 0
+

atropos/tools/__init__.py

@@ -0,0 +1,26 @@
+"""
+Tool abstractions for atropos-agent.
+
+Provides base Tool class and common tool implementations.
+"""
+
+from .base import Tool, ToolCall, ToolRegistry, ToolResult, ToolSchema
+from .build_registry import build_tool_registry
+from .sandbox_stubs import BashTool, ReadFileTool, TerminalTool, WriteFileTool
+from .terminal_stateful_tool import TerminalStatefulTool
+from .tmux_tool import TmuxTool
+
+__all__ = [
+    "Tool",
+    "ToolCall",
+    "ToolRegistry",
+    "ToolResult",
+    "ToolSchema",
+    "BashTool",
+    "ReadFileTool",
+    "WriteFileTool",
+    "TerminalTool",
+    "TerminalStatefulTool",
+    "TmuxTool",
+    "build_tool_registry",
+]

atropos/tools/base.py

@@ -0,0 +1,423 @@
+"""
+Base Tool abstraction for atropos-agent.
+
+Tools follow a simple pattern:
+1. Define schema (name, description, parameters)
+2. Implement execute() method
+3. Return ToolResult with output/error
+
+Tool calls use Hermes-style XML tags:
+<tool_call>{"name": "bash", "arguments": {"command": "ls"}}</tool_call>
+"""
+
+import json
+import re
+import uuid
+from abc import ABC, abstractmethod
+from dataclasses import dataclass, field
+from typing import Any, Dict, List, Literal, Optional
+
+from pydantic import BaseModel, Field
+
+
+@dataclass
+class ToolSchema:
+    """JSON Schema for a tool's parameters."""
+    
+    name: str
+    description: str
+    parameters: Dict[str, Any] = field(default_factory=dict)
+    required: List[str] = field(default_factory=list)
+    external: bool = False  # Whether the tool must be executed via an external ToolServer (secret proxy) and not inside the sandbox.
+    
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert to OpenAI-compatible function schema."""
+        return {
+            "type": "function",
+            "function": {
+                "name": self.name,
+                "description": self.description,
+                "parameters": {
+                    "type": "object",
+                    "properties": self.parameters,
+                    "required": self.required,
+                },
+            },
+        }
+    
+    def to_prompt_description(self) -> str:
+        """Convert to human-readable description for system prompt."""
+        params_desc = []
+        for name, spec in self.parameters.items():
+            req = "(required)" if name in self.required else "(optional)"
+            desc = spec.get("description", "")
+            param_type = spec.get("type", "string")
+            params_desc.append(f"  - {name} ({param_type}) {req}: {desc}")
+        
+        params_str = "\n".join(params_desc) if params_desc else "  (no parameters)"
+        return f"**{self.name}**: {self.description}\nParameters:\n{params_str}"
+
+
+@dataclass
+class ToolCall:
+    """A parsed tool call from model output."""
+    
+    name: str
+    arguments: Dict[str, Any]
+    raw_text: str = ""  # Original XML/JSON text
+    uniq_id: str = field(default_factory=lambda: str(uuid.uuid4()))  # Unique tool-call id for traceability/reconstruction.
+    
+    @classmethod
+    def parse_from_text(cls, text: str) -> List["ToolCall"]:
+        """
+        Extract tool calls from text using Hermes-style XML tags.
+        
+        Supported formats (STRICT: requires well-formed closing tags):
+        - Hermes JSON wrapper:
+          <tool_call>{"name": "...", "arguments": {...}}</tool_call>
+        - GLM/llama.cpp style:
+          <tool_call>terminal{"command":"ls -la"}</tool_call>
+        """
+        calls: List["ToolCall"] = []
+
+        if not text:
+            return calls
+
+        def _append_from_payload(*, name: str, arguments: Dict[str, Any], raw: str, uniq_id: Optional[str] = None) -> None:
+            if not isinstance(name, str) or not name:
+                return
+            if not isinstance(arguments, dict):
+                return
+            calls.append(
+                cls(
+                    name=name,
+                    arguments=arguments,
+                    raw_text=raw,
+                    uniq_id=uniq_id or str(uuid.uuid4()),
+                )
+            )
+
+        # STRICT parsing: only accept well-formed <tool_call>...</tool_call> blocks.
+        pattern = r"<tool_call>\s*(.*?)\s*</tool_call>"
+        for inner in re.findall(pattern, text, re.DOTALL):
+            cleaned = (inner or "").strip()
+            if not cleaned:
+                continue
+
+            # Hermes JSON wrapper.
+            if cleaned.startswith("{"):
+                try:
+                    data = json.loads(cleaned)
+                except json.JSONDecodeError:
+                    continue
+                uniq_id = data.get("uniq_id") or data.get("id") or None
+                _append_from_payload(
+                    name=data.get("name", ""),
+                    arguments=data.get("arguments", {}),
+                    raw=inner,
+                    uniq_id=uniq_id,
+                )
+                continue
+
+            # GLM/llama.cpp style: terminal{...}
+            m = re.match(r"^\s*([A-Za-z0-9_.:\\-]+)\s*(\{.*\})\s*$", cleaned, re.DOTALL)
+            if not m:
+                continue
+            name = m.group(1)
+            args_text = m.group(2)
+            try:
+                args = json.loads(args_text)
+            except json.JSONDecodeError:
+                continue
+            _append_from_payload(name=name, arguments=args, raw=inner)
+
+        return calls
+    
+    @classmethod
+    def has_tool_call(cls, text: str) -> bool:
+        """Check if text contains any tool calls."""
+        return bool(re.search(r"<tool_call>", text))
+
+
+@dataclass
+class ToolResult:
+    """Result from executing a tool."""
+    
+    success: bool
+    output: str = ""
+    error: str = ""
+    metadata: Dict[str, Any] = field(default_factory=dict)
+    uniq_id: Optional[str] = None  # Should match ToolCall.uniq_id for async execution tracking.
+    
+    def to_xml(self) -> str:
+        """Format as XML for including in conversation."""
+        data = {
+            "success": self.success,
+            "output": self.output,
+        }
+        if self.uniq_id:
+            data["uniq_id"] = self.uniq_id
+        if self.error:
+            data["error"] = self.error
+        if self.metadata:
+            data["metadata"] = self.metadata
+        return f"<tool_response>{json.dumps(data)}</tool_response>"
+    
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert to dictionary."""
+        return {
+            "success": self.success,
+            "output": self.output,
+            "error": self.error,
+            "metadata": self.metadata,
+            "uniq_id": self.uniq_id,
+        }
+
+
+class Tool(ABC):
+    """
+    Abstract base class for tools.
+    
+    Subclasses must implement:
+    - schema: ToolSchema describing the tool
+    - execute(): async method that performs the tool action
+    """
+    
+    @property
+    @abstractmethod
+    def schema(self) -> ToolSchema:
+        """Return the tool's schema."""
+        pass
+    
+    @property
+    def name(self) -> str:
+        """Tool name (from schema)."""
+        return self.schema.name
+    
+    @abstractmethod
+    async def execute(self, **kwargs) -> ToolResult:
+        """
+        Execute the tool with given arguments.
+        
+        Args:
+            **kwargs: Tool-specific arguments
+            
+        Returns:
+            ToolResult with success/failure and output
+        """
+        pass
+    
+    def is_available(self) -> tuple[bool, str | None]:
+        """
+        Return whether this tool should be exposed/executable in the current process.
+
+        Tools that depend on optional binaries/services/env vars can override this
+        to avoid advertising a tool that will fail at runtime.
+        """
+        return True, None
+
+    async def __call__(self, **kwargs) -> ToolResult:
+        """Allow calling tool instance directly."""
+        return await self.execute(**kwargs)
+
+# Note: This is only wrapping declarations for the external ToolServer (for execution on external process tools), and tools preinstalled in envs
+class ToolRegistry:
+    """Registry of available tools."""
+    
+    def __init__(self):
+        self._tools: Dict[str, Tool] = {}
+    
+    def register(self, tool: Tool) -> None:
+        """Register a tool."""
+        self._tools[tool.name] = tool
+    
+    def get(self, name: str) -> Optional[Tool]:
+        """Get a tool by name."""
+        return self._tools.get(name)
+    
+    def list_tools(self) -> List[Tool]:
+        """List all registered tools."""
+        return list(self._tools.values())
+    
+    def get_schemas(self) -> List[ToolSchema]:
+        """Get schemas for all registered tools."""
+        return [tool.schema for tool in self._tools.values()]
+    
+    def get_prompt_description(self) -> str:
+        """Generate tool descriptions for system prompt."""
+        descriptions = [tool.schema.to_prompt_description() for tool in self._tools.values()]
+        return "\n\n".join(descriptions)
+
+    def get_prompt_tool_definitions_json(self) -> str:
+        """
+        Return a Hermes-style JSON list of tool definitions for use inside a `<tools>...</tools>` block.
+
+        Hermes trajectories historically use a simplified schema list:
+          [{"name": ..., "description": ..., "parameters": {...}, "required": null}, ...]
+        """
+        formatted: List[Dict[str, Any]] = []
+        for tool in self._tools.values():
+            fn = tool.schema.to_dict().get("function", {})
+            formatted.append(
+                {
+                    "name": fn.get("name", tool.name),
+                    "description": fn.get("description", ""),
+                    "parameters": fn.get("parameters", {}),
+                    # Keep parity with Hermes saved trajectories (required is typically null there).
+                    "required": None,
+                }
+            )
+        return json.dumps(formatted, ensure_ascii=False)
+    
+    async def execute(self, call: ToolCall) -> ToolResult:
+        """Execute a tool call."""
+        tool = self.get(call.name)
+        if tool is None:
+            return ToolResult(
+                success=False,
+                error=f"Unknown tool: {call.name}",
+                uniq_id=call.uniq_id,
+            )
+        
+        try:
+            result = await tool.execute(**call.arguments)
+            if result.uniq_id is None:
+                result.uniq_id = call.uniq_id
+            return result
+        except Exception as e:
+            return ToolResult(
+                success=False,
+                error=f"Tool execution error: {str(e)}",
+                uniq_id=call.uniq_id,
+            )
+
+
+# =============================================================================
+# FastAPI / transport models
+# =============================================================================
+
+
+class ToolCallPayload(BaseModel):
+    name: str
+    arguments: Dict[str, Any] = Field(default_factory=dict)
+    uniq_id: str
+
+    @classmethod
+    def from_tool_call(cls, call: ToolCall) -> "ToolCallPayload":
+        return cls(name=call.name, arguments=call.arguments, uniq_id=call.uniq_id)
+
+    def to_tool_call(self) -> ToolCall:
+        return ToolCall(name=self.name, arguments=self.arguments, uniq_id=self.uniq_id)
+
+
+class ToolResultPayload(BaseModel):
+    success: bool
+    output: str = ""
+    error: str = ""
+    metadata: Dict[str, Any] = Field(default_factory=dict)
+    uniq_id: Optional[str] = None
+
+    @classmethod
+    def from_tool_result(cls, result: ToolResult) -> "ToolResultPayload":
+        return cls(
+            success=result.success,
+            output=result.output,
+            error=result.error,
+            metadata=result.metadata,
+            uniq_id=result.uniq_id,
+        )
+
+    def to_tool_result(self) -> ToolResult:
+        return ToolResult(
+            success=self.success,
+            output=self.output,
+            error=self.error,
+            metadata=self.metadata,
+            uniq_id=self.uniq_id,
+        )
+
+
+class ToolExecutorExecuteRequest(BaseModel):
+    trajectory_id: str
+    tool: ToolCallPayload
+    timeout_s: Optional[float] = None
+
+
+class ToolExecutorReleaseRequest(BaseModel):
+    trajectory_id: str
+    reset_workspace: bool = False
+
+
+class ToolServerExecuteRequest(BaseModel):
+    trajectory_id: Optional[str] = None
+    tool: ToolCallPayload
+    timeout_s: Optional[float] = None
+    # Optional sandbox context for tools that need workspace artifacts.
+    # This is set by ToolExecutor and is NOT model-controlled.
+    slot_id: Optional[str] = None
+    container_addr: Optional[str] = None
+
+
+# =============================================================================
+# Artifact transport models
+# =============================================================================
+
+
+class ArtifactReadRequestPayload(BaseModel):
+    trajectory_id: str
+    path: str
+    encoding: Literal["text", "base64"] = "text"
+    max_bytes: Optional[int] = None
+    include_sha256: bool = False
+
+
+class ArtifactReadResponsePayload(BaseModel):
+    success: bool
+    content: str = ""
+    error: str = ""
+    encoding: str = "text"
+    truncated: bool = False
+    bytes: int = 0
+    file_size: Optional[int] = None
+    path: str = ""
+    mime: Optional[str] = None
+    sha256: Optional[str] = None
+
+
+class ArtifactListRequestPayload(BaseModel):
+    trajectory_id: str
+    path: str = "."
+    recursive: bool = False
+    max_entries: Optional[int] = None
+
+
+class ArtifactListEntryPayload(BaseModel):
+    path: str
+    is_dir: bool
+    size: int
+    mtime: float
+
+
+class ArtifactListResponsePayload(BaseModel):
+    success: bool
+    entries: List[ArtifactListEntryPayload] = Field(default_factory=list)
+    truncated: bool = False
+    error: str = ""
+
+
+class ArtifactArchiveRequestPayload(BaseModel):
+    trajectory_id: str
+    path: str = "."
+    format: Literal["tar.gz", "tgz"] = "tar.gz"
+    max_bytes: Optional[int] = None
+    max_entries: Optional[int] = None
+
+
+class ArtifactArchiveResponsePayload(BaseModel):
+    success: bool
+    content: str = ""
+    error: str = ""
+    encoding: str = "base64"
+    format: str = "tar.gz"
+    bytes: int = 0
+    entry_count: int = 0

atropos/tools/build_registry.py

@@ -0,0 +1,64 @@
+"""
+Unified tool registry builder for Hermes-Agent Atropos integration.
+
+This composes:
+- sandbox tool stubs (terminal/bash/read_file/write_file + stateful terminal/tmux)
+- Hermes external tools (web/vision/image/moa/skills/browser), executed via ToolServer
+
+ToolExecutor only needs the schema + `external` routing bit; ToolServer executes
+the external tools via Hermes' existing implementations.
+"""
+
+from __future__ import annotations
+
+from typing import List, Optional
+
+from .base import ToolRegistry
+from .hermes_external_tools import build_external_tools
+from .sandbox_stubs import BashTool, ReadFileTool, TerminalTool, WriteFileTool
+from .terminal_stateful_tool import TerminalStatefulTool
+from .tmux_tool import TmuxTool
+from .toolset_resolver import resolve_multiple_toolsets
+
+
+def build_tool_registry(
+    *,
+    enabled_toolsets: Optional[List[str]] = None,
+    disabled_toolsets: Optional[List[str]] = None,
+    tool_server_url: Optional[str] = None,
+) -> ToolRegistry:
+    """
+    Build a ToolRegistry for AgentEnv / ToolExecutor / ToolServer.
+
+    If `tool_server_url` is not provided, external tools will be omitted so we do
+    not advertise tools that cannot execute.
+    """
+    enabled_toolsets = enabled_toolsets or ["default"]
+
+    # Resolve tool names using Hermes toolsets plus Atropos additions.
+    selected = set(resolve_multiple_toolsets(enabled_toolsets))
+    if disabled_toolsets:
+        selected -= set(resolve_multiple_toolsets(disabled_toolsets))
+
+    reg = ToolRegistry()
+
+    # Always register sandbox tools if selected.
+    sandbox_by_name = {
+        "terminal": TerminalTool(),
+        "bash": BashTool(),
+        "read_file": ReadFileTool(),
+        "write_file": WriteFileTool(),
+        "terminal_stateful": TerminalStatefulTool(),
+        "tmux": TmuxTool(),
+    }
+    for name, tool in sandbox_by_name.items():
+        if name in selected:
+            reg.register(tool)
+
+    # External tools: only include when ToolServer is configured.
+    if tool_server_url:
+        for tool in build_external_tools(selected_tool_names=selected):
+            if tool.name in selected:
+                reg.register(tool)
+
+    return reg

atropos/tools/hermes_external_tools.py

@@ -0,0 +1,90 @@
+"""
+Hermes external tool adapter for Atropos ToolServer.
+
+These tools reuse Hermes-Agent's existing tool runner (`model_tools.handle_function_call`)
+so we don't duplicate external tool implementations.
+
+Important:
+- These are marked `external=True` and should be executed ONLY by ToolServer.
+- We run `handle_function_call` in a worker thread because the Hermes implementation
+  uses `asyncio.run()` internally for some async tools (web_extract, vision, MoA, etc).
+"""
+
+from __future__ import annotations
+
+import asyncio
+import json
+from typing import Any, Dict, List, Optional
+
+import model_tools
+
+from .base import Tool, ToolResult, ToolSchema
+
+
+def _schema_from_openai_tool_dict(tool: Dict[str, Any], *, external: bool) -> ToolSchema:
+    fn = tool.get("function") or {}
+    name = str(fn.get("name") or "")
+    description = str(fn.get("description") or "")
+    params = fn.get("parameters") or {}
+    properties = params.get("properties") or {}
+    required = params.get("required") or []
+    if not isinstance(required, list):
+        required = []
+    return ToolSchema(
+        name=name,
+        description=description,
+        parameters=dict(properties),
+        required=[str(x) for x in required if isinstance(x, (str, int))],
+        external=external,
+    )
+
+
+class HermesExternalTool(Tool):
+    def __init__(self, schema: ToolSchema):
+        self._schema = schema
+
+    @property
+    def schema(self) -> ToolSchema:
+        return self._schema
+
+    async def execute(self, task_id: Optional[str] = None, **kwargs: Any) -> ToolResult:
+        # `model_tools.handle_function_call` returns a JSON string (success or error).
+        # Run in a thread because some Hermes tool handlers call `asyncio.run()`.
+        raw = await asyncio.to_thread(model_tools.handle_function_call, self.name, kwargs, task_id)
+
+        try:
+            parsed = json.loads(raw)
+        except Exception:
+            # Keep as plain string.
+            return ToolResult(success=True, output=str(raw))
+
+        if isinstance(parsed, dict) and parsed.get("error"):
+            return ToolResult(success=False, error=str(parsed.get("error")), output="")
+
+        return ToolResult(success=True, output=json.dumps(parsed, ensure_ascii=False))
+
+
+def build_external_tools(
+    *,
+    selected_tool_names: Optional[set[str]] = None,
+) -> List[HermesExternalTool]:
+    """
+    Build external tool wrappers from Hermes tool declarations.
+
+    Filters out sandbox-oriented tools (e.g. `terminal`) since those should run
+    inside the sandbox via ToolExecutor.
+    """
+    # IMPORTANT: Hermes' `model_tools.get_tool_definitions()` only understands Hermes toolsets.
+    # Atropos envs add extra toolsets (filesystem/sandbox/stateful). To avoid noisy "Unknown toolset"
+    # prints and accidental filtering, we fetch ALL Hermes tool definitions here and filter by name.
+    tools = model_tools.get_tool_definitions(enabled_toolsets=None, disabled_toolsets=None, quiet_mode=True)
+
+    wrappers: List[HermesExternalTool] = []
+    for t in tools:
+        schema = _schema_from_openai_tool_dict(t, external=True)
+        if schema.name in {"terminal"}:
+            continue
+        if selected_tool_names is not None and schema.name not in selected_tool_names:
+            continue
+        wrappers.append(HermesExternalTool(schema))
+    return wrappers

atropos/tools/sandbox_stubs.py

@@ -0,0 +1,99 @@
+"""
+Sandbox tool stubs for Atropos ToolExecutor.
+
+These tools are executed inside the sandbox containers via:
+ToolExecutor -> SlotPool -> sandbox_server.py
+
+They intentionally do NOT execute anything on the host process. If they are
+called directly (outside ToolExecutor), they return a clear error.
+"""
+
+from __future__ import annotations
+
+from typing import Optional
+
+from .base import Tool, ToolResult, ToolSchema
+
+
+class TerminalTool(Tool):
+    @property
+    def schema(self) -> ToolSchema:
+        return ToolSchema(
+            name="terminal",
+            description=(
+                "Execute a command inside the sandbox slot workspace and return stdout/stderr. "
+                "Filesystem persists within a trajectory slot. Background processes are not supported "
+                "in stateless mode. Commands run under POSIX /bin/sh and each tool call runs in a fresh "
+                "shell (no persisted env vars). Avoid bash-only syntax like `source`; prefer `. .venv/bin/activate` "
+                "or invoke `.venv/bin/python ...` directly."
+            ),
+            parameters={
+                "command": {"type": "string", "description": "The command to execute"},
+                "timeout": {
+                    "type": "integer",
+                    "description": "Command timeout in seconds (optional).",
+                    "minimum": 1,
+                },
+                "background": {
+                    "type": "boolean",
+                    "description": "Not supported in sandbox terminal (always false).",
+                    "default": False,
+                },
+            },
+            required=["command"],
+            external=False,
+        )
+
+    async def execute(self, **_kwargs) -> ToolResult:
+        return ToolResult(
+            success=False,
+            error="terminal must be executed via ToolExecutor inside the sandbox",
+        )
+
+
+class BashTool(Tool):
+    @property
+    def schema(self) -> ToolSchema:
+        return ToolSchema(
+            name="bash",
+            description="Execute a bash command inside the sandbox slot workspace.",
+            parameters={"command": {"type": "string", "description": "The bash command to execute"}},
+            required=["command"],
+            external=False,
+        )
+
+    async def execute(self, **_kwargs) -> ToolResult:
+        return ToolResult(success=False, error="bash must be executed via ToolExecutor inside the sandbox")
+
+
+class ReadFileTool(Tool):
+    @property
+    def schema(self) -> ToolSchema:
+        return ToolSchema(
+            name="read_file",
+            description="Read a file from the sandbox slot workspace.",
+            parameters={"path": {"type": "string", "description": "Path to the file"}},
+            required=["path"],
+            external=False,
+        )
+
+    async def execute(self, **_kwargs) -> ToolResult:
+        return ToolResult(success=False, error="read_file must be executed via ToolExecutor inside the sandbox")
+
+
+class WriteFileTool(Tool):
+    @property
+    def schema(self) -> ToolSchema:
+        return ToolSchema(
+            name="write_file",
+            description="Write a file into the sandbox slot workspace.",
+            parameters={
+                "path": {"type": "string", "description": "Path to the file"},
+                "content": {"type": "string", "description": "File content"},
+            },
+            required=["path", "content"],
+            external=False,
+        )
+
+    async def execute(self, **_kwargs) -> ToolResult:
+        return ToolResult(success=False, error="write_file must be executed via ToolExecutor inside the sandbox")

atropos/tools/terminal_stateful_tool.py

@@ -0,0 +1,45 @@
+"""
+Stateful terminal tool schema.
+
+This is a sandbox tool that routes to the sandbox server as `bash_stateful`
+via ToolExecutor mapping. It exists to expose an explicit, opt-in terminal
+primitive suitable for stateful workflows (e.g. tmux sessions / TUIs).
+"""
+
+from __future__ import annotations
+
+from typing import Optional
+
+from .base import Tool, ToolResult, ToolSchema
+
+
+class TerminalStatefulTool(Tool):
+    @property
+    def schema(self) -> ToolSchema:
+        return ToolSchema(
+            name="terminal_stateful",
+            description=(
+                "Execute a command in the sandbox, allowing stateful/background processes to persist "
+                "across tool calls within the same trajectory slot (e.g. tmux sessions). "
+                "Use sparingly; output is still non-interactive."
+            ),
+            parameters={
+                "command": {"type": "string", "description": "The command to execute"},
+                "timeout": {
+                    "type": "integer",
+                    "description": "Command timeout in seconds (optional).",
+                    "minimum": 1,
+                },
+            },
+            required=["command"],
+        )
+
+    def is_available(self) -> tuple[bool, str | None]:
+        return True, None
+
+    async def execute(self, command: str, timeout: Optional[int] = None) -> ToolResult:
+        _ = (command, timeout)
+        return ToolResult(
+            success=False,
+            error="terminal_stateful must be executed via ToolExecutor inside the sandbox",
+        )

atropos/tools/tmux_tool.py

@@ -0,0 +1,89 @@
+"""
+tmux tool schema (sandbox).
+
+This is a sandbox tool that provides basic tmux session control suitable for
+TUI-style terminal interactions:
+- send keys (arrow keys, enter, etc.)
+- capture the current screen buffer
+
+Execution is routed by ToolExecutor to the sandbox server's `tmux` backend.
+"""
+
+from __future__ import annotations
+
+from typing import Any, Dict, Optional
+
+from .base import Tool, ToolResult, ToolSchema
+
+
+class TmuxTool(Tool):
+    @property
+    def schema(self) -> ToolSchema:
+        return ToolSchema(
+            name="tmux",
+            description=(
+                "Control a per-trajectory tmux session inside the sandbox (stateful terminal). "
+                "Use this for TUI-style interactions: send keys and capture the current screen."
+            ),
+            parameters={
+                "action": {
+                    "type": "string",
+                    "description": "Action to perform: start | send_keys | stream | stop.",
+                    "enum": ["start", "send_keys", "stream", "stop", "capture"],
+                },
+                "keys": {
+                    "description": "Keys to send (string or list of strings) when action=send_keys.",
+                },
+                "block": {
+                    "type": "boolean",
+                    "description": "If true, wait for shell command completion (only valid at a shell prompt).",
+                    "default": False,
+                },
+                "min_wait_s": {
+                    "type": "number",
+                    "description": "For non-blocking send_keys, sleep this long after sending keys (seconds).",
+                    "default": 0.0,
+                },
+                "max_wait_s": {
+                    "type": "number",
+                    "description": "For blocking send_keys, max time to wait for completion (seconds).",
+                },
+                "capture_entire": {
+                    "type": "boolean",
+                    "description": "Deprecated. Streaming is preferred.",
+                    "default": False,
+                },
+                "max_bytes": {
+                    "type": "integer",
+                    "description": "Max bytes to return per stream call.",
+                },
+                "reset": {
+                    "type": "boolean",
+                    "description": "If true, reset stream offset to the beginning of the asciinema recording.",
+                    "default": False,
+                },
+                "pane_width": {
+                    "type": "integer",
+                    "description": "Pane width for action=start (columns).",
+                    "minimum": 20,
+                },
+                "pane_height": {
+                    "type": "integer",
+                    "description": "Pane height for action=start (rows).",
+                    "minimum": 10,
+                },
+            },
+            required=["action"],
+        )
+
+    def is_available(self) -> tuple[bool, str | None]:
+        return True, None
+
+    async def execute(self, **kwargs: Dict[str, Any]) -> ToolResult:
+        # This tool is intended to be executed via ToolExecutor -> sandbox server.
+        # We keep a safe fallback for non-sandbox contexts.
+        action = str(kwargs.get("action") or "").strip()
+        return ToolResult(
+            success=False,
+            error=f"tmux tool must be executed in the sandbox (got action={action!r})",
+        )

atropos/tools/tool_executor.py

@@ -0,0 +1,500 @@
+"""
+ToolExecutor - queued, batched tool dispatch for multiplexed agent trajectories.
+
+This component is responsible for:
+- Maintaining trajectory -> Slot affinity (workspace continuity)
+- Batching sandbox tool calls across trajectories to maximize container utilization
+- Routing external tools (ToolSchema.external=True) to a ToolServer (Phase 4.5)
+
+For now, only sandbox tools are executed:
+- bash
+- read_file
+- write_file
+"""
+
+from __future__ import annotations
+
+import asyncio
+import time
+from dataclasses import dataclass
+from typing import Any, Dict, List, Optional
+
+import httpx
+
+from .base import (
+    ArtifactArchiveRequestPayload,
+    ArtifactArchiveResponsePayload,
+    ArtifactListRequestPayload,
+    ArtifactListResponsePayload,
+    ArtifactReadRequestPayload,
+    ArtifactReadResponsePayload,
+    ToolCall,
+    ToolCallPayload,
+    ToolRegistry,
+    ToolResult,
+    ToolResultPayload,
+    ToolServerExecuteRequest,
+)
+from ..backends.base import ToolBackend
+from ..slots import Slot
+
+
+@dataclass
+class ToolExecutorConfig:
+    batch_window_ms: int = 20
+    max_batch_size: int = 200
+    allow_network: bool = True
+    require_sandbox: bool = False
+    require_stateful_sandbox: bool = False
+    tool_server_url: Optional[str] = None
+    tool_server_token: Optional[str] = None
+
+
+@dataclass
+class _QueuedToolRequest:
+    trajectory_id: str
+    call: ToolCall
+    timeout_s: Optional[float]
+    future: asyncio.Future
+
+
+class ToolExecutor:
+    def __init__(
+        self,
+        backend: ToolBackend,
+        tools: ToolRegistry,
+        config: Optional[ToolExecutorConfig] = None,
+    ) -> None:
+        self.backend = backend
+        self.tools = tools
+        self.config = config or ToolExecutorConfig()
+
+        self._queue: asyncio.Queue[Optional[_QueuedToolRequest]] = asyncio.Queue()
+        self._task: Optional[asyncio.Task] = None
+        self._stopping = asyncio.Event()
+
+        self._slots_lock = asyncio.Lock()
+        self._slot_by_trajectory: Dict[str, Slot] = {}
+
+        self._tool_server_client: Optional[httpx.AsyncClient] = None
+        self._tool_server_lock = asyncio.Lock()
+
+        # lightweight stats for status endpoints
+        self.total_requests: int = 0
+        self.total_errors: int = 0
+        self.latencies_s: List[float] = []
+
+    async def start(self) -> None:
+        if self._task is None:
+            self._task = asyncio.create_task(self._run_loop())
+
+    def queue_size(self) -> int:
+        return self._queue.qsize()
+
+    async def close(self) -> None:
+        self._stopping.set()
+        await self._queue.put(None)
+        if self._task:
+            await self._task
+            self._task = None
+
+        client = self._tool_server_client
+        self._tool_server_client = None
+        if client is not None:
+            await client.aclose()
+
+        # Best-effort release any remaining slots.
+        async with self._slots_lock:
+            slots = list(self._slot_by_trajectory.items())
+            self._slot_by_trajectory.clear()
+
+        for _, slot in slots:
+            try:
+                await self.backend.release(slot, reset_workspace=False)
+            except Exception:
+                pass
+
+    async def execute(
+        self,
+        trajectory_id: str,
+        call: ToolCall,
+        timeout_s: Optional[float] = None,
+    ) -> ToolResult:
+        if self._task is None:
+            raise RuntimeError("ToolExecutor not started (call start() first)")
+
+        # Allow tool args to suggest a timeout (Hermes-compatible terminal tool),
+        # but never let the model choose "infinite" timeouts.
+        if timeout_s is None:
+            raw_timeout = call.arguments.get("timeout")
+            if isinstance(raw_timeout, (int, float)):
+                timeout_s = float(raw_timeout)
+        if timeout_s is not None:
+            timeout_s = max(1.0, min(float(timeout_s), 600.0))
+
+        loop = asyncio.get_running_loop()
+        fut: asyncio.Future = loop.create_future()
+        started = time.perf_counter()
+        await self._queue.put(_QueuedToolRequest(trajectory_id=trajectory_id, call=call, timeout_s=timeout_s, future=fut))
+        try:
+            result: ToolResult = await fut
+            return result
+        finally:
+            self.latencies_s.append(time.perf_counter() - started)
+
+    async def release_trajectory(self, trajectory_id: str, reset_workspace: bool = False) -> None:
+        async with self._slots_lock:
+            slot = self._slot_by_trajectory.pop(trajectory_id, None)
+
+        if slot is not None:
+            await self.backend.release(slot, reset_workspace=reset_workspace)
+
+    async def _get_slot_if_present(self, trajectory_id: str) -> Optional[Slot]:
+        async with self._slots_lock:
+            return self._slot_by_trajectory.get(trajectory_id)
+
+    # ---------------------------------------------------------------------
+    # Artifact helpers (optional)
+    # ---------------------------------------------------------------------
+
+    async def read_artifact(self, req: ArtifactReadRequestPayload) -> ArtifactReadResponsePayload:
+        slot = await self._get_slot_if_present(req.trajectory_id)
+        if slot is None:
+            return ArtifactReadResponsePayload(success=False, error="No active slot for trajectory (run a sandbox tool first)")
+        data = await self.backend.read_artifact(
+            slot,
+            req.path,
+            encoding=req.encoding,
+            max_bytes=req.max_bytes,
+            include_sha256=req.include_sha256,
+        )
+        if isinstance(data, dict):
+            data = dict(data)
+            data.pop("http_status", None)
+        try:
+            return ArtifactReadResponsePayload(**(data or {}))
+        except Exception as e:
+            return ArtifactReadResponsePayload(success=False, error=f"Invalid artifact read response: {e}")
+
+    async def list_artifacts(self, req: ArtifactListRequestPayload) -> ArtifactListResponsePayload:
+        slot = await self._get_slot_if_present(req.trajectory_id)
+        if slot is None:
+            return ArtifactListResponsePayload(success=False, error="No active slot for trajectory (run a sandbox tool first)")
+        data = await self.backend.list_artifacts(
+            slot,
+            req.path,
+            recursive=req.recursive,
+            max_entries=req.max_entries,
+        )
+        if isinstance(data, dict):
+            data = dict(data)
+            data.pop("http_status", None)
+        try:
+            return ArtifactListResponsePayload(**(data or {}))
+        except Exception as e:
+            return ArtifactListResponsePayload(success=False, error=f"Invalid artifact list response: {e}")
+
+    async def archive_artifacts(self, req: ArtifactArchiveRequestPayload) -> ArtifactArchiveResponsePayload:
+        slot = await self._get_slot_if_present(req.trajectory_id)
+        if slot is None:
+            return ArtifactArchiveResponsePayload(success=False, error="No active slot for trajectory (run a sandbox tool first)")
+        data = await self.backend.archive_artifacts(
+            slot,
+            req.path,
+            archive_format=req.format,
+            max_bytes=req.max_bytes,
+            max_entries=req.max_entries,
+        )
+        if isinstance(data, dict):
+            data = dict(data)
+            data.pop("http_status", None)
+        try:
+            return ArtifactArchiveResponsePayload(**(data or {}))
+        except Exception as e:
+            return ArtifactArchiveResponsePayload(success=False, error=f"Invalid artifact archive response: {e}")
+
+    async def _get_or_acquire_slot(self, trajectory_id: str) -> Slot:
+        async with self._slots_lock:
+            existing = self._slot_by_trajectory.get(trajectory_id)
+            if existing is not None:
+                return existing
+
+        slot = await self.backend.acquire(trajectory_id)
+
+        async with self._slots_lock:
+            existing = self._slot_by_trajectory.get(trajectory_id)
+            if existing is not None:
+                # Another coroutine won the race; return its slot.
+                await self.backend.release(slot, reset_workspace=False)
+                return existing
+            self._slot_by_trajectory[trajectory_id] = slot
+            return slot
+
+    async def _run_loop(self) -> None:
+        pending: List[_QueuedToolRequest] = []
+        deadline: Optional[float] = None
+
+        batch_window_s = max(0.0, self.config.batch_window_ms / 1000.0)
+        max_batch = max(1, self.config.max_batch_size)
+
+        while True:
+            if self._stopping.is_set() and self._queue.empty() and not pending:
+                break
+
+            timeout = None
+            if pending and deadline is not None:
+                timeout = max(0.0, deadline - time.perf_counter())
+
+            try:
+                item = await asyncio.wait_for(self._queue.get(), timeout=timeout)
+                if item is None:
+                    continue
+                pending.append(item)
+                if len(pending) == 1:
+                    deadline = time.perf_counter() + batch_window_s
+                if len(pending) < max_batch:
+                    continue
+            except asyncio.TimeoutError:
+                # batch window elapsed
+                pass
+
+            if not pending:
+                deadline = None
+                continue
+
+            batch = pending
+            pending = []
+            deadline = None
+
+            await self._execute_batch(batch)
+
+    async def _get_tool_server_client(self) -> httpx.AsyncClient:
+        url = self.config.tool_server_url
+        if not url:
+            raise RuntimeError("ToolServer not configured")
+
+        if self._tool_server_client is not None:
+            return self._tool_server_client
+
+        async with self._tool_server_lock:
+            if self._tool_server_client is None:
+                self._tool_server_client = httpx.AsyncClient(base_url=url.rstrip("/"))
+            return self._tool_server_client
+
+    def _tool_server_headers(self) -> Dict[str, str]:
+        token = self.config.tool_server_token
+        if not token:
+            return {}
+        return {"Authorization": f"Bearer {token}"}
+
+    async def _execute_external(self, req: _QueuedToolRequest) -> ToolResult:
+        client = await self._get_tool_server_client()
+        slot_id: Optional[str] = None
+        container_addr: Optional[str] = None
+        slot = await self._get_slot_if_present(req.trajectory_id)
+        if slot is not None:
+            slot_id = slot.slot_id
+            container_addr = slot.container_addr
+
+        payload = ToolServerExecuteRequest(
+            trajectory_id=req.trajectory_id,
+            tool=ToolCallPayload.from_tool_call(req.call),
+            timeout_s=req.timeout_s,
+            slot_id=slot_id,
+            container_addr=container_addr,
+        )
+
+        try:
+            resp = await client.post(
+                "/execute",
+                json=payload.model_dump(),
+                headers=self._tool_server_headers(),
+                timeout=req.timeout_s,
+            )
+            resp.raise_for_status()
+            data = resp.json()
+            parsed = ToolResultPayload(**data)
+            result = parsed.to_tool_result()
+            if result.uniq_id is None:
+                result.uniq_id = req.call.uniq_id
+            return result
+        except Exception as e:
+            return ToolResult(
+                success=False,
+                error=f"External tool failed: {e}",
+                uniq_id=req.call.uniq_id,
+            )
+
+    async def _execute_batch(self, batch: List[_QueuedToolRequest]) -> None:
+        # Resolve tool schemas once per request and separate sandbox/external/unknown.
+        sandbox_items: List[_QueuedToolRequest] = []
+        external_items: List[_QueuedToolRequest] = []
+        unknown_items: List[_QueuedToolRequest] = []
+
+        for it in batch:
+            tool = self.tools.get(it.call.name)
+            if tool is None:
+                unknown_items.append(it)
+                continue
+
+            schema = tool.schema
+            if not schema.external:
+                sandbox_items.append(it)
+            else:
+                external_items.append(it)
+
+        for it in unknown_items:
+            self.total_requests += 1
+            self.total_errors += 1
+            if not it.future.done():
+                it.future.set_result(
+                    ToolResult(
+                        success=False,
+                        error=f"Unknown tool: {it.call.name}",
+                        uniq_id=it.call.uniq_id,
+                    )
+                )
+
+        if external_items:
+            if not self.config.tool_server_url:
+                for it in external_items:
+                    self.total_requests += 1
+                    self.total_errors += 1
+                    if not it.future.done():
+                        it.future.set_result(
+                            ToolResult(
+                                success=False,
+                                error=f"External tool not available (ToolServer not configured): {it.call.name}",
+                                uniq_id=it.call.uniq_id,
+                            )
+                        )
+            else:
+                results = await asyncio.gather(*[self._execute_external(it) for it in external_items])
+                for it, res in zip(external_items, results):
+                    self.total_requests += 1
+                    if not getattr(res, "success", False):
+                        self.total_errors += 1
+                    if not it.future.done():
+                        it.future.set_result(res)
+
+        if not sandbox_items:
+            return
+
+        # Acquire slots for the distinct trajectories in this batch.
+        try:
+            traj_ids = list({it.trajectory_id for it in sandbox_items})
+            slots = await asyncio.gather(*[self._get_or_acquire_slot(tid) for tid in traj_ids])
+            slot_by_traj = dict(zip(traj_ids, slots))
+        except Exception as e:
+            for it in sandbox_items:
+                self.total_requests += 1
+                self.total_errors += 1
+                if not it.future.done():
+                    it.future.set_result(
+                        ToolResult(
+                            success=False,
+                            error=f"Failed to acquire slot: {e}",
+                            uniq_id=it.call.uniq_id,
+                        )
+                    )
+            return
+
+        # Group by timeout so we don't accidentally make short timeouts wait on long ones.
+        by_timeout: Dict[float, List[_QueuedToolRequest]] = {}
+        default_timeout = self.backend.default_timeout_s
+
+        for it in sandbox_items:
+            t = it.timeout_s
+            if t is None:
+                t = default_timeout
+            if t is None:
+                t = 30.0
+            by_timeout.setdefault(float(t), []).append(it)
+
+        for timeout_s, items in by_timeout.items():
+            requests = []
+            dispatched: List[_QueuedToolRequest] = []
+            for it in items:
+                slot = slot_by_traj[it.trajectory_id]
+                tool_name = it.call.name
+                args = dict(it.call.arguments)
+
+                # Hermes compatibility: treat `terminal` as an alias of sandbox `bash`.
+                if tool_name == "terminal":
+                    if args.get("background"):
+                        self.total_requests += 1
+                        self.total_errors += 1
+                        if not it.future.done():
+                            it.future.set_result(
+                                ToolResult(
+                                    success=False,
+                                    error="terminal background execution is not supported in sandbox",
+                                    uniq_id=it.call.uniq_id,
+                                )
+                            )
+                        continue
+                    tool_name = "bash"
+                    # `timeout` is handled at the ToolExecutor level, not passed to the sandbox tool args.
+                    args.pop("timeout", None)
+                elif tool_name == "terminal_stateful":
+                    tool_name = "bash_stateful"
+                    args.pop("timeout", None)
+                elif tool_name == "tmux":
+                    # `tmux` is a sandbox tool backed by the stateful session manager.
+                    # Network policy is env-controlled.
+                    args.pop("allow_network", None)
+
+                if tool_name == "bash":
+                    # Network policy is set by the environment/executor, not by the model.
+                    args.pop("allow_network", None)
+                    args.pop("require_sandbox", None)
+                    args["allow_network"] = bool(self.config.allow_network)
+                    args["require_sandbox"] = bool(self.config.require_sandbox)
+                    # `timeout` is handled at the ToolExecutor level, not passed to the sandbox tool args.
+                    args.pop("timeout", None)
+                elif tool_name == "bash_stateful":
+                    # Network policy is set by the environment/executor, not by the model.
+                    args.pop("allow_network", None)
+                    args.pop("require_sandbox", None)
+                    args.pop("require_stateful_sandbox", None)
+                    args["allow_network"] = bool(self.config.allow_network)
+                    args["require_stateful_sandbox"] = bool(self.config.require_stateful_sandbox)
+                    args.pop("timeout", None)
+                elif tool_name == "tmux":
+                    # Network policy applies to the underlying stateful session.
+                    args.pop("allow_network", None)
+                    args.pop("require_sandbox", None)
+                    args.pop("require_stateful_sandbox", None)
+                    args["allow_network"] = bool(self.config.allow_network)
+                    args["require_stateful_sandbox"] = bool(self.config.require_stateful_sandbox)
+
+                requests.append((slot, tool_name, args))
+                dispatched.append(it)
+
+            results = None
+            try:
+                if not dispatched:
+                    continue
+                results = await self.backend.execute_batch(requests, timeout_s=timeout_s)
+            except Exception as e:
+                for it in items:
+                    self.total_requests += 1
+                    self.total_errors += 1
+                    if not it.future.done():
+                        it.future.set_result(
+                            ToolResult(
+                                success=False,
+                                error=f"Batch execution failed: {e}",
+                                uniq_id=it.call.uniq_id,
+                            )
+                        )
+                continue
+
+            for it, res in zip(dispatched, results):
+                self.total_requests += 1
+                if not getattr(res, "success", False):
+                    self.total_errors += 1
+                tool_result = res.to_tool_result()
+                tool_result.uniq_id = it.call.uniq_id
+                if not it.future.done():
+                    it.future.set_result(tool_result)

atropos/tools/toolset_resolver.py

@@ -0,0 +1,88 @@
+"""
+Toolset resolution for Hermes-Agent Atropos integration.
+
+We primarily reuse Hermes-Agent toolsets (`toolsets.py`), but Atropos training/envs
+need a few extra sandbox-oriented toolsets that Hermes doesn't expose by default
+(e.g. filesystem + stateful terminal).
+"""
+
+from __future__ import annotations
+
+from typing import Any, Dict, List, Optional, Set
+
+import toolsets as hermes_toolsets
+
+
+ATROPOS_TOOLSETS: Dict[str, Dict[str, Any]] = {
+    "filesystem": {
+        "description": "Read/write files in the sandbox workspace.",
+        "tools": ["read_file", "write_file"],
+        "includes": [],
+    },
+    "terminal_stateful": {
+        "description": "Stateful terminal execution (tmux/TUI support) inside the sandbox.",
+        "tools": ["terminal_stateful", "tmux"],
+        "includes": [],
+    },
+    "sandbox": {
+        "description": "Sandbox tools (terminal + filesystem).",
+        "tools": [],
+        "includes": ["terminal", "filesystem"],
+    },
+    "default": {
+        "description": "Default toolset for Atropos AgentEnv tasks.",
+        "tools": [],
+        "includes": ["sandbox"],
+    },
+    "full": {
+        "description": "All Hermes tools plus Atropos sandbox additions.",
+        "tools": [],
+        "includes": ["all", "filesystem", "sandbox", "terminal_stateful"],
+    },
+}
+
+
+def validate_toolset(name: str) -> bool:
+    if name in {"all", "*"}:
+        return True
+    return hermes_toolsets.validate_toolset(name) or name in ATROPOS_TOOLSETS
+
+
+def resolve_toolset(name: str, visited: Optional[Set[str]] = None) -> List[str]:
+    if visited is None:
+        visited = set()
+
+    if name in {"all", "*"}:
+        # Union Hermes + Atropos toolsets.
+        all_tools: Set[str] = set()
+        for tname in hermes_toolsets.get_toolset_names():
+            all_tools.update(resolve_toolset(tname, visited=set()))
+        for tname, spec in ATROPOS_TOOLSETS.items():
+            # Avoid recursion: some Atropos toolsets (e.g. "full") include "all".
+            if tname == "full" or "all" in (spec.get("includes") or []):
+                continue
+            all_tools.update(resolve_toolset(tname, visited=set()))
+        return sorted(all_tools)
+
+    if name in ATROPOS_TOOLSETS:
+        if name in visited:
+            return []
+        visited.add(name)
+        spec = ATROPOS_TOOLSETS[name]
+        tools: Set[str] = set(spec.get("tools", []))
+        for inc in spec.get("includes", []):
+            tools.update(resolve_toolset(inc, visited=set(visited)))
+        return sorted(tools)
+
+    # Fall back to Hermes toolsets.
+    # IMPORTANT: do not pre-add `name` to `visited` here; Hermes' resolver uses
+    # `visited` for its own cycle detection and will treat the presence of `name`
+    # as a circular dependency.
+    return sorted(hermes_toolsets.resolve_toolset(name, visited=set(visited)))
+
+
+def resolve_multiple_toolsets(names: List[str]) -> List[str]:
+    tools: Set[str] = set()
+    for name in names:
+        tools.update(resolve_toolset(name, visited=set()))
+    return sorted(tools)

atropos_compatible_agent.py

@@ -0,0 +1,415 @@
+#!/usr/bin/env python3
+"""
+Atropos-compatible Hermes agent runner.
+
+This is a minimal subclass of Hermes-Agent's `AIAgent` that swaps the OpenAI
+function-calling backend for Atroposlib's `ManagedServer`/`ServerManager` backend
+and uses Hermes-style XML tool tags:
+
+- <tool_call>{"name": "...", "arguments": {...}}</tool_call>
+- <tool_response>{...}</tool_response>
+
+Tool observations are appended as `role="user"` messages containing one or more
+`<tool_response>` blocks so they survive common chat templates during tokenization.
+"""
+
+from __future__ import annotations
+
+import asyncio
+import json
+import re
+import time
+import warnings
+import os
+from contextlib import asynccontextmanager
+from typing import Any, AsyncGenerator, Dict, List, Optional, Tuple
+
+from model_tools import cleanup_vm, handle_function_call
+from run_agent import AIAgent
+
+_TOOL_CALL_RE = re.compile(r"<tool_call>\\s*(.*?)\\s*</tool_call>", re.DOTALL)
+
+
+ATROPOS_TOOL_SYSTEM_PROMPT = """You are a helpful AI assistant with access to tools.
+
+## Available Tools
+<tools>
+{tool_descriptions}
+</tools>
+
+## How to Use Tools
+To call a tool, output:
+<tool_call>{{"name": "tool_name", "arguments": {{"arg1": "value1"}}}}</tool_call>
+
+You may include optional reasoning in <think>...</think> before tool calls.
+
+After each tool call, you will receive tool results as:
+<tool_response>{{...}}</tool_response>
+
+Continue until finished, then provide a final response with no <tool_call> blocks.
+"""
+
+
+class AtroposAIAgent(AIAgent):
+    """
+    Hermes `AIAgent` variant that uses Atroposlib ServerManager/ManagedServer.
+
+    Notes:
+    - The default Hermes `AIAgent` remains unchanged; this class is opt-in.
+    - The underlying server must expose `managed_server(tokenizer=...)` OR be a single
+      APIServer-compatible object usable by Atroposlib's `ManagedServer`.
+    """
+
+    def __init__(
+        self,
+        *,
+        server: Any,
+        tokenizer: Any = None,
+        model: str = "local",
+        max_iterations: int = 10,
+        tool_delay: float = 0.0,
+        enabled_toolsets: Optional[List[str]] = None,
+        disabled_toolsets: Optional[List[str]] = None,
+        save_trajectories: bool = False,
+        verbose_logging: bool = False,
+        quiet_mode: bool = False,
+        ephemeral_system_prompt: Optional[str] = None,
+        log_prefix_chars: int = 100,
+        log_prefix: str = "",
+        session_id: Optional[str] = None,
+        temperature: Optional[float] = None,
+        max_tokens: Optional[int] = None,
+    ):
+        # Call parent init mainly to reuse tool selection + trajectory saving utilities.
+        super().__init__(
+            base_url="http://unused",
+            api_key="dummy-key",
+            model=model,
+            max_iterations=max_iterations,
+            tool_delay=tool_delay,
+            enabled_toolsets=enabled_toolsets,
+            disabled_toolsets=disabled_toolsets,
+            save_trajectories=save_trajectories,
+            verbose_logging=verbose_logging,
+            quiet_mode=quiet_mode,
+            ephemeral_system_prompt=ephemeral_system_prompt,
+            log_prefix_chars=log_prefix_chars,
+            log_prefix=log_prefix,
+            session_id=session_id,
+        )
+
+        self.server = server
+        self.tokenizer = tokenizer
+        self.temperature = temperature
+        self.max_tokens = max_tokens
+
+    @asynccontextmanager
+    async def _managed(self) -> AsyncGenerator[Any, None]:
+        if hasattr(self.server, "managed_server"):
+            with warnings.catch_warnings():
+                warnings.filterwarnings(
+                    "ignore",
+                    message=r"Using OpenAIServer with managed_server does not allow for state tracking",
+                    category=UserWarning,
+                )
+                async with self.server.managed_server(tokenizer=self.tokenizer) as managed:
+                    yield managed
+            return
+
+        # Fall back to directly wrapping a single server object.
+        from atroposlib.envs.server_handling.managed_server import ManagedServer
+
+        managed = ManagedServer(server=self.server, tokenizer=self.tokenizer)
+        try:
+            yield managed
+        finally:
+            managed.reset()
+
+    def _tool_descriptions_text(self) -> str:
+        if not self.tools:
+            return "(no tools available)"
+
+        parts: List[str] = []
+        for tool in self.tools:
+            fn = (tool or {}).get("function", {})
+            name = fn.get("name", "")
+            desc = (fn.get("description") or "").strip()
+            if not name:
+                continue
+            if desc:
+                parts.append(f"- {name}: {desc}")
+            else:
+                parts.append(f"- {name}")
+        return "\n".join(parts) if parts else "(no tools available)"
+
+    def _build_system_prompt(self, system_message: Optional[str]) -> Optional[str]:
+        tool_prompt = ATROPOS_TOOL_SYSTEM_PROMPT.format(
+            tool_descriptions=self._tool_descriptions_text()
+        )
+
+        parts: List[str] = []
+        if system_message:
+            parts.append(system_message)
+        if self.ephemeral_system_prompt:
+            parts.append(self.ephemeral_system_prompt)
+        parts.append(tool_prompt)
+
+        return "\n\n".join(parts)
+
+    def _parse_tool_calls(self, content: str) -> Tuple[List[Tuple[str, Dict[str, Any]]], List[str]]:
+        """
+        Returns:
+          (calls, errors)
+        """
+        calls: List[Tuple[str, Dict[str, Any]]] = []
+        errors: List[str] = []
+
+        for raw in _TOOL_CALL_RE.findall(content or ""):
+            try:
+                payload = json.loads(raw)
+            except json.JSONDecodeError as exc:
+                errors.append(f"Invalid JSON inside <tool_call>: {exc}")
+                continue
+
+            name = payload.get("name")
+            args = payload.get("arguments", {})
+            if not isinstance(name, str) or not name:
+                errors.append("Tool call missing 'name' string")
+                continue
+            if not isinstance(args, dict):
+                errors.append("Tool call 'arguments' must be an object")
+                continue
+
+            calls.append((name, args))
+
+        return calls, errors
+
+    async def run_conversation_async(
+        self,
+        user_message: str,
+        system_message: Optional[str] = None,
+        conversation_history: Optional[List[Dict[str, Any]]] = None,
+        task_id: Optional[str] = None,
+    ) -> Dict[str, Any]:
+        import uuid
+
+        effective_task_id = task_id or str(uuid.uuid4())
+
+        messages: List[Dict[str, Any]] = conversation_history.copy() if conversation_history else []
+        messages.append({"role": "user", "content": user_message})
+
+        active_system_prompt = self._build_system_prompt(system_message)
+
+        api_call_count = 0
+        final_response: Optional[str] = None
+        managed_state: Optional[Dict[str, Any]] = None
+        completed = False
+
+        try:
+            async with self._managed() as managed:
+                while api_call_count < self.max_iterations:
+                    api_call_count += 1
+
+                    api_messages = messages.copy()
+                    if active_system_prompt:
+                        api_messages = [{"role": "system", "content": active_system_prompt}] + api_messages
+
+                    chat_kwargs: Dict[str, Any] = {"messages": api_messages, "n": 1}
+                    if self.max_tokens is not None:
+                        chat_kwargs["max_tokens"] = self.max_tokens
+                    if self.temperature is not None:
+                        chat_kwargs["temperature"] = self.temperature
+
+                    # Prefer OpenAI tool calling when supported by the backend:
+                    # - Many providers normalize Hermes-style <tool_call> tags into tool_calls when `tools` is provided.
+                    # - ManagedServer (atroposlib) does prompt->completion conversion and does not support `tools`.
+                    #   Only pass `tools` when we're calling an OpenAI-compatible chat endpoint directly.
+                    tool_schemas = self.tools if self.tools else None
+                    managed_cls = type(managed).__name__
+                    if tool_schemas and managed_cls != "ManagedServer":
+                        chat_kwargs["tools"] = tool_schemas
+
+                    if os.getenv("HERMES_DEBUG_ATROPOS_REQUEST") == "1":
+                        meta = {
+                            "managed_type": managed_cls,
+                            "model": getattr(getattr(managed, "config", None), "model_name", self.model),
+                            "base_url": getattr(getattr(managed, "config", None), "base_url", None),
+                            "kwargs": chat_kwargs,
+                        }
+                        # Avoid dumping megabytes of data accidentally.
+                        # (Messages can be large; this is still "full" but bounded.)
+                        print("\n=== HERMES_DEBUG_ATROPOS_REQUEST ===", flush=True)
+                        print(json.dumps(meta, ensure_ascii=False, indent=2)[:200_000], flush=True)
+
+                    response = await managed.chat_completion(**chat_kwargs)
+
+                    if os.getenv("HERMES_DEBUG_ATROPOS_RESPONSE") == "1":
+                        try:
+                            dumped = response.model_dump()  # openai pydantic model
+                        except Exception:
+                            dumped = getattr(response, "__dict__", {"repr": repr(response)})
+                        print("\n=== HERMES_DEBUG_ATROPOS_RESPONSE: ChatCompletion (raw) ===", flush=True)
+                        print(json.dumps(dumped, ensure_ascii=False, indent=2), flush=True)
+
+                    if hasattr(managed, "get_state"):
+                        managed_state = managed.get_state()
+
+                    msg = response.choices[0].message
+                    assistant_content = (msg.content or "")
+                    msg_reasoning = getattr(msg, "reasoning", None)
+
+                    # Use tool_calls if the backend provides them (preferred).
+                    structured_tool_calls = getattr(msg, "tool_calls", None)
+
+                    # If the backend emits content="" but includes useful text in reasoning,
+                    # use it for parsing *only if needed* (e.g. tool tags).
+                    if assistant_content == "" and isinstance(msg_reasoning, str) and msg_reasoning:
+                        if os.getenv("HERMES_DEBUG_ATROPOS_RESPONSE") == "1":
+                            print("\n=== HERMES_DEBUG_ATROPOS_RESPONSE: message.reasoning present (content empty) ===", flush=True)
+                            print(msg_reasoning, flush=True)
+
+                    assistant_msg: Dict[str, Any] = {"role": "assistant", "content": assistant_content}
+                    if structured_tool_calls:
+                        # Preserve tool_calls so the next request is consistent with OpenAI protocol.
+                        try:
+                            assistant_msg["tool_calls"] = [
+                                {
+                                    "id": tc.id,
+                                    "type": tc.type,
+                                    "function": {"name": tc.function.name, "arguments": tc.function.arguments},
+                                }
+                                for tc in structured_tool_calls
+                            ]
+                        except Exception:
+                            # Best-effort; keep conversation moving.
+                            pass
+                    messages.append(assistant_msg)
+
+                    # Mode A: OpenAI tool calling (preferred when supported)
+                    if structured_tool_calls:
+                        for tc in structured_tool_calls:
+                            tool_start = time.time()
+                            try:
+                                tool_args = json.loads(tc.function.arguments or "{}")
+                            except Exception:
+                                tool_args = {}
+                            tool_result = handle_function_call(tc.function.name, tool_args, effective_task_id)
+                            tool_duration = time.time() - tool_start
+
+                            # Keep the raw tool result as tool content (OpenAI protocol expects role=tool).
+                            messages.append(
+                                {
+                                    "role": "tool",
+                                    "tool_call_id": tc.id,
+                                    "content": tool_result,
+                                }
+                            )
+
+                            if self.tool_delay and self.tool_delay > 0:
+                                await asyncio.sleep(self.tool_delay)
+
+                        # Continue loop after tool execution.
+                        continue
+
+                    # Mode B: Hermes XML tool tags in assistant text (fallback).
+                    parse_source = assistant_content or (msg_reasoning or "")
+                    tool_calls, parse_errors = self._parse_tool_calls(parse_source)
+
+                    if parse_errors and not tool_calls:
+                        # Ask the model to retry with valid tool JSON.
+                        err_text = "; ".join(parse_errors[:3])
+                        messages.append(
+                            {
+                                "role": "user",
+                                "content": (
+                                    f"<tool_response>{json.dumps({'error': err_text}, ensure_ascii=False)}</tool_response>\n"
+                                    "The previous <tool_call> blocks were invalid. Please output valid JSON inside <tool_call>."
+                                ),
+                            }
+                        )
+                        continue
+
+                    if not tool_calls:
+                        # No tool calls: treat as final answer.
+                        final_response = (assistant_content or "").strip()
+                        completed = True
+                        break
+
+                    tool_responses: List[str] = []
+                    for tool_name, tool_args in tool_calls:
+                        tool_start = time.time()
+                        tool_result = handle_function_call(tool_name, tool_args, effective_task_id)
+                        tool_duration = time.time() - tool_start
+
+                        try:
+                            parsed = json.loads(tool_result)
+                            payload: Any = parsed
+                        except Exception:
+                            payload = tool_result
+
+                        tool_payload = {
+                            "name": tool_name,
+                            "duration_s": round(tool_duration, 3),
+                            "result": payload,
+                        }
+                        tool_responses.append(
+                            f"<tool_response>{json.dumps(tool_payload, ensure_ascii=False)}</tool_response>"
+                        )
+
+                        if self.tool_delay and self.tool_delay > 0:
+                            await asyncio.sleep(self.tool_delay)
+
+                    messages.append({"role": "user", "content": "\n".join(tool_responses)})
+
+                if final_response is None:
+                    final_response = "I've reached the maximum number of iterations."
+
+        finally:
+            try:
+                cleanup_vm(effective_task_id)
+            except Exception:
+                pass
+
+        # Save trajectory using Hermes formatting (optional).
+        self._save_trajectory(messages, user_message, completed=completed)
+
+        return {
+            "final_response": final_response,
+            "messages": messages,
+            "api_calls": api_call_count,
+            "completed": completed,
+            "managed_state": managed_state,
+            "system_prompt": active_system_prompt,
+            "task_id": effective_task_id,
+        }
+
+    def run_conversation(self, *args: Any, **kwargs: Any) -> Dict[str, Any]:
+        """
+        Sync wrapper for convenience.
+
+        If called from within a running event loop (e.g. prompt_toolkit), this
+        runs the async conversation in a dedicated thread to avoid nested loops.
+        """
+        try:
+            asyncio.get_running_loop()
+        except RuntimeError:
+            return asyncio.run(self.run_conversation_async(*args, **kwargs))
+
+        import queue
+        import threading
+
+        out: "queue.Queue[object]" = queue.Queue(maxsize=1)
+
+        def runner() -> None:
+            try:
+                out.put(asyncio.run(self.run_conversation_async(*args, **kwargs)))
+            except BaseException as exc:  # noqa: BLE001
+                out.put(exc)
+
+        thread = threading.Thread(target=runner, daemon=True)
+        thread.start()
+
+        result = out.get()
+        if isinstance(result, BaseException):
+            raise result
+        return result  # type: ignore[return-value]

batch_runner.py

@@ -27,29 +27,38 @@ import time
 from pathlib import Path
 from typing import List, Dict, Any, Optional, Tuple
 from datetime import datetime
-from multiprocessing import Pool, Lock
+from multiprocessing import Pool, Manager, 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())
+# All possible tools - used to ensure consistent schema across all trajectory entries
+# This is required because Arrow/Parquet (used by HuggingFace datasets) needs identical schemas
+ALL_POSSIBLE_TOOLS = {
+    'terminal', 'web_search', 'web_extract',
+    'vision_analyze', 'image_generate', 'mixture_of_agents',
+    # Skills tools
+    'skills_categories', 'skills_list', 'skill_view',
+    # Browser automation tools
+    'browser_navigate', 'browser_snapshot', 'browser_click',
+    'browser_type', 'browser_scroll', 'browser_back',
+    'browser_press', 'browser_close', 'browser_get_images',
+    'browser_vision'
+}
 
 # Default stats for tools that weren't used
 DEFAULT_TOOL_STATS = {'count': 0, 'success': 0, 'failure': 0}
@@ -171,7 +180,7 @@ 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 (json.JSONDecodeError, ValueError, TypeError):
+            except:
                 # 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:
@@ -191,42 +200,6 @@ 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],
@@ -238,7 +211,7 @@ def _process_single_prompt(
     
     Args:
         prompt_index (int): Index of prompt in dataset
-        prompt_data (Dict): Prompt data containing 'prompt' field and optional 'image' field
+        prompt_data (Dict): Prompt data containing 'prompt' field
         batch_num (int): Batch number
         config (Dict): Configuration dict with agent parameters
         
@@ -246,58 +219,6 @@ 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
@@ -323,22 +244,14 @@ def _process_single_prompt(
             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=task_id)
+        result = agent.run_conversation(prompt, task_id=f"task_{prompt_index}")
         
         # 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"],
@@ -351,7 +264,6 @@ 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"],
@@ -420,9 +332,7 @@ 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:
@@ -436,13 +346,6 @@ 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)
@@ -483,10 +386,6 @@ 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"]
         
-        # 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)
@@ -502,8 +401,6 @@ 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
     }
 
@@ -531,10 +428,6 @@ class BatchRunner:
         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.
@@ -556,10 +449,6 @@ class BatchRunner:
             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
@@ -577,10 +466,6 @@ class BatchRunner:
         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):
@@ -596,17 +481,13 @@ class BatchRunner:
         # Statistics file
         self.stats_file = self.output_dir / "statistics.json"
         
-        # Load dataset (and optionally truncate to max_samples)
+        # Load dataset
         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("📊 Batch Runner Initialized")
+        print(f"📊 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)}")
@@ -700,13 +581,14 @@ 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:
-                atomic_json_write(self.checkpoint_file, checkpoint_data)
+                with open(self.checkpoint_file, 'w', encoding='utf-8') as f:
+                    json.dump(checkpoint_data, f, indent=2, ensure_ascii=False)
         else:
-            atomic_json_write(self.checkpoint_file, checkpoint_data)
+            with open(self.checkpoint_file, 'w', encoding='utf-8') as f:
+                json.dump(checkpoint_data, f, indent=2, ensure_ascii=False)
     
     def _scan_completed_prompts_by_content(self) -> set:
         """
@@ -826,20 +708,18 @@ class BatchRunner:
             print("=" * 70)
             print(f"   Original dataset size:     {len(self.dataset):,} prompts")
             print(f"   Already completed:         {len(skipped_indices):,} prompts")
-            print("   ─────────────────────────────────────────")
+            print(f"   ─────────────────────────────────────────")
             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
-            }
+        # Initialize checkpoint data (needed for saving at the end)
+        checkpoint_data = {
+            "run_name": self.run_name,
+            "completed_prompts": [],
+            "batch_stats": {},
+            "last_updated": None
+        }
         
         # Prepare configuration for workers
         config = {
@@ -855,13 +735,10 @@ class BatchRunner:
             "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,
         }
         
         # For backward compatibility, still track by index (but this is secondary to content matching)
-        completed_prompts_set = set(checkpoint_data.get("completed_prompts", []))
+        completed_prompts_set = set()
         
         # Aggregate statistics across all batches
         total_tool_stats = {}
@@ -870,9 +747,6 @@ class BatchRunner:
         
         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
@@ -888,7 +762,7 @@ class BatchRunner:
             ]
             
             print(f"✅ Created {len(tasks)} batch tasks")
-            print("🚀 Starting parallel batch processing...\n")
+            print(f"🚀 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
@@ -918,35 +792,11 @@ class BatchRunner:
                     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", []))
@@ -963,17 +813,10 @@ 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 (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}")
+        # Save final checkpoint
+        checkpoint_data["completed_prompts"] = all_completed_prompts
+        self._save_checkpoint(checkpoint_data)
         
         # Calculate success rates
         for tool_name in total_tool_stats:
@@ -992,8 +835,15 @@ class BatchRunner:
         combined_file = self.output_dir / "trajectories.jsonl"
         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
+        VALID_TOOLS = {'web_search', 'web_extract', 'terminal', 'vision_analyze', 
+                       'image_generate', 'mixture_of_agents',
+                       # Skills tools
+                       'skills_categories', 'skills_list', 'skill_view',
+                       # Browser automation tools
+                       'browser_navigate', 'browser_snapshot', 'browser_click',
+                       'browser_type', 'browser_scroll', 'browser_back',
+                       'browser_press', 'browser_close', 'browser_get_images',
+                       'browser_vision'}
         
         total_entries = 0
         filtered_entries = 0
@@ -1042,8 +892,7 @@ class BatchRunner:
             "model": self.model,
             "completed_at": datetime.now().isoformat(),
             "duration_seconds": round(time.time() - start_time, 2),
-            "tool_statistics": total_tool_stats,
-            "reasoning_statistics": total_reasoning_stats,
+            "tool_statistics": total_tool_stats
         }
         
         with open(self.stats_file, 'w', encoding='utf-8') as f:
@@ -1057,7 +906,7 @@ class BatchRunner:
         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("\n📈 Tool Usage Statistics:")
+        print(f"\n📈 Tool Usage Statistics:")
         print("-" * 70)
         
         if total_tool_stats:
@@ -1081,28 +930,9 @@ 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("   - Trajectories: trajectories.jsonl (combined)")
-        print("   - Individual batches: batch_*.jsonl (for debugging)")
+        print(f"   - Trajectories: trajectories.jsonl (combined)")
+        print(f"   - Individual batches: batch_*.jsonl (for debugging)")
         print(f"   - Statistics: {self.stats_file.name}")
         print(f"   - Checkpoint: {self.checkpoint_file.name}")
 
@@ -1112,7 +942,7 @@ def main(
     batch_size: int = None,
     run_name: str = None,
     distribution: str = "default",
-    model: str = "anthropic/claude-sonnet-4.6",
+    model: str = "anthropic/claude-sonnet-4-20250514",
     api_key: str = None,
     base_url: str = "https://openrouter.ai/api/v1",
     max_turns: int = 10,
@@ -1126,11 +956,6 @@ def main(
     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.
@@ -1154,11 +979,6 @@ def main(
         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
@@ -1170,13 +990,9 @@ def main(
         # Use specific distribution
         python batch_runner.py --dataset_file=data.jsonl --batch_size=10 --run_name=image_test --distribution=image_gen
         
-        # With disabled reasoning and max tokens
+        # With ephemeral system prompt (not saved to dataset)
         python batch_runner.py --dataset_file=data.jsonl --batch_size=10 --run_name=my_run \\
-                               --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
+                               --ephemeral_system_prompt="You are a helpful assistant focused on image generation."
         
         # List available distributions
         python batch_runner.py --list_distributions
@@ -1215,36 +1031,6 @@ def main(
     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(
@@ -1264,10 +1050,6 @@ def main(
             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

@@ -7,74 +7,12 @@
 # =============================================================================
 model:
   # Default model to use (can be overridden with --model flag)
-  default: "anthropic/claude-opus-4.6"
-  
-  # Inference provider selection:
-  #   "auto"       - Use Nous Portal if logged in, otherwise OpenRouter/env vars (default)
-  #   "nous-api"   - Use Nous Portal via API key (requires: NOUS_API_KEY)
-  #   "openrouter" - Always use OpenRouter API key from OPENROUTER_API_KEY
-  #   "nous"       - Always use Nous Portal (requires: hermes login)
-  #   "zai"        - Use z.ai / ZhipuAI GLM models (requires: GLM_API_KEY)
-  #   "kimi-coding"- Use Kimi / Moonshot AI models (requires: KIMI_API_KEY)
-  #   "minimax"    - Use MiniMax global endpoint (requires: MINIMAX_API_KEY)
-  #   "minimax-cn" - Use MiniMax China endpoint (requires: MINIMAX_CN_API_KEY)
-  # Can also be overridden with --provider flag or HERMES_INFERENCE_PROVIDER env var.
-  provider: "auto"
+  default: "anthropic/claude-sonnet-4"
   
   # 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
 # =============================================================================
@@ -85,14 +23,10 @@ model:
 # 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.
+  env_type: "local"
+  cwd: "."  # Use "." for current directory, or specify absolute path
   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!
 
@@ -102,8 +36,8 @@ terminal:
 # 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
+#   env_type: "ssh"
+#   cwd: "/home/myuser/project"
 #   timeout: 180
 #   lifetime_seconds: 300
 #   ssh_host: "my-server.example.com"
@@ -117,18 +51,11 @@ terminal:
 # Great for: reproducible environments, testing, isolation
 # -----------------------------------------------------------------------------
 # terminal:
-#   backend: "docker"
-#   cwd: "/workspace"  # Path INSIDE the container (default: /)
+#   env_type: "docker"
+#   cwd: "/workspace"
 #   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"
+#   docker_image: "python:3.11"
 
 # -----------------------------------------------------------------------------
 # OPTION 4: Singularity/Apptainer container
@@ -136,11 +63,11 @@ terminal:
 # Great for: HPC clusters, shared compute environments
 # -----------------------------------------------------------------------------
 # terminal:
-#   backend: "singularity"
-#   cwd: "/workspace"  # Path INSIDE the container (default: /root)
+#   env_type: "singularity"
+#   cwd: "/workspace"
 #   timeout: 180
 #   lifetime_seconds: 300
-#   singularity_image: "docker://nikolaik/python-nodejs:python3.11-nodejs20"
+#   singularity_image: "docker://python:3.11"
 
 # -----------------------------------------------------------------------------
 # OPTION 5: Modal cloud execution
@@ -148,34 +75,11 @@ terminal:
 # Great for: GPU access, scalable compute, serverless execution
 # -----------------------------------------------------------------------------
 # terminal:
-#   backend: "modal"
-#   cwd: "/workspace"  # Path INSIDE the sandbox (default: /root)
+#   env_type: "modal"
+#   cwd: "/workspace"
 #   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)
+#   modal_image: "python:3.11"
 
 # -----------------------------------------------------------------------------
 # SUDO SUPPORT (works with ALL backends above)
@@ -200,20 +104,6 @@ terminal:
 # 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
 # =============================================================================
@@ -222,186 +112,19 @@ browser:
   # 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 4 turns (recent context is most relevant)
-# 5. Summarizes middle turns using a fast/cheap model
-# 6. Inserts summary as a user message, continues conversation seamlessly
-#
-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: 0.85 = 85%)
-  # Lower values = more aggressive compression, higher values = compress later
-  threshold: 0.85
-  
-  # 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"
-#
-#   # 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
-
 # =============================================================================
 # 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
+  # Maximum conversation turns before stopping
+  max_turns: 20
   
   # 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"
+  # Custom system prompt (personality, instructions, etc.)
+  # Leave empty or remove to use default agent behavior
+  system_prompt: ""
   
   # Predefined personalities (use with /personality command)
   personalities:
@@ -426,111 +149,19 @@ agent:
 # Control which tools the agent has access to.
 # Use "all" to enable everything, or specify individual toolsets.
 
-# =============================================================================
-# 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
+# Available toolsets:
 #
 #   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
+#   terminal     - Command execution (terminal)
 #   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)
+#   skills       - Load skill documents (skills_categories, 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)
+#   debugging    - terminal + web (for troubleshooting)
 #   safe         - web + vision + moa (no terminal access)
 
 # -----------------------------------------------------------------------------
@@ -581,74 +212,6 @@ toolsets:
 # toolsets:
 #   - safe
 
-# =============================================================================
-# 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
 # =============================================================================
@@ -664,131 +227,9 @@ stt:
 # 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
-
-  # 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.
-  # Disabled by default — enable to try the streaming UX.
-  streaming: false
-
-  # ───────────────────────────────────────────────────────────────────────────
-  # 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

configs/run_browser_tasks.sh

@@ -0,0 +1,42 @@
+#!/bin/bash
+
+# Browser-focused data generation run
+# Uses browser-use-tasks.jsonl (6504 tasks)
+# Distribution: browser 97%, web 20%, vision 12%, terminal 15%
+
+# Create logs directory if it doesn't exist
+mkdir -p logs
+
+# Generate log filename with timestamp
+LOG_FILE="logs/browser_tasks_$(date +%Y%m%d_%H%M%S).log"
+
+echo "📝 Logging output to: $LOG_FILE"
+echo "🌐 Running browser-focused tasks with browser_tasks distribution"
+
+python batch_runner.py \
+  --dataset_file="browser-use-tasks.jsonl" \
+  --batch_size=20 \
+  --run_name="browser_tasks" \
+  --distribution="browser_tasks" \
+  --model="moonshotai/kimi-k2.5" \
+  --verbose \
+  --base_url="https://openrouter.ai/api/v1" \
+  --num_workers=50 \
+  --max_turns=60 \
+  --resume \
+  --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 try to search directly on Google or other search engines via the browser - they block automated searches. Instead, ALWAYS use the web_search tool first to find URLs for any pages you need to visit, then use browser tools to navigate to those URLs.
+
+2. COOKIE/PRIVACY DIALOGS: After navigating to a page, ALWAYS check if there are cookie consent dialogs, privacy popups, or overlay modals blocking the page. These appear in snapshots as 'dialog' elements with buttons like 'Close', 'Accept', 'Accept All', 'Decline', 'I Agree', 'Got it', 'OK', or 'X'. You MUST dismiss these dialogs FIRST by clicking the appropriate button before trying to interact with other page elements. After dismissing a dialog, take a fresh browser_snapshot to get updated element references.
+
+3. HANDLING TIMEOUTS: If an action times out, it often means the element is blocked by an overlay or the page state has changed. Take a new snapshot to see the current page state and look for any dialogs or popups that need to be dismissed. If there is no dialog box to bypass, then try a new method or report the error to the user and complete the task.
+
+4. GENERAL: Use browser tools to click elements, fill forms, extract information, and perform web-based tasks. If terminal is available, use it for any local file operations or computations needed to support your web tasks. Be thorough in verifying your actions and handle any errors gracefully by retrying or trying alternative approaches." \
+  2>&1 | tee "$LOG_FILE"
+
+echo "✅ Log saved to: $LOG_FILE"
+
+#  --providers_allowed="gmicloud,siliconflow,atlas-cloud,z-ai,novita" \

configs/run_datagen_glm4.7-imagen.sh

@@ -0,0 +1,26 @@
+#!/bin/bash
+
+# Create logs directory if it doesn't exist
+mkdir -p logs
+
+# Generate a timestamp for the log file
+TIMESTAMP=$(date +%Y%m%d_%H%M%S)
+LOG_FILE="logs/imagen_eval_gpt5_${TIMESTAMP}.log"
+
+echo "📝 Logging output to: $LOG_FILE"
+
+python batch_runner.py \
+  --dataset_file="source-data/hermes-agent-imagen-data/hermes_agent_imagen_train_sft.jsonl" \
+  --batch_size=20 \
+  --run_name="imagen_train_sft_glm4.7" \
+  --distribution="image_gen" \
+  --model="z-ai/glm-4.7" \
+  --base_url="https://openrouter.ai/api/v1" \
+  --providers_allowed="gmicloud,siliconflow,atlas-cloud,z-ai,novita" \
+  --num_workers=50 \
+  --max_turns=25 \
+  --ephemeral_system_prompt="When generating an image for the user view the image by using the vision_analyze tool to ensure it is what the user wanted. If it isn't feel free to retry a few times. If none are perfect, choose the best option that is the closest match, and explain its imperfections. If the image generation tool fails, try again a few times. If the vision analyze tool fails, provide the image to the user and explain it is your best effort attempt." \
+  2>&1 | tee "$LOG_FILE"
+
+echo "✅ Log saved to: $LOG_FILE"
+#  --verbose \

configs/run_datagen_glm4.7.sh

@@ -0,0 +1,26 @@
+#!/bin/bash
+
+# Create logs directory if it doesn't exist
+mkdir -p logs
+
+# Generate log filename with timestamp
+LOG_FILE="logs/glm4.7-thinking-sft1_$(date +%Y%m%d_%H%M%S).log"
+
+echo "📝 Logging output to: $LOG_FILE"
+
+python batch_runner.py \
+  --dataset_file="source-data/hermes-agent-agent-tasks-1/agent_tasks_sft_2.jsonl" \
+  --batch_size=20 \
+  --run_name="megascience_glm4.7-thinking-sft2" \
+  --distribution="science" \
+  --model="z-ai/glm-4.7" \
+  --base_url="https://openrouter.ai/api/v1" \
+  --providers_allowed="gmicloud,siliconflow,atlas-cloud,z-ai,novita" \
+  --num_workers=15 \
+  --max_turns=60 \
+  --ephemeral_system_prompt="You have access to a variety of tools to help you solve scientific, math, and technology problems presented to you. You can use them in sequence and build off of the results of prior tools you've used results. Always use the terminal or search tool if it can provide additional context, verify formulas, double check concepts and recent studies and understanding, doing all calculations, etc. You should only be confident in your own reasoning, knowledge, or calculations if you've exhaustively used all tools available to you to that can help you verify or validate your work. Always pip install any packages you need to use the python scripts you want to run. If you need to use a tool that isn't available, you can use the terminal tool to install or create it in many cases as well. Do not use the terminal tool to communicate with the user, as they cannot see your commands, only your final response after completing the task. Search for at least 3 sources, but not more than 12, so you can maintain focused context." \
+  2>&1 | tee "$LOG_FILE"
+
+echo "✅ Log saved to: $LOG_FILE"
+
+#  --verbose \

configs/run_datagen_glm4.7_megascience.sh

@@ -0,0 +1,27 @@
+#!/bin/bash
+
+# Create logs directory if it doesn't exist
+mkdir -p logs
+
+# Generate log filename with timestamp
+LOG_FILE="logs/glm4.7-thinking-sft1-10k_$(date +%Y%m%d_%H%M%S).log"
+
+echo "📝 Logging output to: $LOG_FILE"
+
+python batch_runner.py \
+  --dataset_file="source-data/hermes-agent-megascience-data/hermes_agent_megascience_sft_train_1_10k.jsonl" \
+  --batch_size=20 \
+  --run_name="megascience_glm4.7-thinking-sft1" \
+  --distribution="science" \
+  --model="z-ai/glm-4.7" \
+  --base_url="https://openrouter.ai/api/v1" \
+  --providers_allowed="gmicloud,siliconflow,atlas-cloud,z-ai,novita" \
+  --num_workers=50 \
+  --max_turns=60 \
+  --resume \
+  --ephemeral_system_prompt="You have access to a variety of tools to help you solve scientific, math, and technology problems presented to you. You can use them in sequence and build off of the results of prior tools you've used for furthering results. Always use the terminal or search tool if it can provide additional context, verify formulas, double check concepts and recent studies and understanding, doing all calculations, etc. You should only be confident in your own reasoning, knowledge, or calculations if you've exhaustively used all tools available to you to that can help you verify or validate your work. Always pip install any packages you need to use the python scripts you want to run. If you need to use a tool that isn't available, you can use the terminal tool to install or create it in many cases as well. Do not use the terminal tool to communicate with the user, as they cannot see your commands, only your final response after completing the task. Search for at least 3 sources, but not more than 12, so you can maintain a focused context." \
+  2>&1 | tee "$LOG_FILE"
+
+echo "✅ Log saved to: $LOG_FILE"
+
+#  --verbose \

configs/run_datagen_glm4.7_raw_tasks.sh

@@ -0,0 +1,28 @@
+#!/bin/bash
+
+# Create logs directory if it doesn't exist
+mkdir -p logs
+
+# Generate log filename with timestamp
+LOG_FILE="logs/glm4.7-terminal-tasks_$(date +%Y%m%d_%H%M%S).log"
+
+echo "📝 Logging output to: $LOG_FILE"
+
+python batch_runner.py \
+  --dataset_file="source-data/raw_tasks_prompts.jsonl" \
+  --batch_size=20 \
+  --run_name="terminal-tasks-glm4.7-thinking" \
+  --distribution="default" \
+  --model="z-ai/glm-4.7" \
+  --base_url="https://openrouter.ai/api/v1" \
+  --providers_allowed="gmicloud,siliconflow,atlas-cloud,z-ai,novita" \
+  --num_workers=50 \
+  --max_turns=60 \
+  --ephemeral_system_prompt="You have access to a variety of tools to help you complete coding, system administration, and general computing tasks. You can use them in sequence and build off of the results of prior tools you've used. Always use the terminal tool to execute commands, write code, install packages, and verify your work. You should test and validate everything you create. Always pip install any packages you need (use --break-system-packages if needed). If you need a tool that isn't available, you can use the terminal to install or create it. Do not use the terminal tool to communicate with the user, as they cannot see your commands, only your final response after completing the task. Use web search when you need to look up documentation, APIs, or current best practices." \
+  2>&1 | tee "$LOG_FILE"
+
+echo "✅ Log saved to: $LOG_FILE"
+
+#  --verbose \
+#  --resume \
+

configs/run_datagen_megascience.sh

@@ -0,0 +1,12 @@
+python batch_runner.py \
+  --dataset_file="hermes-agent-megascience-data/hermes_agent_megascience_eval.jsonl" \
+  --batch_size=10 \
+  --run_name="megascience_eval_gpt5_2" \
+  --distribution="science" \
+  --model="gpt-5" \
+  --base_url="https://api.openai.com/v1" \
+  --api_key="${OPENAI_API_KEY}" \
+  --num_workers=5 \
+  --max_turns=30 \
+  --verbose \
+  --ephemeral_system_prompt="You have access to a variety of tools to help you solve scientific, math, and technology problems presented to you. You can use them in sequence and build off of the results of prior tools you've used results. Always use a tool if it can provide additional context, verify formulas, double check concepts and recent studies and understanding, doing all calculations, etc. You should not be confident in your own reasoning, knowledge, or calculations without using a tool to verify or validate your work."

configs/run_datagen_minimax-3.1.sh

@@ -0,0 +1,12 @@
+python batch_runner.py \
+  --dataset_file="source-data/hermes-agent-agent-tasks-1/agent_tasks_eval.jsonl" \
+  --batch_size=50 \
+  --run_name="megascience_sft_minimax-m2.1-thinking-2-eval" \
+  --distribution="science" \
+  --model="minimax/minimax-m2.1" \
+  --base_url="https://openrouter.ai/api/v1" \
+  --providers_allowed="minimax" \
+  --num_workers=1 \
+  --max_turns=40 \
+  --verbose \
+  --ephemeral_system_prompt="You have access to a variety of tools to help you solve scientific, math, and technology problems presented to you. You can use them in sequence and build off of the results of prior tools you've used results. Always use the terminal or search tool if it can provide additional context, verify formulas, double check concepts and recent studies and understanding, doing all calculations, etc. You should only be confident in your own reasoning, knowledge, or calculations if you've exhaustively used all tools available to you to that can help you verify or validate your work. Always pip install any packages you need to use the python scripts you want to run. If you need to use a tool that isn't available, you can use the terminal tool to install or create it in many cases as well. Do not use the terminal tool to communicate with the user, as they cannot see your commands, only your final response after completing the task. Search for at least 3 sources, but not more than 12."

configs/run_eval_glm4.7_newterm.sh

@@ -0,0 +1,29 @@
+#!/bin/bash
+
+# Create logs directory if it doesn't exist
+mkdir -p logs
+
+# Generate log filename with timestamp
+LOG_FILE="logs/glm4.7-terminal-tasks-newterm_$(date +%Y%m%d_%H%M%S).log"
+
+echo "📝 Logging output to: $LOG_FILE"
+
+python batch_runner.py \
+  --dataset_file="source-data/hermes-agent-agent-tasks-1/agent_tasks_eval.jsonl" \
+  --batch_size=1 \
+  --run_name="terminal-tasks-test-newterm" \
+  --distribution="terminal_only" \
+  --verbose \
+  --model="z-ai/glm-4.7" \
+  --base_url="https://openrouter.ai/api/v1" \
+  --providers_allowed="gmicloud,siliconflow,atlas-cloud,z-ai,novita" \
+  --num_workers=5 \
+  --max_turns=60 \
+  --ephemeral_system_prompt="You have access to a variety of tools to help you complete coding, system administration, and general computing tasks. You can use them in sequence and build off of the results of prior tools you've used. Always use the terminal tool to execute commands, write code, install packages, and verify your work. You should test and validate everything you create. Always pip install any packages you need (use --break-system-packages if needed). If you need a tool that isn't available, you can use the terminal to install or create it. Do not use the terminal tool to communicate with the user, as they cannot see your commands, only your final response after completing the task. Use web search when you need to look up documentation, APIs, or current best practices." \
+  2>&1 | tee "$LOG_FILE"
+
+echo "✅ Log saved to: $LOG_FILE"
+
+#  --verbose \
+#  --resume \
+

configs/run_eval_terminal.sh

@@ -0,0 +1,33 @@
+#!/bin/bash
+
+# Terminal-only evaluation run using Modal sandboxes
+# Uses 10 sample tasks from nous-terminal-tasks
+
+# Create logs directory if it doesn't exist
+mkdir -p logs
+
+# Generate log filename with timestamp
+LOG_FILE="logs/terminal_eval_$(date +%Y%m%d_%H%M%S).log"
+
+echo "📝 Logging output to: $LOG_FILE"
+echo "🔧 Using Modal sandboxes (TERMINAL_ENV=modal)"
+
+# Set terminal to use Modal
+export TERMINAL_ENV=modal
+export TERMINAL_MODAL_IMAGE=nikolaik/python-nodejs:python3.11-nodejs20
+export TERMINAL_TIMEOUT=300
+
+python batch_runner.py \
+  --dataset_file="nous-terminal-tasks_eval.jsonl" \
+  --batch_size=5 \
+  --run_name="terminal_eval" \
+  --distribution="terminal_only" \
+  --model="z-ai/glm-4.7" \
+  --base_url="https://openrouter.ai/api/v1" \
+  --providers_allowed="gmicloud,siliconflow,atlas-cloud,z-ai,novita" \
+  --num_workers=2 \
+  --max_turns=30 \
+  --ephemeral_system_prompt="You have access to a terminal tool for executing commands. Use it to complete the task. Install any packages you need with apt-get or pip (use --break-system-packages if needed). Do not use interactive tools (vim, nano, python repl). If git output is large, pipe to cat." \
+  2>&1 | tee "$LOG_FILE"
+
+echo "✅ Log saved to: $LOG_FILE"