fix(a2a): client tools take args-as-dict positional; accept agent_name alias

Live Tier-3 testing (CLI agent -> a2a tools -> live peer gateway -> model)
surfaced two bugs the kwarg-style unit tests masked:

1. registry.dispatch calls handlers as handler(args, **kwargs) — args is the
   whole dict positional. The handlers used keyword params (url=, agent=), so
   the dict bound to the first param and .strip() raised
   'dict object has no attribute strip'. Rewrote all three handlers to take
   args: dict (matching the spotify/google_meet convention). Added a
   registry-dispatch regression test that exercises the real call path the
   direct-kwarg tests never hit.

2. The model repeatedly reached for agent_name= instead of agent= (6 retries
   before success). Accept agent_name/name and message/text/task aliases so a
   reasonable guess succeeds first try.

Verified live: client agent discovers the peer's Agent Card, calls it, and
gets the reply back (PONG round-trip confirmed on both client audit log and
peer conversation log). 39 plugin tests pass.

.dockerignore

@@ -63,45 +63,3 @@ data/
 # Compose/profile runtime state (bind-mounted; avoid ownership/secret issues)
 hermes-config/
 runtime/
-
-# ---------- Not needed inside the Docker image ----------
-
-# Desktop app source (Tauri/Electron); never installed in the container
-apps/
-
-# Test suite — not shipped in production images
-tests/
-
-# Documentation site (Docusaurus) and supplementary docs
-website/
-docs/
-
-# Assets only used by the GitHub README
-assets/
-infographic/
-
-# Plugin-level docs (hermes-achievements ships docs/ but the runtime doesn't read them)
-plugins/hermes-achievements/docs/
-
-# Nix / Homebrew / AUR packaging metadata — irrelevant to Docker
-nix/
-flake.nix
-flake.lock
-packaging/
-
-# Design and planning documents
-plans/
-.plans/
-
-# ACP registry manifest (icon + agent.json) — not consumed at runtime
-acp_registry/
-
-# Repo-level dotfiles that are git-only or dev-tooling config
-.env.example
-.envrc
-.gitattributes
-.hadolint.yaml
-.mailmap
-
-# Top-level LICENSE (not matched by *.md); not needed inside the container
-LICENSE

.github/workflows/deploy-site.yml

@@ -44,7 +44,7 @@ jobs:
 
       - uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020  # v4
         with:
-          node-version: 22
+          node-version: 20
           cache: npm
           cache-dependency-path: website/package-lock.json
 
@@ -59,22 +59,12 @@ jobs:
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
         run: |
-          # Rebuild the unified catalog. The file is gitignored, so a fresh
-          # checkout starts without it and we want the freshest crawl in
-          # every deploy.
-          #
-          # This MUST be fatal. build_skills_index.py runs a health check and
-          # exits non-zero WITHOUT writing the output file when a source
-          # collapses (e.g. a GitHub API rate limit zeroes the github /
-          # claude-marketplace / well-known taps all at once). Letting the
-          # deploy continue would either (a) ship a degenerate index missing
-          # whole hubs — the June 2026 regression where OpenAI/Anthropic/
-          # HuggingFace/NVIDIA tabs vanished — or (b) fall through to a
-          # local-only catalog. Failing here keeps the last good deployment
-          # live (GitHub Pages serves the previous build) instead of
-          # publishing a broken catalog. Re-run the workflow once the
-          # transient rate limit clears.
-          python3 scripts/build_skills_index.py
+          # Always rebuild — the file isn't committed (gitignored), so a
+          # fresh checkout starts without it and we want the freshest crawl
+          # in every deploy. Failure is non-fatal: extract-skills.py will
+          # fall back to the legacy snapshot cache and the Skills Hub page
+          # still renders, just without the latest community catalog.
+          python3 scripts/build_skills_index.py || echo "Skills index build failed (non-fatal)"
 
       - name: Extract skill metadata for dashboard
         run: python3 website/scripts/extract-skills.py

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

@@ -18,7 +18,7 @@ jobs:
 
       - uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020  # v4
         with:
-          node-version: 22
+          node-version: 20
           cache: npm
           cache-dependency-path: website/package-lock.json
 

.github/workflows/nix-lockfile-fix.yml

@@ -75,10 +75,9 @@ jobs:
         run: |
           set -euo pipefail
 
-          # Ensure only nix/lib.nix (home of the single npmDepsHash) was
-          # modified — prevents accidental self-triggering if fix-lockfiles
-          # ever touches package files.
-          unexpected="$(git diff --name-only | grep -Ev '^nix/lib\.nix$' || true)"
+          # Ensure only nix files were modified — prevents accidental
+          # self-triggering if fix-lockfiles ever touches package files.
+          unexpected="$(git diff --name-only | grep -Ev '^nix/(tui|web)\.nix$' || true)"
           if [ -n "$unexpected" ]; then
             echo "::error::Unexpected modified files: $unexpected"
             exit 1
@@ -90,7 +89,7 @@ jobs:
 
           git config user.name 'github-actions[bot]'
           git config user.email '41898282+github-actions[bot]@users.noreply.github.com'
-          git add nix/lib.nix
+          git add nix/tui.nix nix/web.nix
           git commit -m "fix(nix): auto-refresh npm lockfile hashes" \
             -m "Source: $GITHUB_SHA" \
             -m "Run: $GITHUB_SERVER_URL/$GITHUB_REPOSITORY/actions/runs/$GITHUB_RUN_ID"
@@ -217,7 +216,7 @@ jobs:
           set -euo pipefail
           git config user.name 'github-actions[bot]'
           git config user.email '41898282+github-actions[bot]@users.noreply.github.com'
-          git add nix/lib.nix
+          git add nix/tui.nix nix/web.nix
           git commit -m "fix(nix): refresh npm lockfile hashes"
           git push
 

.github/workflows/tests.yml

@@ -55,31 +55,15 @@ jobs:
 
       - name: Install uv
         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
-          # pyproject.toml or uv.lock changes. `uv sync` still runs every
-          # time, but resolves from the warm cache instead of re-downloading
-          # and re-building wheels.
-          enable-cache: true
-          cache-dependency-glob: |
-            pyproject.toml
-            uv.lock
 
       - name: Set up Python 3.11
         run: uv python install 3.11
 
       - name: Install dependencies
-        # `uv sync --locked` installs the exact pinned set from uv.lock (and
-        # fails if the lock is out of sync with pyproject.toml), giving a
-        # reproducible env. It also creates .venv itself, so no separate
-        # `uv venv` step is needed.
-        run: uv sync --locked --python 3.11 --extra all --extra dev
-
-      - name: Minimize uv cache
-        # Optimized for CI: prunes pre-built wheels that are cheap to
-        # re-download, keeping the persisted cache small and fast to restore.
-        run: uv cache prune --ci
+        run: |
+          uv venv .venv --python 3.11
+          source .venv/bin/activate
+          uv pip install -e ".[all,dev]"
 
       - name: Run tests (slice ${{ matrix.slice }}/6)
         # Per-file isolation via scripts/run_tests_parallel.py: discovers
@@ -177,31 +161,15 @@ jobs:
 
       - name: Install uv
         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
-          # pyproject.toml or uv.lock changes. `uv sync` still runs every
-          # time, but resolves from the warm cache instead of re-downloading
-          # and re-building wheels.
-          enable-cache: true
-          cache-dependency-glob: |
-            pyproject.toml
-            uv.lock
 
       - name: Set up Python 3.11
         run: uv python install 3.11
 
       - name: Install dependencies
-        # `uv sync --locked` installs the exact pinned set from uv.lock (and
-        # fails if the lock is out of sync with pyproject.toml), giving a
-        # reproducible env. It also creates .venv itself, so no separate
-        # `uv venv` step is needed.
-        run: uv sync --locked --python 3.11 --extra all --extra dev
-
-      - name: Minimize uv cache
-        # Optimized for CI: prunes pre-built wheels that are cheap to
-        # re-download, keeping the persisted cache small and fast to restore.
-        run: uv cache prune --ci
+        run: |
+          uv venv .venv --python 3.11
+          source .venv/bin/activate
+          uv pip install -e ".[all,dev]"
 
       - name: Packaged-wheel i18n smoke test
         run: |

.github/workflows/typecheck.yml

@@ -1,25 +0,0 @@
-# .github/workflows/typecheck.yml
-name: Typecheck
-
-on:
-  push:
-    branches: [main]
-  pull_request:
-    branches: [main]
-
-jobs:
-  typecheck:
-    runs-on: ubuntu-latest
-    strategy:
-      matrix:
-        package:
-          [ui-tui, web, apps/bootstrap-installer, apps/desktop, apps/shared]
-      fail-fast: false # report all failures, not just the first one
-    steps:
-      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
-      - uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020 # v4
-        with:
-          node-version: 22
-          cache: npm
-      - run: npm ci
-      - run: npm run --prefix ${{ matrix.package }} typecheck

.gitignore

@@ -114,12 +114,6 @@ docs/superpowers/*
 # treat it as a local edit and autostash it on every run (#38529).
 .hermes-bootstrap-complete
 
-# Interrupted-update breadcrumb + recovery lock written next to the shared venv
-# by `hermes update` / launch-time self-heal. Runtime state, never a code change
-# — ignore so `git status` stays clean and update's autostash skips them.
-.update-incomplete
-.update-incomplete.lock
-
 # Tool Search live-test harness output — non-deterministic model transcripts,
 # regenerated by scripts/tool_search_livetest.py. Never an artifact of the repo.
 scripts/out/

AGENTS.md

@@ -4,201 +4,6 @@ Instructions for AI coding assistants and developers working on the hermes-agent
 
 **Never give up on the right solution.**
 
-## What Hermes Is
-
-Hermes is a personal AI agent that runs the same agent core across a CLI, a
-messaging gateway (Telegram, Discord, Slack, and ~20 other platforms), a TUI,
-and an Electron desktop app. It learns across sessions (memory + skills),
-delegates to subagents, runs scheduled jobs, and drives a real terminal and
-browser. It is extended primarily through **plugins and skills**, not by
-growing the core.
-
-Two properties shape almost every design decision and are the lens for
-reviewing any change:
-
-- **Per-conversation prompt caching is sacred.** A long-lived conversation
-  reuses a cached prefix every turn. Anything that mutates past context,
-  swaps toolsets, or rebuilds the system prompt mid-conversation invalidates
-  that cache and multiplies the user's cost. We do not do it (the one
-  exception is context compression).
-- **The core is a narrow waist; capability lives at the edges.** Every model
-  tool we add is sent on every API call, so the bar for a new *core* tool is
-  high. Most new capability should arrive as a CLI command + skill, a
-  service-gated tool, or a plugin — not as core surface.
-
-## Contribution Rubric — What We Want / What We Don't
-
-This is the project's intent layer. Use it two ways:
-
-1. **For humans and for your own work** — what gets merged and what gets
-   rejected, so a contribution aims at the target.
-2. **For automated review (the triage sweeper)** — guidance on when a PR is
-   safe to close on the three allowed reasons (`implemented_on_main`,
-   `cannot_reproduce`, `incoherent`) and, just as important, **when NOT to
-   close** one. Taste-based "we don't want this / out of scope" closes are NOT
-   an automated decision — those stay with a human maintainer. The sweeper's
-   job here is to recognize design intent and *avoid wrongly closing a
-   legitimate contribution*, not to make the won't-implement call itself.
-
-Read the balance right: Hermes ships a **lot** — most merges are bug fixes to
-real reported behavior, and the product surface (platforms, channels,
-providers, models, desktop/TUI features) expands aggressively and on purpose.
-The restraint below is aimed squarely at the **core agent + the model tool
-schema**, the one place where every addition is paid for on every API call.
-"Smallest footprint" governs *how a capability is wired into the core*, NOT
-whether the product is allowed to grow. We are expansive at the edges and
-conservative at the waist.
-
-### What we want
-
-- **Fix real bugs, well.** The bulk of what lands is `fix(...)` against an
-  actual reported symptom. A good fix reproduces the symptom on current
-  `main`, points to the exact line where it manifests, and fixes the whole bug
-  class — sibling call paths included — not just the one site the reporter hit.
-- **Expand reach at the edges.** New platform adapters, channels, providers,
-  models, and desktop/TUI/dashboard features are welcome and land routinely,
-  including large ones (a new messaging channel, a session-cap feature, a
-  Windows PTY bridge). Breadth in the product is a goal, not a footprint
-  concern — as long as it integrates with the existing setup/config UX
-  (`hermes tools`, `hermes setup`, auto-install) rather than bolting on a raw
-  env var.
-- **Refactor god-files into clean modules.** Extracting a multi-thousand-line
-  cluster out of `cli.py` / `run_agent.py` / `gateway/run.py` into a focused
-  mixin or module is wanted work, even when the diff is huge and mechanical
-  (large `+N/-N` refactors merge regularly). The "every line traces to the
-  request" test applies to *feature* PRs; a declared refactor's request IS the
-  extraction.
-- **Keep the core narrow.** New *model tools* are the expensive exception —
-  every tool ships on every API call. Prefer, in order: extend existing code →
-  CLI command + skill → service-gated tool (`check_fn`) → plugin → MCP server
-  in the catalog → new core tool (last resort). See "The Footprint Ladder."
-- **Extend, don't duplicate.** Before adding a module/manager/hook, check
-  whether existing infrastructure already covers the use case. When several PRs
-  integrate the same *category*, design one shared interface instead of merging
-  them one at a time (see the ABC + orchestrator note under the Footprint
-  Ladder).
-- **Behavior contracts over snapshots.** Tests should assert how two pieces of
-  data must relate (invariants), not freeze a current value (model lists,
-  config version literals, enumeration counts). See "Don't write
-  change-detector tests."
-- **E2E validation, not just green unit mocks.** For anything touching
-  resolution chains, config propagation, security boundaries, remote
-  backends, or file/network I/O, exercise the real path with real imports
-  against a temp `HERMES_HOME`. Mocks hide integration bugs.
-- **Cache-, alternation-, and invariant-safe.** Preserve prompt caching, strict
-  message role alternation (never two same-role messages in a row; never a
-  synthetic user message injected mid-loop), and a system prompt that is
-  byte-stable for the life of a conversation.
-- **Contributor credit preserved.** Salvage external work by cherry-picking
-  (rebase-merge) so authorship survives in git history; don't reimplement from
-  scratch when you can build on top.
-
-### What we don't want (rejected even when well-built)
-
-- **Speculative infrastructure.** Hooks, callbacks, or extension points with no
-  concrete consumer. Adding a hook is easy; removing one after plugins depend
-  on it is hard. A hook is NOT speculative if a contributor has a real, stated
-  use case — even if the consumer ships separately.
-- **New `HERMES_*` env vars for non-secret config.** `.env` is for secrets
-  only (API keys, tokens, passwords). All behavioral settings — timeouts,
-  thresholds, feature flags, display prefs — go in `config.yaml`. Bridge to an
-  internal env var if the mechanism needs one, but user-facing docs point to
-  `config.yaml`. Reject PRs that tell users to "set X in your .env" unless X
-  is a credential.
-- **A new core tool when terminal + file already do the job, or when a skill
-  would.** If the only barrier is file visibility on a remote backend, fix the
-  mount, not the toolset.
-- **Lazy-reading escape hatches on instructional tools.** No `offset`/`limit`
-  pagination on tools that load content the agent must read fully (skills,
-  prompts, playbooks). Models will read page 1 and skip the rest.
-- **"Fixes" that destroy the feature they secure.** A mitigation that kills the
-  feature's purpose is the wrong mitigation. Read the original commit's intent
-  (`git log -p -S`) before restricting behavior; find a fix that preserves the
-  feature.
-- **Outbound telemetry / usage attribution without opt-in gating.** No new
-  analytics, third-party identifier tagging, or attribution tags until a
-  generic user-facing opt-in (config gate + setup prompt + `hermes tools`
-  toggle) exists. Park behind a label, do not merge.
-- **Change-detector tests, cache-breaking mid-conversation, dead code wired in
-  without E2E proof, and plugins that touch core files.** Plugins live in their
-  own directory and work within the ABCs/hooks we provide; if a plugin needs
-  more, widen the generic plugin surface, don't special-case it in core.
-
-### Before you call it a bug — verify the premise (and when NOT to close)
-
-The most common reason a well-written PR gets closed is not code quality — it
-is that the change is built on a **wrong premise**, or it treats an
-**intentional design as a gap**. These patterns cut both ways: they tell a
-human reviewer what to scrutinize, and they tell the automated sweeper when a
-PR is NOT safe to close as `implemented_on_main` / `cannot_reproduce` (when in
-doubt, leave it open for a human). They are distilled from real closes.
-
-- **"Intentional design, not a gap."** A limitation that looks like an
-  oversight is often deliberate. Before "fixing" a missing link or a
-  restriction, ask whether the isolation IS the design. Example: profiles are
-  independent islands on purpose — a PR adding live config inheritance from the
-  default profile was closed because coupling profiles together is exactly what
-  the design prevents (the copy-at-creation `--clone` path already covers the
-  legitimate "start from my default" case). Read the original commit's intent
-  (`git log -p -S "<symbol>"`) before assuming something is unfinished.
-- **"The premise doesn't hold against how X actually works."** A PR's
-  justification frequently rests on a wrong mental model of an existing
-  mechanism. Trace the real code/runtime before accepting the rationale. Two
-  real closes: a rate-limit "re-probe during cooldown" PR (the breaker only
-  trips on a *confirmed-empty* account bucket, so re-probing just hammers a
-  bucket we've already proven empty); a usage-accumulation fix whose new branch
-  **never executes at runtime** because an earlier guard already popped the
-  state it depended on. If you can't point to the exact line where the bug
-  manifests AND show the fix changes that line's behavior, you haven't verified
-  the premise.
-- **"This fix was wrong — the absence/omission was deliberate."** Adding the
-  obvious-looking missing piece can break things the omission was protecting.
-  Example: restoring "missing" `__init__.py` files made a test tree importable
-  as a dotted package that shadowed the real plugin, deleting its `register()`
-  at import time. The absence was load-bearing.
-- **"Overreached / resurrected an approach we'd moved past."** Scope creep that
-  supersedes an agreed-on base, or revives a direction the maintainers
-  deliberately closed, gets rejected even when the code works. Keep the change
-  to the narrow piece that was actually agreed; offer the rest as a focused
-  follow-up.
-
-The throughline: **verify the claim AND the intent against the codebase before
-writing or merging a fix.** A confirmed reproduction on current `main` plus a
-line-level account of where the fix acts beats a plausible-sounding rationale
-every time. When in doubt about intent, it is cheaper to ask than to ship a
-fix that fights the design.
-
-### The Footprint Ladder (new capability decision)
-
-Each rung adds more permanent surface than the one above. Choose the highest
-(least-footprint) rung that correctly solves the problem:
-
-1. **Extend existing code** — the capability is a variation of something that
-   already exists. Zero new surface.
-2. **CLI command + skill** — manages config/state/infra expressible as shell
-   commands. The agent runs `hermes <subcommand>` guided by a skill. Zero
-   model-tool footprint. Default choice for subscriptions, scheduled tasks,
-   service setup. Examples: `hermes webhook`, `hermes cron`, `hermes tools`.
-3. **Service-gated tool (`check_fn`)** — needs structured params/returns AND
-   only appears when a prerequisite is configured. Zero footprint otherwise.
-   Examples: Home Assistant tools (gated on token), memory-provider tools.
-4. **Plugin** — third-party/niche/user-specific capability that doesn't ship in
-   core. Lives in `~/.hermes/plugins/` or a pip package, discovered at runtime.
-5. **MCP server (in the catalog)** — if the capability genuinely needs to be a
-   tool (structured I/O the agent invokes) but isn't core-fundamental, prefer
-   building it as an MCP server and adding it to the MCP catalog over growing
-   the core toolset. The agent connects to it through the built-in MCP client;
-   zero permanent core-schema footprint, and it's reusable by any MCP host.
-6. **New core tool** — only when the capability is fundamental, broadly useful
-   to nearly every user, and unreachable via terminal + file (or an MCP server).
-   Examples of correct core tools: terminal, read_file, web_search,
-   browser_navigate.
-
-When 3+ open PRs try to integrate the same *category* of thing (memory
-backends, providers, notifiers), don't merge them one at a time — design an
-ABC + orchestrator, wrap the existing built-in as the first provider, and turn
-the competing PRs into plugins against that interface.
-
 ## Development Environment
 
 ```bash
@@ -459,7 +264,7 @@ npm install       # first time
 npm run dev       # watch mode (rebuilds hermes-ink + tsx --watch)
 npm start         # production
 npm run build     # full build (hermes-ink + tsc)
-npm run typecheck # typecheck only (tsc --noEmit)
+npm run type-check # typecheck only (tsc --noEmit)
 npm run lint      # eslint
 npm run fmt       # prettier
 npm test          # vitest
@@ -497,11 +302,9 @@ A **separate** chat surface from both the classic CLI and the dashboard's embedd
 
 ## Adding New Tools
 
-Before adding any tool, settle the footprint question first (see "The
-Footprint Ladder" in the Contribution Rubric): most capabilities should NOT
-be core tools. For custom or local-only tools, do **not** edit Hermes core.
-Use the plugin route instead: create `~/.hermes/plugins/<name>/plugin.yaml`
-and `~/.hermes/plugins/<name>/__init__.py`, then register tools with
+For most custom or local-only tools, do **not** edit Hermes core. Use the plugin
+route instead: create `~/.hermes/plugins/<name>/plugin.yaml` and
+`~/.hermes/plugins/<name>/__init__.py`, then register tools with
 `ctx.register_tool(...)`. Plugin toolsets are discovered automatically and can be
 enabled or disabled without touching `tools/` or `toolsets.py`.
 

Dockerfile

@@ -25,7 +25,7 @@ ENV PLAYWRIGHT_BROWSERS_PATH=/opt/hermes/.playwright
 # hermes process, the dashboard, and per-profile gateways.
 RUN apt-get update && \
     apt-get install -y --no-install-recommends \
-    ca-certificates curl iputils-ping python3 python-is-python3 ripgrep ffmpeg gcc g++ make cmake python3-dev python3-venv libffi-dev libolm-dev procps git openssh-client docker-cli xz-utils && \
+    ca-certificates curl iputils-ping python3 python-is-python3 ripgrep ffmpeg gcc python3-dev python3-venv libffi-dev libolm-dev procps git openssh-client docker-cli xz-utils && \
     rm -rf /var/lib/apt/lists/*
 
 # ---------- s6-overlay install ----------
@@ -146,9 +146,9 @@ RUN npm install --prefer-offline --no-audit && \
 #
 # `uv sync --frozen --no-install-project --extra all --extra messaging`
 # installs the deps reachable through the composite `[all]` extra
-# (handpicked set intended for the production image — excludes `[dev]`),
-# plus gateway messaging adapters that should work in the published image
-# without a first-boot lazy install.  We do NOT use `--all-extras`:
+# (handpicked set intended for the production image), plus gateway
+# messaging adapters that should work in the published image without a
+# first-boot lazy install.  We do NOT use `--all-extras`:
 # that would pull in `[rl]` (atroposlib + tinker + torch + wandb from
 # git), `[yc-bench]` (another git dep), and `[termux-all]` (Android
 # redundancy), none of which belong in the published container.
@@ -164,30 +164,19 @@ RUN npm install --prefer-offline --no-audit && \
 # image update and recall/retain then fails with
 # `ModuleNotFoundError: No module named 'hindsight_client'` (#38128).
 #
-# The Matrix gateway's deps ([matrix] extra) are baked in because
-# python-olm (transitive via mautrix[encryption]) builds from source on
-# Python/image combinations without usable wheels.  The Docker image is
-# Linux-only, so keeping the native libolm/build-toolchain packages here
-# avoids the cross-platform failures that kept [matrix] out of [all]
-# while still making Matrix work in the published container. Fixes #30399.
-#
 # The editable link is created after the source copy below.
 COPY pyproject.toml uv.lock ./
 RUN touch ./README.md
-RUN uv sync --frozen --no-install-project --extra all --extra messaging --extra anthropic --extra bedrock --extra azure-identity --extra hindsight --extra matrix
-
-# ---------- Frontend build (cached independently from Python source) ----------
-# Copy only the frontend source trees first so that Python-only changes don't
-# invalidate the (relatively slow) web + ui-tui build layer.
-COPY web/ web/
-COPY ui-tui/ ui-tui/
-RUN cd web && npm run build && \
-    cd ../ui-tui && npm run build
+RUN uv sync --frozen --no-install-project --extra all --extra messaging --extra anthropic --extra bedrock --extra azure-identity --extra hindsight
 
 # ---------- Source code ----------
 # .dockerignore excludes node_modules, so the installs above survive.
 COPY --chown=hermes:hermes . .
 
+# Build browser dashboard and terminal UI assets.
+RUN cd web && npm run build && \
+    cd ../ui-tui && npm run build
+
 # ---------- Permissions ----------
 # Make install dir world-readable so any HERMES_UID can read it at runtime.
 # The venv needs to be traversable too.

MANIFEST.in

@@ -1,6 +1,5 @@
 graft skills
 graft optional-skills
-graft optional-mcps
 graft locales
 # Bundled plugin manifests (plugin.yaml / plugin.yml). Without these the
 # PluginManager scan (hermes_cli/plugins.py) finds zero plugins on installs

README.md

@@ -3,9 +3,7 @@
 </p>
 
 # Hermes Agent ☤
-<p align="center">
-  <a href="https://hermes-agent.nousresearch.com/">Hermes Agent</a> | <a href="https://hermes-agent.nousresearch.com/">Hermes Desktop</a>
-</p>
+
 <p align="center">
   <a href="https://hermes-agent.nousresearch.com/docs/"><img src="https://img.shields.io/badge/Docs-hermes--agent.nousresearch.com-FFD700?style=for-the-badge" alt="Documentation"></a>
   <a href="https://discord.gg/NousResearch"><img src="https://img.shields.io/badge/Discord-5865F2?style=for-the-badge&logo=discord&logoColor=white" alt="Discord"></a>
@@ -55,7 +53,7 @@ If you already have Git installed, the installer detects it and uses that instea
 
 > **Android / Termux:** The tested manual path is documented in the [Termux guide](https://hermes-agent.nousresearch.com/docs/getting-started/termux). On Termux, Hermes installs a curated `.[termux]` extra because the full `.[all]` extra currently pulls Android-incompatible voice dependencies.
 >
-> **Windows:** Native Windows is fully supported — the PowerShell one-liner above installs everything. If you'd rather use WSL2, the Linux command works there too. Native Windows install lives under `%LOCALAPPDATA%\hermes`; WSL2 installs under `~/.hermes` as on Linux.
+> **Windows:** Native Windows is fully supported — the PowerShell one-liner above installs everything. If you'd rather use WSL2, the Linux command works there too. Native Windows install lives under `%LOCALAPPDATA%\hermes`; WSL2 installs under `~/.hermes` as on Linux. The only Hermes feature that currently needs WSL2 specifically is the browser-based dashboard chat pane (it uses a POSIX PTY — classic CLI and gateway both run natively).
 
 After installation:
 

acp_adapter/provenance.py

@@ -1,127 +0,0 @@
-"""Derive ACP session-provenance metadata from the existing compression chain.
-
-This is an additive Hermes extension surfaced under ACP ``_meta.hermes`` so
-existing ACP clients ignore it. It carries no new persisted state: everything
-is derived on demand from the ``sessions`` table (``parent_session_id`` /
-``end_reason``), which already models compression-continuation chains.
-
-The ACP/editor ``session_id`` stays the stable public handle. When context
-compression rotates the internal Hermes head, ``build_session_provenance`` lets
-a client see the previous/current internal ids and the lineage root without
-parsing status text, guessing from token drops, or reading ``state.db``.
-"""
-
-from __future__ import annotations
-
-from typing import Any, Dict, Optional
-
-# Bound defensive walks; compression chains this deep are pathological.
-_MAX_WALK = 100
-
-
-def build_session_provenance(
-    db: Any,
-    acp_session_id: str,
-    current_hermes_session_id: str,
-    *,
-    previous_hermes_session_id: Optional[str] = None,
-) -> Optional[Dict[str, Any]]:
-    """Build ``_meta.hermes.sessionProvenance`` for an ACP session.
-
-    Args:
-        db: A ``SessionDB`` (must expose ``get_session``).
-        acp_session_id: The stable ACP/editor-facing session handle.
-        current_hermes_session_id: The live internal Hermes DB session id
-            (``state.agent.session_id``).
-        previous_hermes_session_id: The internal id from before the most recent
-            turn, when known. Supplied by ``prompt()`` to flag a rotation.
-
-    Returns:
-        A dict suitable for ``{"hermes": {"sessionProvenance": <dict>}}`` under
-        ACP ``_meta``, or ``None`` if the session can't be read.
-    """
-    try:
-        row = db.get_session(current_hermes_session_id)
-    except Exception:
-        return None
-    if not row:
-        return None
-
-    parent_id = row.get("parent_session_id")
-    end_reason = row.get("end_reason")
-
-    # Walk parents to the lineage root and count compression depth. Only
-    # compression-split parents (parent.end_reason == 'compression') count
-    # toward depth — delegate/branch children share the parent_session_id
-    # column but are not compaction boundaries.
-    root_id = current_hermes_session_id
-    compression_depth = 0
-    cursor_parent = parent_id
-    seen = {current_hermes_session_id}
-    for _ in range(_MAX_WALK):
-        if not cursor_parent or cursor_parent in seen:
-            break
-        seen.add(cursor_parent)
-        try:
-            prow = db.get_session(cursor_parent)
-        except Exception:
-            prow = None
-        if not prow:
-            break
-        root_id = cursor_parent
-        if prow.get("end_reason") == "compression":
-            compression_depth += 1
-        cursor_parent = prow.get("parent_session_id")
-
-    # A session is a compression continuation when its parent was ended with
-    # end_reason='compression'. Determine that from the immediate parent.
-    is_continuation = False
-    if parent_id:
-        try:
-            immediate_parent = db.get_session(parent_id)
-        except Exception:
-            immediate_parent = None
-        if immediate_parent and immediate_parent.get("end_reason") == "compression":
-            is_continuation = True
-
-    rotated = bool(
-        previous_hermes_session_id
-        and previous_hermes_session_id != current_hermes_session_id
-    )
-
-    provenance: Dict[str, Any] = {
-        "acpSessionId": acp_session_id,
-        "currentHermesSessionId": current_hermes_session_id,
-        "rootHermesSessionId": root_id,
-        "parentHermesSessionId": parent_id,
-        "sessionKind": "continuation" if is_continuation else "root",
-        "compressionDepth": compression_depth,
-    }
-    if previous_hermes_session_id:
-        provenance["previousHermesSessionId"] = previous_hermes_session_id
-    if rotated:
-        # The head moved during the last turn. The only mechanism that rotates
-        # the internal id mid-turn is compression-driven session splitting.
-        provenance["reason"] = "compression"
-        provenance["creatorKind"] = "compression"
-
-    return provenance
-
-
-def session_provenance_meta(
-    db: Any,
-    acp_session_id: str,
-    current_hermes_session_id: str,
-    *,
-    previous_hermes_session_id: Optional[str] = None,
-) -> Optional[Dict[str, Any]]:
-    """Return a ready ``_meta`` payload: ``{"hermes": {"sessionProvenance": ...}}``."""
-    prov = build_session_provenance(
-        db,
-        acp_session_id,
-        current_hermes_session_id,
-        previous_hermes_session_id=previous_hermes_session_id,
-    )
-    if prov is None:
-        return None
-    return {"hermes": {"sessionProvenance": prov}}

acp_adapter/server.py

@@ -71,7 +71,6 @@ from acp_adapter.events import (
     make_tool_progress_cb,
 )
 from acp_adapter.permissions import make_approval_callback
-from acp_adapter.provenance import session_provenance_meta
 from acp_adapter.session import SessionManager, SessionState, _expand_acp_enabled_toolsets
 from acp_adapter.tools import build_tool_complete, build_tool_start
 
@@ -710,39 +709,8 @@ class HermesACPAgent(acp.Agent):
                 exc_info=True,
             )
 
-    def _provenance_meta(
-        self,
-        acp_session_id: str,
-        current_hermes_session_id: str,
-        previous_hermes_session_id: Optional[str] = None,
-    ) -> Optional[dict]:
-        """Best-effort ``_meta.hermes.sessionProvenance`` for an ACP session."""
-        try:
-            return session_provenance_meta(
-                self.session_manager._get_db(),
-                acp_session_id,
-                current_hermes_session_id,
-                previous_hermes_session_id=previous_hermes_session_id,
-            )
-        except Exception:
-            logger.debug(
-                "Could not build ACP session provenance for %s", acp_session_id, exc_info=True
-            )
-            return None
-
-    async def _send_session_info_update(
-        self,
-        session_id: str,
-        *,
-        current_hermes_session_id: Optional[str] = None,
-        previous_hermes_session_id: Optional[str] = None,
-    ) -> None:
-        """Send ACP native session metadata after Hermes changes it.
-
-        When the internal Hermes head rotated (e.g. compression-driven session
-        split during a turn), pass ``previous_hermes_session_id`` so the
-        attached ``_meta.hermes.sessionProvenance`` flags the rotation reason.
-        """
+    async def _send_session_info_update(self, session_id: str) -> None:
+        """Send ACP native session metadata after Hermes changes it."""
         if not self._conn:
             return
         try:
@@ -759,16 +727,10 @@ class HermesACPAgent(acp.Agent):
         # the updated_at since we're emitting this notification precisely
         # because the title was just refreshed.
         updated_at = datetime.now(timezone.utc).isoformat()
-        meta = self._provenance_meta(
-            session_id,
-            current_hermes_session_id or session_id,
-            previous_hermes_session_id,
-        )
         update = SessionInfoUpdate(
             session_update="session_info_update",
             title=title if isinstance(title, str) and title.strip() else None,
             updated_at=updated_at,
-            field_meta=meta,
         )
         try:
             await self._conn.session_update(
@@ -1119,9 +1081,6 @@ class HermesACPAgent(acp.Agent):
             session_id=state.session_id,
             models=self._build_model_state(state),
             modes=self._session_modes(state),
-            field_meta=self._provenance_meta(
-                state.session_id, getattr(state.agent, "session_id", state.session_id)
-            ),
         )
 
     async def load_session(
@@ -1166,9 +1125,6 @@ class HermesACPAgent(acp.Agent):
         return LoadSessionResponse(
             models=self._build_model_state(state),
             modes=self._session_modes(state),
-            field_meta=self._provenance_meta(
-                session_id, getattr(state.agent, "session_id", session_id)
-            ),
         )
 
     async def resume_session(
@@ -1201,9 +1157,6 @@ class HermesACPAgent(acp.Agent):
         return ResumeSessionResponse(
             models=self._build_model_state(state),
             modes=self._session_modes(state),
-            field_meta=self._provenance_meta(
-                state.session_id, getattr(state.agent, "session_id", state.session_id)
-            ),
         )
 
     async def cancel(self, session_id: str, **kwargs: Any) -> None:
@@ -1541,11 +1494,6 @@ class HermesACPAgent(acp.Agent):
                         logger.debug("Could not clear ACP session context", exc_info=True)
 
         try:
-            # Snapshot the internal Hermes DB session id before the turn so we
-            # can detect a compression-driven session rotation afterwards. The
-            # ACP `session_id` stays the stable client handle; agent.session_id
-            # is the live internal head that compression may rotate.
-            pre_turn_hermes_id = getattr(state.agent, "session_id", None)
             # Wrap the executor call in a fresh copy of the current context so
             # concurrent ACP sessions on the shared ThreadPoolExecutor don't
             # stomp on each other's ContextVar writes (HERMES_SESSION_KEY in
@@ -1564,41 +1512,8 @@ class HermesACPAgent(acp.Agent):
             # Persist updated history so sessions survive process restarts.
             self.session_manager.save_session(session_id)
 
-        # Detect a compression-driven internal session rotation. If the agent's
-        # DB head moved during the turn, emit a session_info_update carrying
-        # _meta.hermes.sessionProvenance so ACP clients can render the boundary
-        # and keep old/new ids in lineage. The ACP session_id is unchanged.
-        post_turn_hermes_id = getattr(state.agent, "session_id", None)
-        if (
-            conn
-            and post_turn_hermes_id
-            and pre_turn_hermes_id
-            and post_turn_hermes_id != pre_turn_hermes_id
-        ):
-            try:
-                await self._send_session_info_update(
-                    session_id,
-                    current_hermes_session_id=post_turn_hermes_id,
-                    previous_hermes_session_id=pre_turn_hermes_id,
-                )
-            except Exception:
-                logger.debug(
-                    "Could not emit ACP provenance update after rotation for %s",
-                    session_id,
-                    exc_info=True,
-                )
-
         final_response = result.get("final_response", "")
-        cancelled = bool(state.cancel_event and state.cancel_event.is_set())
-        interrupted = bool(result.get("interrupted")) or cancelled
-        # Hermes' local "waiting for model response" interrupt status is metadata,
-        # not assistant prose — clients get cancellation from stop_reason instead.
-        from agent.conversation_loop import INTERRUPT_WAITING_FOR_MODEL_PREFIX
-
-        suppress_interrupt_response = interrupted and final_response.startswith(
-            INTERRUPT_WAITING_FOR_MODEL_PREFIX
-        )
-        if final_response and not suppress_interrupt_response:
+        if final_response:
             try:
                 from agent.title_generator import maybe_auto_title
 
@@ -1619,12 +1534,7 @@ class HermesACPAgent(acp.Agent):
                 )
             except Exception:
                 logger.debug("Failed to auto-title ACP session %s", session_id, exc_info=True)
-        if (
-            final_response
-            and conn
-            and not suppress_interrupt_response
-            and (not streamed_message or result.get("response_transformed"))
-        ):
+        if final_response and conn and (not streamed_message or result.get("response_transformed")):
             # Deliver the final response when streaming did not already send it,
             # or when a plugin hook transformed the response after streaming
             # finished (e.g. transform_llm_output) — otherwise the appended /
@@ -1666,7 +1576,7 @@ class HermesACPAgent(acp.Agent):
 
         await self._send_usage_update(state)
 
-        stop_reason = "cancelled" if cancelled else "end_turn"
+        stop_reason = "cancelled" if state.cancel_event and state.cancel_event.is_set() else "end_turn"
         return PromptResponse(stop_reason=stop_reason, usage=usage)
 
     # ---- Slash commands (headless) -------------------------------------------

agent/agent_init.py

@@ -169,7 +169,6 @@ def init_agent(
     save_trajectories: bool = False,
     verbose_logging: bool = False,
     quiet_mode: bool = False,
-    tool_progress_mode: str = "all",
     ephemeral_system_prompt: str = None,
     log_prefix_chars: int = 100,
     log_prefix: str = "",
@@ -187,7 +186,6 @@ def init_agent(
     thinking_callback: callable = None,
     reasoning_callback: callable = None,
     clarify_callback: callable = None,
-    read_terminal_callback: callable = None,
     step_callback: callable = None,
     stream_delta_callback: callable = None,
     interim_assistant_callback: callable = None,
@@ -282,7 +280,6 @@ def init_agent(
     agent.save_trajectories = save_trajectories
     agent.verbose_logging = verbose_logging
     agent.quiet_mode = quiet_mode
-    agent.tool_progress_mode = tool_progress_mode
     agent.ephemeral_system_prompt = ephemeral_system_prompt
     agent.platform = platform  # "cli", "telegram", "discord", "whatsapp", etc.
     agent._user_id = user_id  # Platform user identifier (gateway sessions)
@@ -418,7 +415,6 @@ def init_agent(
     agent.thinking_callback = thinking_callback
     agent.reasoning_callback = reasoning_callback
     agent.clarify_callback = clarify_callback
-    agent.read_terminal_callback = read_terminal_callback
     agent.step_callback = step_callback
     agent.stream_delta_callback = stream_delta_callback
     agent.interim_assistant_callback = interim_assistant_callback

agent/agent_runtime_helpers.py

@@ -49,7 +49,7 @@ def _ra():
 
 
 AGENT_RUNTIME_POST_HOOK_TOOL_NAMES = frozenset(
-    {"todo", "session_search", "memory", "clarify", "read_terminal", "delegate_task"}
+    {"todo", "session_search", "memory", "clarify", "delegate_task"}
 )
 
 
@@ -1784,17 +1784,6 @@ def invoke_tool(agent, function_name: str, function_args: dict, effective_task_i
                 ),
                 next_args,
             )
-    elif function_name == "read_terminal":
-        def _execute(next_args: dict) -> Any:
-            from tools.read_terminal_tool import read_terminal_tool as _read_terminal_tool
-            return _finish_agent_tool(
-                _read_terminal_tool(
-                    start_line=next_args.get("start_line"),
-                    count=next_args.get("count"),
-                    callback=getattr(agent, "read_terminal_callback", None),
-                ),
-                next_args,
-            )
     elif function_name == "delegate_task":
         def _execute(next_args: dict) -> Any:
             return _finish_agent_tool(agent._dispatch_delegate_task(next_args), next_args)
@@ -1857,27 +1846,6 @@ def repair_tool_call(agent, tool_name: str) -> str | None:
     if not tool_name:
         return None
 
-    # VolcEngine api/plan workaround (issue #33007): the endpoint's
-    # protocol-translation layer occasionally leaks raw XML attribute
-    # fragments into tool_use.name, e.g.
-    #   `terminal" parameter="command" string="true`
-    #   `execute_code" parameter="code" string="true`
-    #   `session_search" parameter="session_id" string="true`
-    # We trim at the first unambiguous XML/quote character so the rest
-    # of the repair pipeline (lowercase / snake_case / fuzzy match)
-    # can resolve the cleaned name to a real tool.
-    #
-    # Crucially we DO NOT split on whitespace: legitimate inputs like
-    # "write file" must keep flowing through ``_norm`` -> ``write_file``
-    # (covered by test_space_to_underscore in
-    # tests/run_agent/test_repair_tool_call_name.py).
-    for _xml_sep in ('"', "'", "<", ">"):
-        _idx = tool_name.find(_xml_sep)
-        if _idx > 0:
-            tool_name = tool_name[:_idx]
-    if not tool_name:
-        return None
-
     def _norm(s: str) -> str:
         return s.lower().replace("-", "_").replace(" ", "_")
 

agent/anthropic_adapter.py

@@ -73,50 +73,20 @@ ADAPTIVE_EFFORT_MAP = {
     "minimal": "low",
 }
 
-# ── Anthropic thinking-mode classification ────────────────────────────
-# Claude 4.6 replaced budget-based extended thinking with *adaptive* thinking,
-# and 4.7 additionally forbids the manual ``thinking`` block entirely and drops
-# temperature/top_p/top_k.  Newer Claude releases (4.8, and named models like
-# claude-fable-5) follow the same modern contract — but they share no common
-# version substring, so an allowlist of version numbers ("4.6", "4.7", …) goes
-# stale the moment a model ships without a recognized number and silently
-# routes it down the legacy manual-thinking path.
-#
-# Instead we DEFAULT unknown Claude models to the modern contract and keep an
-# explicit *legacy* list of the older Claude families that still require manual
-# thinking.  This mirrors _get_anthropic_max_output's "default to newest" design
-# (future models are unlikely to regress to the older contract), so each new
-# Claude release works without a code change.
-#
-# Non-Claude Anthropic-Messages models (minimax, qwen3, GLM, …) are NOT Claude,
-# so they fall through to the legacy path automatically — exactly what those
-# manual-thinking endpoints need.
-
-# Older Claude families that DON'T support adaptive thinking (manual thinking
-# with budget_tokens only). Substring-matched against the model name.
-_LEGACY_MANUAL_THINKING_CLAUDE_SUBSTRINGS = (
-    "claude-3",          # 3, 3.5, 3.7
-    "claude-opus-4-0", "claude-opus-4.0", "claude-opus-4-1", "claude-opus-4.1",
-    "claude-sonnet-4-0", "claude-sonnet-4.0",
-    "claude-opus-4-2025", "claude-sonnet-4-2025",  # date-stamped 4.0 IDs
-    "claude-opus-4-5", "claude-opus-4.5",
-    "claude-sonnet-4-5", "claude-sonnet-4.5",
-    "claude-haiku-4-5", "claude-haiku-4.5",
-)
-
-# Older Claude families that DON'T accept the "xhigh" effort level (4.6 only
-# supports low/medium/high/max). xhigh arrived with Opus 4.7. Adaptive models
-# not in this list (4.7, 4.8, fable, future) accept xhigh.
-_NO_XHIGH_CLAUDE_SUBSTRINGS = (
-    "claude-opus-4-6", "claude-opus-4.6",
-    "claude-sonnet-4-6", "claude-sonnet-4.6",
-)
-
-
-def _is_claude_model(model: str | None) -> bool:
-    return "claude" in (model or "").lower()
+# Models that accept the "xhigh" output_config.effort level.  Opus 4.7 added
+# xhigh as a distinct level between high and max; older adaptive-thinking
+# models (4.6) reject it with a 400.  Keep this substring list in sync with
+# the Anthropic migration guide as new model families ship.
+_XHIGH_EFFORT_SUBSTRINGS = ("4-7", "4.7", "4-8", "4.8")
 
+# Models where extended thinking is deprecated/removed (4.6+ behavior: adaptive
+# is the only supported mode; 4.7 additionally forbids manual thinking entirely
+# and drops temperature/top_p/top_k).
+_ADAPTIVE_THINKING_SUBSTRINGS = ("4-6", "4.6", "4-7", "4.7", "4-8", "4.8")
 
+# Models where temperature/top_p/top_k return 400 if set to non-default values.
+# This is the Opus 4.7 contract; future 4.x+ models are expected to follow it.
+_NO_SAMPLING_PARAMS_SUBSTRINGS = ("4-7", "4.7", "4-8", "4.8")
 _FAST_MODE_SUPPORTED_SUBSTRINGS = ("opus-4-6", "opus-4.6")
 
 # ── Max output token limits per Anthropic model ───────────────────────
@@ -124,8 +94,6 @@ _FAST_MODE_SUPPORTED_SUBSTRINGS = ("opus-4-6", "opus-4.6")
 # max_tokens as a mandatory field.  Previously we hardcoded 16384, which
 # starves thinking-enabled models (thinking tokens count toward the limit).
 _ANTHROPIC_OUTPUT_LIMITS = {
-    # Mythos-class named models (claude-fable-5, …) — 1M context, reasoning
-    "claude-fable":      128_000,
     # Claude 4.8
     "claude-opus-4-8":   128_000,
     # Claude 4.7
@@ -240,17 +208,8 @@ def _resolve_anthropic_messages_max_tokens(
 
 
 def _supports_adaptive_thinking(model: str) -> bool:
-    """Return True for Claude models that use adaptive thinking (4.6+).
-
-    Defaults *unknown* Claude models to adaptive (the modern contract) and
-    only returns False for the explicit legacy list of older Claude families
-    that require manual budget-based thinking. Non-Claude Anthropic-Messages
-    models (minimax, qwen3, …) return False so they keep the manual path.
-    """
-    if not _is_claude_model(model):
-        return False
-    m = model.lower()
-    return not any(v in m for v in _LEGACY_MANUAL_THINKING_CLAUDE_SUBSTRINGS)
+    """Return True for Claude 4.6+ models that support adaptive thinking."""
+    return any(v in model for v in _ADAPTIVE_THINKING_SUBSTRINGS)
 
 
 def _supports_xhigh_effort(model: str) -> bool:
@@ -260,33 +219,18 @@ def _supports_xhigh_effort(model: str) -> bool:
     Pre-4.7 adaptive models (Opus/Sonnet 4.6) only accept low/medium/high/max
     and reject xhigh with an HTTP 400. Callers should downgrade xhigh→max
     when this returns False.
-
-    Defaults unknown adaptive Claude models to accepting xhigh (4.7+ contract);
-    only the 4.6 family and legacy manual-thinking models are excluded.
     """
-    if not _supports_adaptive_thinking(model):
-        return False
-    m = model.lower()
-    return not any(v in m for v in _NO_XHIGH_CLAUDE_SUBSTRINGS)
+    return any(v in model for v in _XHIGH_EFFORT_SUBSTRINGS)
 
 
 def _forbids_sampling_params(model: str) -> bool:
     """Return True for models that 400 on any non-default temperature/top_p/top_k.
 
-    Opus 4.7 introduced this restriction; later Claude releases follow it.
-    Defaults unknown Claude models to forbidding sampling params (the modern
-    contract). The 4.6 family still accepts them, and the legacy manual-thinking
-    families (4.5 and older) accept them too, so both are excluded. Non-Claude
-    models are unaffected. Callers should omit these fields entirely rather than
-    passing zero/default values (the API rejects anything non-null).
+    Opus 4.7 explicitly rejects sampling parameters; later Claude releases are
+    expected to follow suit.  Callers should omit these fields entirely rather
+    than passing zero/default values (the API rejects anything non-null).
     """
-    if not _is_claude_model(model):
-        return False
-    m = model.lower()
-    # 4.6 family is adaptive but still accepts sampling params.
-    if any(v in m for v in _NO_XHIGH_CLAUDE_SUBSTRINGS):
-        return False
-    return not any(v in m for v in _LEGACY_MANUAL_THINKING_CLAUDE_SUBSTRINGS)
+    return any(v in model for v in _NO_SAMPLING_PARAMS_SUBSTRINGS)
 
 
 def _supports_fast_mode(model: str) -> bool:
@@ -877,7 +821,6 @@ def _read_claude_code_credentials_from_keychain() -> Optional[Dict[str, Any]]:
             capture_output=True,
             text=True,
             timeout=5,
-            stdin=subprocess.DEVNULL,
         )
     except (OSError, subprocess.TimeoutExpired):
         logger.debug("Keychain: security command not available or timed out")
@@ -1220,10 +1163,7 @@ def run_oauth_setup_token() -> Optional[str]:
             "Install it with: npm install -g @anthropic-ai/claude-code"
         )
 
-    # Run interactively — stdin/stdout/stderr inherited so the user can
-    # complete the OAuth login prompt. Must keep inherited stdin; the TUI-EOF
-    # concern does not apply to an interactive login the user explicitly
-    # invokes.  noqa: subprocess-stdin
+    # Run interactively — stdin/stdout/stderr inherited so user can interact
     try:
         subprocess.run([claude_path, "setup-token"])
     except (KeyboardInterrupt, EOFError):
@@ -1571,15 +1511,6 @@ def _convert_content_part_to_anthropic(part: Any) -> Optional[Dict[str, Any]]:
 
     if ptype == "input_text":
         block: Dict[str, Any] = {"type": "text", "text": part.get("text", "")}
-    elif ptype == "text":
-        # A stored Anthropic text block. Rebuild from whitelisted fields only —
-        # SDK response text blocks carry output-only siblings (parsed_output,
-        # citations=None) that the Messages INPUT schema rejects with HTTP 400
-        # "Extra inputs are not permitted". Do NOT dict(part) it verbatim.
-        block = {"type": "text", "text": part.get("text", "")}
-        cits = part.get("citations")
-        if isinstance(cits, list) and cits:
-            block["citations"] = cits
     elif ptype in {"image_url", "input_image"}:
         image_value = part.get("image_url", {})
         url = image_value.get("url", "") if isinstance(image_value, dict) else str(image_value or "")
@@ -1694,58 +1625,6 @@ def _content_parts_to_anthropic_blocks(parts: Any) -> List[Dict[str, Any]]:
     return out
 
 
-def _sanitize_replay_block(b: Dict[str, Any]) -> Optional[Dict[str, Any]]:
-    """Strip output-only fields from a stored Anthropic content block so it is
-    valid as REQUEST input on replay.
-
-    The SDK response objects carry output-only attributes that the Messages
-    *input* schema forbids ("Extra inputs are not permitted"): text blocks get
-    ``parsed_output``/``citations`` (when null), tool_use blocks get ``caller``,
-    etc. ``normalize_response`` captured blocks verbatim via ``_to_plain_data``,
-    so these leak back as input on the next turn → HTTP 400.
-
-    Whitelist per type (NOT a blacklist) so future SDK output-only fields can't
-    reintroduce the bug. Returns a clean block, or None to drop it.
-    """
-    if not isinstance(b, dict):
-        return None
-    btype = b.get("type")
-    if btype == "text":
-        out: Dict[str, Any] = {"type": "text", "text": b.get("text", "")}
-        # citations is input-valid ONLY when it's a non-empty list; the SDK
-        # emits citations=None on responses, which the input schema rejects.
-        cits = b.get("citations")
-        if isinstance(cits, list) and cits:
-            out["citations"] = cits
-        if isinstance(b.get("cache_control"), dict):
-            out["cache_control"] = b["cache_control"]
-        return out
-    if btype == "thinking":
-        out = {"type": "thinking", "thinking": b.get("thinking", "")}
-        if b.get("signature"):
-            out["signature"] = b["signature"]
-        return out
-    if btype == "redacted_thinking":
-        # Only valid with its data payload; drop if missing.
-        return {"type": "redacted_thinking", "data": b["data"]} if b.get("data") else None
-    if btype == "tool_use":
-        out = {
-            "type": "tool_use",
-            "id": _sanitize_tool_id(b.get("id", "")),
-            "name": b.get("name", ""),
-            "input": b.get("input", {}),
-        }
-        if isinstance(b.get("cache_control"), dict):
-            out["cache_control"] = b["cache_control"]
-        return out
-    if btype == "image":
-        src = b.get("source")
-        return {"type": "image", "source": src} if isinstance(src, dict) else None
-    # Unknown/unsupported block type on the input path — drop rather than risk
-    # another "Extra inputs are not permitted".
-    return None
-
-
 def _convert_assistant_message(m: Dict[str, Any]) -> Dict[str, Any]:
     """Convert an assistant message to Anthropic content blocks.
 
@@ -1753,55 +1632,6 @@ def _convert_assistant_message(m: Dict[str, Any]) -> Dict[str, Any]:
     reasoning_content injection for Kimi/DeepSeek endpoints.
     """
     content = m.get("content", "")
-    # Anthropic interleaved-thinking fast path: when this turn carries a
-    # verbatim, order-preserving block list (set by normalize_response only
-    # for turns that interleave SIGNED thinking with tool_use), replay it.
-    # Each block is run through _sanitize_replay_block to strip output-only
-    # SDK fields (parsed_output, caller, citations=None, …) that the Messages
-    # INPUT schema forbids — replaying them verbatim caused HTTP 400 "Extra
-    # inputs are not permitted" (text.parsed_output). Block ORDER is preserved
-    # (the reason this channel exists); only forbidden sibling fields are
-    # dropped, leaving thinking signatures and tool_use id/name/input intact.
-    ordered_blocks = m.get("anthropic_content_blocks")
-    if isinstance(ordered_blocks, list) and ordered_blocks:
-        # Re-source each tool_use input from the stored tool_calls map rather
-        # than the captured block. The ordered-blocks list captures tool_use
-        # input from the RAW API response (normalize_response), which is NOT
-        # credential-redacted; tool_calls[].function.arguments IS redacted at
-        # storage time (build_assistant_message, #19798). Replaying the raw
-        # block input would resurrect a secret the model inlined into a tool
-        # call (e.g. terminal(command="curl -H 'Authorization: Bearer sk-...'")
-        # onto the wire, even though the same value is redacted everywhere else
-        # in history. Keying by sanitized tool id preserves interleave order
-        # (the reason this channel exists) while swapping in the redacted
-        # input. Adapted from #36071 (replay-time tool-input re-sourcing).
-        redacted_input_by_id: Dict[str, Any] = {}
-        for tc in m.get("tool_calls", []) or []:
-            if not isinstance(tc, dict):
-                continue
-            fn = tc.get("function", {}) or {}
-            raw_args = fn.get("arguments", "{}")
-            try:
-                parsed_args = json.loads(raw_args) if isinstance(raw_args, str) else raw_args
-            except (json.JSONDecodeError, ValueError):
-                parsed_args = {}
-            redacted_input_by_id[_sanitize_tool_id(tc.get("id", ""))] = parsed_args
-        replayed: List[Dict[str, Any]] = []
-        for b in ordered_blocks:
-            clean = _sanitize_replay_block(b)
-            if clean is None:
-                continue
-            if clean.get("type") == "tool_use":
-                # Override raw (un-redacted) input with the redacted copy when
-                # we have one for this id; fall back to the sanitized block
-                # input only if the tool_call is missing (shape mismatch).
-                redacted = redacted_input_by_id.get(clean.get("id", ""))
-                if redacted is not None:
-                    clean["input"] = redacted
-            replayed.append(clean)
-        if replayed:
-            return {"role": "assistant", "content": replayed}
-
     blocks = _extract_preserved_thinking_blocks(m)
     if content:
         if isinstance(content, list):
@@ -2471,43 +2301,3 @@ def build_anthropic_kwargs(
         kwargs["extra_headers"] = {"anthropic-beta": ",".join(betas)}
 
     return kwargs
-
-
-# Keys that belong exclusively to the OpenAI Responses / Codex API shape.
-# The Anthropic Messages SDK (``messages.create()`` / ``messages.stream()``)
-# raises ``TypeError: ... got an unexpected keyword argument`` on any of them.
-_RESPONSES_ONLY_KWARGS = frozenset(
-    {"instructions", "input", "store", "parallel_tool_calls"}
-)
-
-
-def sanitize_anthropic_kwargs(api_kwargs: Any, *, log_prefix: str = "") -> Any:
-    """Drop Responses-API-only keys before an Anthropic Messages SDK call.
-
-    Defensive boundary guard for #31673: under rare api_mode-flip races
-    (e.g. a concurrent auxiliary call mutating a shared agent between the
-    kwargs build and the stream dispatch), a Responses-shaped payload
-    carrying ``instructions=`` can reach ``messages.stream()`` /
-    ``messages.create()``. The Anthropic SDK rejects it with a
-    non-retryable ``TypeError`` that nukes the whole turn and propagates
-    the entire fallback chain.
-
-    Mutates ``api_kwargs`` in place and returns it. When a foreign key is
-    present we log a WARNING so the underlying race stays visible in the
-    wild instead of being silently papered over.
-    """
-    if not isinstance(api_kwargs, dict):
-        return api_kwargs
-    leaked = _RESPONSES_ONLY_KWARGS.intersection(api_kwargs)
-    if leaked:
-        for _key in leaked:
-            api_kwargs.pop(_key, None)
-        logger.warning(
-            "%sStripped Responses-only kwarg(s) %s from an Anthropic Messages "
-            "call (api_mode flip race — see #31673). The call will proceed; "
-            "this breadcrumb means a kwargs build ran under a Responses "
-            "api_mode while dispatch ran under anthropic_messages.",
-            log_prefix,
-            sorted(leaked),
-        )
-    return api_kwargs

agent/auxiliary_client.py

@@ -102,7 +102,7 @@ OpenAI = _OpenAIProxy()  # module-level name, resolves lazily on call/isinstance
 from agent.credential_pool import load_pool
 from hermes_cli.config import get_hermes_home
 from hermes_constants import OPENROUTER_BASE_URL
-from utils import base_url_host_matches, base_url_hostname, model_forces_max_completion_tokens, normalize_proxy_env_vars
+from utils import base_url_host_matches, base_url_hostname, normalize_proxy_env_vars
 
 logger = logging.getLogger(__name__)
 
@@ -637,6 +637,54 @@ def _pool_runtime_base_url(entry: Any, fallback: str = "") -> str:
 # calls to the Codex Responses API so callers don't need any changes.
 
 
+def _convert_content_for_responses(content: Any) -> Any:
+    """Convert chat.completions content to Responses API format.
+
+    chat.completions uses:
+      {"type": "text", "text": "..."}
+      {"type": "image_url", "image_url": {"url": "data:image/png;base64,..."}}
+
+    Responses API uses:
+      {"type": "input_text", "text": "..."}
+      {"type": "input_image", "image_url": "data:image/png;base64,..."}
+
+    If content is a plain string, it's returned as-is (the Responses API
+    accepts strings directly for text-only messages).
+    """
+    if isinstance(content, str):
+        return content
+    if not isinstance(content, list):
+        return str(content) if content else ""
+
+    converted: List[Dict[str, Any]] = []
+    for part in content:
+        if not isinstance(part, dict):
+            continue
+        ptype = part.get("type", "")
+        if ptype == "text":
+            converted.append({"type": "input_text", "text": part.get("text", "")})
+        elif ptype == "image_url":
+            # chat.completions nests the URL: {"image_url": {"url": "..."}}
+            image_data = part.get("image_url", {})
+            url = image_data.get("url", "") if isinstance(image_data, dict) else str(image_data)
+            entry: Dict[str, Any] = {"type": "input_image", "image_url": url}
+            # Preserve detail if specified
+            detail = image_data.get("detail") if isinstance(image_data, dict) else None
+            if detail:
+                entry["detail"] = detail
+            converted.append(entry)
+        elif ptype in {"input_text", "input_image"}:
+            # Already in Responses format — pass through
+            converted.append(part)
+        else:
+            # Unknown content type — try to preserve as text
+            text = part.get("text", "")
+            if text:
+                converted.append({"type": "input_text", "text": text})
+
+    return converted or ""
+
+
 class _CodexCompletionsAdapter:
     """Drop-in shim that accepts chat.completions.create() kwargs and
     routes them through the Codex Responses streaming API."""
@@ -649,37 +697,26 @@ class _CodexCompletionsAdapter:
         messages = kwargs.get("messages", [])
         model = kwargs.get("model", self._model)
 
-        # Separate system/instructions from replayable conversation messages,
-        # then route the rest through the SINGLE shared chat->Responses
-        # converter used by the main agent transport
-        # (agent/transports/codex.py). Maintaining a private conversion loop
-        # here let chat-style messages with role="tool" leak straight into
-        # Responses input[] — which the Responses API rejects with
-        # "Invalid value: 'tool'. Supported values are: 'assistant', 'system',
-        # 'developer', and 'user'." (issue #5709, hit hard by flush_memories()
-        # / compression replaying real session history that includes assistant
-        # tool_calls + role="tool" results). The shared converter encodes
-        # assistant tool calls as `function_call` items and tool results as
-        # `function_call_output` items with a valid call_id, so every
-        # Responses path normalizes tool history identically and cannot drift.
-        from agent.codex_responses_adapter import _chat_messages_to_responses_input
-
+        # Separate system/instructions from conversation messages.
+        # Convert chat.completions multimodal content blocks to Responses
+        # API format (input_text / input_image instead of text / image_url).
         instructions = "You are a helpful assistant."
-        replay_messages: List[Dict[str, Any]] = []
+        input_msgs: List[Dict[str, Any]] = []
         for msg in messages:
             role = msg.get("role", "user")
             content = msg.get("content") or ""
             if role == "system":
                 instructions = content if isinstance(content, str) else str(content)
             else:
-                replay_messages.append(msg)
-
-        input_items = _chat_messages_to_responses_input(replay_messages)
+                input_msgs.append({
+                    "role": role,
+                    "content": _convert_content_for_responses(content),
+                })
 
         resp_kwargs: Dict[str, Any] = {
             "model": model,
             "instructions": instructions,
-            "input": input_items or [{"role": "user", "content": ""}],
+            "input": input_msgs or [{"role": "user", "content": ""}],
             "store": False,
         }
 
@@ -2476,25 +2513,6 @@ def _is_connection_error(exc: Exception) -> bool:
     return False
 
 
-def _is_transient_transport_error(exc: Exception) -> bool:
-    """Return True for a one-off transport blip worth retrying ONCE on the
-    same provider before any provider/model fallback.
-
-    Covers connection/streaming-close errors (via the canonical
-    ``_is_connection_error`` detector, shared so the two cannot drift) plus a
-    pure 5xx/408 HTTP status. Deliberately narrow: this is the "retry the
-    same target once" gate, distinct from ``_is_payment_error`` /
-    ``_is_auth_error`` / ``_is_rate_limit_error`` which the except-chain
-    handles by switching provider, refreshing creds, or rotating the pool.
-    """
-    if _is_connection_error(exc):
-        return True
-    status = getattr(exc, "status_code", None) or getattr(
-        getattr(exc, "response", None), "status_code", None
-    )
-    return isinstance(status, int) and (status == 408 or 500 <= status < 600)
-
-
 def _is_auth_error(exc: Exception) -> bool:
     """Detect auth failures that should trigger provider-specific refresh."""
     status = getattr(exc, "status_code", None)
@@ -4300,15 +4318,13 @@ def get_auxiliary_extra_body() -> dict:
     return _nous_extra_body() if auxiliary_is_nous else {}
 
 
-def auxiliary_max_tokens_param(value: int, *, model: Optional[str] = None) -> dict:
+def auxiliary_max_tokens_param(value: int) -> dict:
     """Return the correct max tokens kwarg for the auxiliary client's provider.
-
+    
     OpenRouter and local models use 'max_tokens'. Direct OpenAI with newer
-    models (gpt-4o, gpt-4.1, gpt-5+, o-series) requires 'max_completion_tokens'.
+    models (gpt-4o, o-series, gpt-5+) requires 'max_completion_tokens'.
     The Codex adapter translates max_tokens internally, so we use max_tokens
-    for it as well. Pass ``model`` so third-party OpenAI-compatible endpoints
-    fronting the newer families are also recognised — URL-only detection
-    misses the case where a custom base URL serves e.g. ``gpt-5.4``.
+    for it as well.
     """
     custom_base = _current_custom_base_url()
     or_key = os.getenv("OPENROUTER_API_KEY")
@@ -4318,9 +4334,6 @@ def auxiliary_max_tokens_param(value: int, *, model: Optional[str] = None) -> di
             and _read_nous_auth() is None
             and base_url_hostname(custom_base) in {"api.openai.com", "api.githubcopilot.com"}):
         return {"max_completion_tokens": value}
-    # ...and for any caller serving a newer OpenAI-family model by name.
-    if model_forces_max_completion_tokens(model):
-        return {"max_completion_tokens": value}
     return {"max_tokens": value}
 
 
@@ -5171,28 +5184,8 @@ def call_llm(
     # Handle unsupported temperature, max_tokens vs max_completion_tokens retry,
     # then payment fallback.
     try:
-        # Retry ONCE on the same provider for a one-off transient transport
-        # blip (streaming-close / incomplete chunked read / 5xx / 408) before
-        # the except-chain below escalates to provider/model fallback. A
-        # single dropped connection shouldn't abandon an otherwise-healthy
-        # provider. A second failure (or any non-transient error) falls
-        # through to ``first_err`` and the existing fallback handling
-        # unchanged. This is the unified home for the transient retry that
-        # every auxiliary task (compression, memory flush, title-gen,
-        # session-search, vision) shares. (PR #16587)
-        try:
-            return _validate_llm_response(
-                client.chat.completions.create(**kwargs), task)
-        except Exception as transient_err:
-            if not _is_transient_transport_error(transient_err):
-                raise
-            logger.info(
-                "Auxiliary %s: transient transport error; retrying once on "
-                "the same provider before fallback: %s",
-                task or "call", transient_err,
-            )
-            return _validate_llm_response(
-                client.chat.completions.create(**kwargs), task)
+        return _validate_llm_response(
+            client.chat.completions.create(**kwargs), task)
     except Exception as first_err:
         if "temperature" in kwargs and _is_unsupported_temperature_error(first_err):
             retry_kwargs = dict(kwargs)
@@ -5658,22 +5651,8 @@ async def async_call_llm(
         kwargs["messages"] = _convert_openai_images_to_anthropic(kwargs["messages"])
 
     try:
-        # Retry ONCE on the same provider for a transient transport blip
-        # before the except-chain escalates to fallback — see call_llm()
-        # for the rationale. (PR #16587)
-        try:
-            return _validate_llm_response(
-                await client.chat.completions.create(**kwargs), task)
-        except Exception as transient_err:
-            if not _is_transient_transport_error(transient_err):
-                raise
-            logger.info(
-                "Auxiliary %s (async): transient transport error; retrying "
-                "once on the same provider before fallback: %s",
-                task or "call", transient_err,
-            )
-            return _validate_llm_response(
-                await client.chat.completions.create(**kwargs), task)
+        return _validate_llm_response(
+            await client.chat.completions.create(**kwargs), task)
     except Exception as first_err:
         if "temperature" in kwargs and _is_unsupported_temperature_error(first_err):
             retry_kwargs = dict(kwargs)

agent/background_review.py

@@ -449,17 +449,6 @@ def _run_review_in_thread(
             # if a future code path bypasses the cache.
             review_agent.session_start = agent.session_start
             review_agent.session_id = agent.session_id
-            # Never let the review fork compress. It shares the parent's
-            # session_id, so if it won a compression race it would rotate the
-            # parent into a NEW child that the gateway never adopts (the fork
-            # is single-lifecycle and dies right after this run_conversation).
-            # The foreground turn would then start from the stale parent and
-            # compress it again, leaving the same parent with two sibling
-            # children (issue #38727). Review also needs full context to
-            # produce a good memory/skill summary — compressing would strip
-            # detail. Both compression triggers in conversation_loop.py gate on
-            # agent.compression_enabled, so this short-circuits both paths.
-            review_agent.compression_enabled = False
 
             from model_tools import get_tool_definitions
             from hermes_cli.plugins import (

agent/chat_completion_helpers.py

@@ -139,15 +139,6 @@ def interruptible_api_call(agent, api_kwargs: dict):
     result = {"response": None, "error": None}
     request_client_holder = {"client": None, "owner_tid": None}
     request_client_lock = threading.Lock()
-    # Request-local cancellation flag. Distinct from agent._interrupt_requested
-    # because that flag is cleared at run_conversation() turn boundaries, but
-    # this daemon worker thread can outlive the turn (the gateway caches
-    # AIAgent instances per session). Tracks whether THIS specific request was
-    # cancelled by the main thread's interrupt handler, so the transport error
-    # that is the expected consequence of our own force-close isn't misread as
-    # a network bug and surfaced to the caller. (PR #6600 — cascading interrupt
-    # hang.)
-    _request_cancelled = {"value": False}
 
     def _set_request_client(client):
         with request_client_lock:
@@ -238,17 +229,6 @@ def interruptible_api_call(agent, api_kwargs: dict):
                 )
                 result["response"] = request_client.chat.completions.create(**api_kwargs)
         except Exception as e:
-            # If the request was cancelled by the main thread's interrupt
-            # handler, the transport error is the expected consequence of our
-            # own force-close, NOT a network bug. Swallow it instead of
-            # surfacing — the main thread raises InterruptedError. (#6600)
-            if _request_cancelled["value"]:
-                logger.debug(
-                    "Non-streaming worker caught %s after request cancellation — "
-                    "exiting without surfacing a network error.",
-                    type(e).__name__,
-                )
-                return
             result["error"] = e
         finally:
             _close_request_client_once("request_complete")
@@ -526,14 +506,6 @@ def interruptible_api_call(agent, api_kwargs: dict):
             break
 
         if agent._interrupt_requested:
-            # Mark THIS request cancelled before force-closing so the worker's
-            # exception handler recognizes the forced transport error as a
-            # cancel and exits cleanly instead of surfacing a network error or
-            # (in the streaming path) burning full retry cycles. (#6600)
-            _request_cancelled["value"] = True
-            logger.debug(
-                "Force-closing httpx client due to interrupt (not a network error)."
-            )
             # Force-close the in-flight worker-local HTTP connection to stop
             # token generation without poisoning the shared client used to
             # seed future retries.
@@ -952,18 +924,6 @@ def build_assistant_message(agent, assistant_message, finish_reason: str) -> dic
         if preserved:
             msg["reasoning_details"] = preserved
 
-    # Anthropic interleaved-thinking replay: when a turn interleaves signed
-    # thinking blocks with tool_use, the parallel reasoning_details +
-    # tool_calls fields lose the cross-type ordering, and reconstruction
-    # front-loads thinking — reordering signed blocks and triggering HTTP 400
-    # ("thinking ... blocks in the latest assistant message cannot be
-    # modified"). Carry the verbatim ordered block list so the adapter can
-    # replay the latest assistant message unchanged. See
-    # agent/transports/anthropic.py and agent/anthropic_adapter.py.
-    ordered_blocks = getattr(assistant_message, "anthropic_content_blocks", None)
-    if ordered_blocks:
-        msg["anthropic_content_blocks"] = ordered_blocks
-
     # Codex Responses API: preserve encrypted reasoning items for
     # multi-turn continuity. These get replayed as input on the next turn.
     codex_items = getattr(assistant_message, "codex_reasoning_items", None)
@@ -1665,14 +1625,6 @@ def interruptible_streaming_api_call(agent, api_kwargs: dict, *, on_first_delta=
     result = {"response": None, "error": None, "partial_tool_names": []}
     request_client_holder = {"client": None, "diag": None, "owner_tid": None}
     request_client_lock = threading.Lock()
-    # Request-local cancellation flag — see interruptible_api_call for the full
-    # rationale. The streaming retry loop is where the 7-minute cascading-
-    # interrupt hang originated: a force-close raised RemoteProtocolError, the
-    # loop classified it as a transient network error, and burned full retry
-    # cycles (and emitted "reconnecting" noise) on a request the user already
-    # cancelled. The token lets the worker recognize its own forced close and
-    # exit immediately instead of retrying. (PR #6600.)
-    _request_cancelled = {"value": False}
 
     def _set_request_client(client):
         with request_client_lock:
@@ -1710,14 +1662,6 @@ def interruptible_streaming_api_call(agent, api_kwargs: dict, *, on_first_delta=
     # poll loop uses this to detect stale connections that keep receiving
     # SSE keep-alive pings but no actual data.
     last_chunk_time = {"t": time.time()}
-    # Stale-stream patience, shared between the httpx socket read timeout
-    # (built in ``_call_chat_completions`` below) and the stale-stream detector
-    # (computed further down, before the worker thread starts).  Initialized
-    # here so the read-timeout builder can floor itself at the stale value and
-    # never fire before the detector.  ``None`` until the detector value is
-    # resolved, so the builder degrades to its plain default if it ever runs
-    # first.
-    _stream_stale_timeout = None
 
     def _fire_first_delta():
         if not first_delta_fired["done"] and on_first_delta:
@@ -1754,26 +1698,6 @@ def interruptible_streaming_api_call(agent, api_kwargs: dict, *, on_first_delta=
                     "Local provider detected (%s) — stream read timeout raised to %.0fs",
                     agent.base_url, _stream_read_timeout,
                 )
-            elif (
-                _stream_read_timeout == 120.0
-                and _stream_stale_timeout is not None
-                and _stream_stale_timeout != float("inf")
-                and _stream_stale_timeout > _stream_read_timeout
-            ):
-                # Cloud reasoning models (e.g. Opus) routinely pause mid-stream
-                # for minutes during extended thinking.  The stale-stream
-                # detector is deliberately scaled up to tolerate this (180–300s,
-                # see the stale-timeout block below), but the raw httpx socket
-                # read timeout defaulted to a flat 120s and fired *first* —
-                # tearing down a healthy reasoning stream before the stale
-                # detector (which owns retry + diagnostics) could act.  Keep the
-                # socket read timeout in step with the detector so it no longer
-                # preempts it.
-                _stream_read_timeout = _stream_stale_timeout
-                logger.debug(
-                    "Cloud reasoning stream — read timeout raised to %.0fs to "
-                    "match stale-stream detector", _stream_read_timeout,
-                )
         # Cap connect/pool at 60s even when provider timeout is higher.
         # connect/pool cover TCP handshake, not model inference.
         _conn_cap = min(_base_timeout, 60.0) if _provider_timeout_cfg is not None else 30.0
@@ -2026,58 +1950,6 @@ def interruptible_streaming_api_call(agent, api_kwargs: dict, *, on_first_delta=
                 "(possible upstream error or malformed SSE response)."
             )
 
-        # A stream that delivered a tool call but only partial/unparseable
-        # JSON args splits into two very different cases:
-        #
-        #   1. Provider sent finish_reason="length" → a genuine output-cap
-        #      truncation.  Boosting max_tokens on retry is the right move.
-        #
-        #   2. Provider sent NO finish_reason (the SSE simply stopped after
-        #      the opening "{" with no terminator and no [DONE]) → the
-        #      upstream dropped/stalled the connection mid tool-call.  This
-        #      is NOT an output cap — the model never reported hitting one.
-        #      Some dedicated endpoints (e.g. NVIDIA Nemotron Ultra on the
-        #      Nous dedicated endpoint) stall for minutes during large
-        #      tool-arg generation, then close the stream cleanly without a
-        #      finish_reason.  Stamping "length" here sends it down the
-        #      max_tokens-boost truncation path, which retries 3× to no
-        #      effect and finally reports the misleading "Response truncated
-        #      due to output length limit" — the red herring this guards
-        #      against.  Route it through the partial-stream-stub path
-        #      instead so the loop reports an honest mid-tool-call stream
-        #      drop and fails fast rather than escalating output budget.
-        _tool_args_dropped_no_finish = has_truncated_tool_args and finish_reason is None
-        if _tool_args_dropped_no_finish:
-            _dropped_names = [
-                (tool_calls_acc[idx]["function"]["name"] or "?")
-                for idx in sorted(tool_calls_acc)
-            ]
-            logger.warning(
-                "Stream ended with no finish_reason while a tool call's "
-                "arguments were still incomplete (tools=%s); treating as a "
-                "mid-tool-call stream drop, not an output-length truncation.",
-                _dropped_names,
-            )
-            full_reasoning = "".join(reasoning_parts) or None
-            mock_message = SimpleNamespace(
-                role=role,
-                content=full_content,
-                tool_calls=None,
-                reasoning_content=full_reasoning,
-            )
-            mock_choice = SimpleNamespace(
-                index=0,
-                message=mock_message,
-                finish_reason=FINISH_REASON_LENGTH,
-            )
-            return SimpleNamespace(
-                id=PARTIAL_STREAM_STUB_ID,
-                model=model_name,
-                choices=[mock_choice],
-                usage=usage_obj,
-                _dropped_tool_names=_dropped_names or None,
-            )
-
         effective_finish_reason = finish_reason or "stop"
         if has_truncated_tool_args:
             effective_finish_reason = "length"
@@ -2116,14 +1988,6 @@ def interruptible_streaming_api_call(agent, api_kwargs: dict, *, on_first_delta=
         # Per-attempt diagnostic dict for the retry block to consume.
         _diag = agent._stream_diag_init()
         request_client_holder["diag"] = _diag
-        # Defensive: strip Responses-only kwargs (instructions, input, ...)
-        # that can leak in under an api_mode-flip race. The Anthropic SDK
-        # raises a non-retryable TypeError on them, killing the turn. See
-        # #31673 / sanitize_anthropic_kwargs().
-        from agent.anthropic_adapter import sanitize_anthropic_kwargs
-        sanitize_anthropic_kwargs(
-            api_kwargs, log_prefix=getattr(agent, "log_prefix", "")
-        )
         # Use the Anthropic SDK's streaming context manager
         with agent._anthropic_client.messages.stream(**api_kwargs) as stream:
             # The Anthropic SDK exposes the raw httpx response on
@@ -2214,21 +2078,6 @@ def interruptible_streaming_api_call(agent, api_kwargs: dict, *, on_first_delta=
                         result["response"] = _call_chat_completions()
                     return  # success
                 except Exception as e:
-                    # If the main poll loop force-closed this request because
-                    # of an interrupt, the resulting transport error is the
-                    # expected consequence of our own close — NOT a transient
-                    # network error. Exit immediately: no retry, no fallback,
-                    # no "reconnecting" status. The outer poll loop raises
-                    # InterruptedError. This is the fix for the cascading-
-                    # interrupt hang where doomed retries burned full
-                    # stream-stale-timeout cycles. (#6600)
-                    if _request_cancelled["value"]:
-                        logger.debug(
-                            "Streaming worker caught %s after request "
-                            "cancellation — exiting without retry.",
-                            type(e).__name__,
-                        )
-                        return
                     _is_timeout = isinstance(
                         e, (_httpx.ReadTimeout, _httpx.ConnectTimeout, _httpx.PoolTimeout)
                     )
@@ -2538,15 +2387,6 @@ def interruptible_streaming_api_call(agent, api_kwargs: dict, *, on_first_delta=
             )
 
         if agent._interrupt_requested:
-            # Mark THIS request cancelled before force-closing so the worker's
-            # exception handler recognizes the forced transport error as a
-            # cancel and exits without retrying or surfacing a network error.
-            # (#6600)
-            _request_cancelled["value"] = True
-            logger.debug(
-                "Force-closing streaming httpx client due to interrupt "
-                "(not a network error)."
-            )
             try:
                 if agent.api_mode == "anthropic_messages":
                     agent._anthropic_client.close()

agent/codex_runtime.py

@@ -25,154 +25,6 @@ from typing import Any, Dict, List
 logger = logging.getLogger(__name__)
 
 
-def _coerce_usage_int(value: Any) -> int:
-    if isinstance(value, bool):
-        return 0
-    if isinstance(value, int):
-        return max(value, 0)
-    if isinstance(value, float):
-        return max(int(value), 0)
-    if isinstance(value, str):
-        try:
-            return max(int(value), 0)
-        except ValueError:
-            return 0
-    return 0
-
-
-def _record_codex_app_server_usage(agent, turn) -> dict[str, Any]:
-    """Translate Codex app-server token usage into Hermes accounting.
-
-    Codex app-server reports usage via thread/tokenUsage/updated as:
-    inputTokens, cachedInputTokens, outputTokens, reasoningOutputTokens,
-    totalTokens.
-
-    Hermes' canonical prompt bucket includes uncached input + cached input.
-    The Codex app-server protocol does not currently expose cache-write tokens,
-    so that bucket remains zero on this runtime.
-
-    Even when Codex omits usage for a turn, Hermes should still count that turn
-    as one API call for session/status accounting.
-    """
-    agent.session_api_calls += 1
-
-    usage = getattr(turn, "token_usage_last", None)
-    if not isinstance(usage, dict) or not usage:
-        if agent._session_db and agent.session_id:
-            try:
-                if not agent._session_db_created:
-                    agent._ensure_db_session()
-                agent._session_db.update_token_counts(
-                    agent.session_id,
-                    model=agent.model,
-                    api_call_count=1,
-                )
-            except Exception as exc:
-                logger.debug(
-                    "Codex app-server api-call persistence failed (session=%s): %s",
-                    agent.session_id, exc,
-                )
-        return {}
-
-    from agent.usage_pricing import CanonicalUsage, estimate_usage_cost
-
-    input_tokens = _coerce_usage_int(usage.get("inputTokens"))
-    cache_read_tokens = _coerce_usage_int(usage.get("cachedInputTokens"))
-    output_tokens = _coerce_usage_int(usage.get("outputTokens"))
-    reasoning_tokens = _coerce_usage_int(usage.get("reasoningOutputTokens"))
-    reported_total = _coerce_usage_int(usage.get("totalTokens"))
-
-    canonical_usage = CanonicalUsage(
-        input_tokens=input_tokens,
-        output_tokens=output_tokens,
-        cache_read_tokens=cache_read_tokens,
-        cache_write_tokens=0,
-        reasoning_tokens=reasoning_tokens,
-        raw_usage=usage,
-    )
-    prompt_tokens = canonical_usage.prompt_tokens
-    completion_tokens = canonical_usage.output_tokens
-    total_tokens = reported_total or canonical_usage.total_tokens
-    usage_dict = {
-        "prompt_tokens": prompt_tokens,
-        "completion_tokens": completion_tokens,
-        "total_tokens": total_tokens,
-        "input_tokens": canonical_usage.input_tokens,
-        "output_tokens": canonical_usage.output_tokens,
-        "cache_read_tokens": canonical_usage.cache_read_tokens,
-        "cache_write_tokens": canonical_usage.cache_write_tokens,
-        "reasoning_tokens": canonical_usage.reasoning_tokens,
-    }
-
-    compressor = getattr(agent, "context_compressor", None)
-    if compressor is not None:
-        try:
-            compressor.update_from_response(usage_dict)
-            context_window = getattr(turn, "model_context_window", None)
-            if isinstance(context_window, int) and context_window > 0:
-                compressor.context_length = context_window
-        except Exception:
-            logger.debug("codex app-server usage update failed", exc_info=True)
-
-    agent.session_prompt_tokens += prompt_tokens
-    agent.session_completion_tokens += completion_tokens
-    agent.session_total_tokens += total_tokens
-    agent.session_input_tokens += canonical_usage.input_tokens
-    agent.session_output_tokens += canonical_usage.output_tokens
-    agent.session_cache_read_tokens += canonical_usage.cache_read_tokens
-    agent.session_cache_write_tokens += canonical_usage.cache_write_tokens
-    agent.session_reasoning_tokens += canonical_usage.reasoning_tokens
-
-    cost_result = estimate_usage_cost(
-        agent.model,
-        canonical_usage,
-        provider=agent.provider,
-        base_url=agent.base_url,
-        api_key=getattr(agent, "api_key", ""),
-    )
-    if cost_result.amount_usd is not None:
-        agent.session_estimated_cost_usd += float(cost_result.amount_usd)
-    agent.session_cost_status = cost_result.status
-    agent.session_cost_source = cost_result.source
-
-    if agent._session_db and agent.session_id:
-        try:
-            if not agent._session_db_created:
-                agent._ensure_db_session()
-            agent._session_db.update_token_counts(
-                agent.session_id,
-                input_tokens=canonical_usage.input_tokens,
-                output_tokens=canonical_usage.output_tokens,
-                cache_read_tokens=canonical_usage.cache_read_tokens,
-                cache_write_tokens=canonical_usage.cache_write_tokens,
-                reasoning_tokens=canonical_usage.reasoning_tokens,
-                estimated_cost_usd=float(cost_result.amount_usd)
-                if cost_result.amount_usd is not None else None,
-                cost_status=cost_result.status,
-                cost_source=cost_result.source,
-                billing_provider=agent.provider,
-                billing_base_url=agent.base_url,
-                billing_mode="subscription_included"
-                if cost_result.status == "included" else None,
-                model=agent.model,
-                api_call_count=1,
-            )
-        except Exception as exc:
-            logger.debug(
-                "Codex app-server token persistence failed (session=%s, tokens=%d): %s",
-                agent.session_id, total_tokens, exc,
-            )
-
-    return {
-        **usage_dict,
-        "last_prompt_tokens": prompt_tokens,
-        "estimated_cost_usd": float(cost_result.amount_usd)
-        if cost_result.amount_usd is not None else None,
-        "cost_status": cost_result.status,
-        "cost_source": cost_result.source,
-    }
-
-
 def run_codex_app_server_turn(
     agent,
     *,
@@ -268,8 +120,6 @@ def run_codex_app_server_turn(
     agent._iters_since_skill = (
         getattr(agent, "_iters_since_skill", 0) + turn.tool_iterations
     )
-    usage_result = _record_codex_app_server_usage(agent, turn)
-    api_calls = 1
 
     # Now check the skill nudge AFTER iters were incremented — same
     # pattern the chat_completions path uses (line ~15432).
@@ -314,13 +164,12 @@ def run_codex_app_server_turn(
     return {
         "final_response": turn.final_text,
         "messages": messages,
-        "api_calls": api_calls,
+        "api_calls": 1,  # one app-server "turn" maps to one logical API call
         "completed": not turn.interrupted and turn.error is None,
         "partial": turn.interrupted or turn.error is not None,
         "error": turn.error,
         "codex_thread_id": turn.thread_id,
         "codex_turn_id": turn.turn_id,
-        **usage_result,
     }
 
 

agent/coding_context.py

@@ -1,700 +0,0 @@
-"""Coding-context awareness — base Hermes, every interactive surface.
-
-When the user runs Hermes inside a code workspace (CLI, TUI, desktop app, or an
-editor over ACP), Hermes shifts into a **coding posture**. This module is the
-single place that decides whether we're in that posture and what it implies,
-so the rest of the codebase never re-derives "are we coding?" on its own.
-
-Architecture — one seam, many consumers
-----------------------------------------
-The posture is modelled as a frozen :class:`RuntimeMode` selected from a small
-:class:`ContextProfile` registry (today: ``coding`` and ``general``). A profile
-is *data* — it declares the toolset to collapse to, the operating brief to
-inject, and hints for other domains (model routing, memory, subagents). Every
-domain reads the same resolved object instead of probing git/config itself:
-
-  * **System prompt** — ``RuntimeMode.system_blocks()`` → the operating brief +
-    a live git/workspace snapshot (``agent/system_prompt.py``).
-  * **Toolset** — ``RuntimeMode.toolset_selection()`` → the ``coding`` toolset
-    plus the user's enabled MCP servers (``cli.py`` / ``tui_gateway``). Only
-    under the opt-in ``focus`` mode: the default posture is prompt-only and
-    never touches the user's configured toolsets (toolsets like messaging /
-    smart-home / music are off-by-default anyway, and someone who explicitly
-    enabled image-gen or Spotify shouldn't lose it for being in a git repo).
-  * **Delegation** — subagents inherit the parent's toolset and run through the
-    same prompt builder, so the coding posture propagates to children for free.
-  * **Model / memory / compression** — declared on the profile
-    (``model_hint``, ``memory_policy``) as the extension seam; consumers read
-    ``mode.profile`` rather than re-deciding.
-
-Cache safety
-------------
-The mode is resolved **once** and is immutable. The workspace snapshot is built
-once at prompt-build time and baked into the *stable* system-prompt tier — never
-re-probed per turn (that would shatter the prompt cache). Branch and dirty state
-drift mid-session, so the brief tells the model to re-check with ``git`` before
-acting on the snapshot. A ``/coding`` flip therefore only takes effect next
-session (deferred), the same contract as ``/skills install`` vs ``--now``.
-
-Activation (config ``agent.coding_context``):
-
-  * ``auto`` (default) — posture (brief + snapshot) on an interactive coding
-    surface sitting in a code workspace (git repo or recognised project root).
-    Prompt-only; toolsets untouched.
-  * ``focus`` — like ``auto``, but additionally collapses the toolset to the
-    ``coding`` set + enabled MCP servers. Explicit opt-in for a lean schema.
-  * ``on`` — force the posture anywhere (incl. non-workspaces). Prompt-only.
-  * ``off`` — disable entirely.
-"""
-
-from __future__ import annotations
-
-import json
-import logging
-import os
-import re
-import subprocess
-from dataclasses import dataclass
-from pathlib import Path
-from typing import Any, Optional
-
-logger = logging.getLogger("hermes.coding_context")
-
-CODING_TOOLSET = "coding"
-
-# Surfaces where a coding posture makes sense under ``auto``. Messaging
-# platforms (telegram, discord, slack, …) are intentionally absent — a chat bot
-# in a group is not pair-programming.
-INTERACTIVE_CODING_PLATFORMS = {"cli", "tui", "acp", "desktop", ""}
-
-# Project-root signals that mark a directory as a code workspace even when it
-# isn't (yet) a git repo. Cheap filename checks — no parsing.
-_PROJECT_MARKERS = (
-    "pyproject.toml", "setup.py", "setup.cfg", "requirements.txt",
-    "package.json", "tsconfig.json", "deno.json",
-    "Cargo.toml", "go.mod", "pom.xml", "build.gradle", "build.gradle.kts",
-    "Gemfile", "composer.json", "mix.exs", "pubspec.yaml",
-    "CMakeLists.txt", "Makefile", "Dockerfile",
-    "AGENTS.md", "CLAUDE.md", ".cursorrules",
-)
-
-# Agent-instruction files surfaced separately from manifests in the snapshot.
-_CONTEXT_FILES = ("AGENTS.md", "CLAUDE.md", ".cursorrules")
-
-# Lockfile → package manager, checked in priority order.
-_PY_LOCKFILES = (("uv.lock", "uv"), ("poetry.lock", "poetry"), ("Pipfile.lock", "pipenv"))
-_JS_LOCKFILES = (
-    ("pnpm-lock.yaml", "pnpm"), ("bun.lockb", "bun"), ("bun.lock", "bun"),
-    ("yarn.lock", "yarn"), ("package-lock.json", "npm"),
-)
-
-# package.json scripts / Makefile targets worth surfacing as verify commands.
-_VERIFY_TARGETS = ("test", "tests", "lint", "typecheck", "check", "build", "fmt", "format")
-_MAX_VERIFY_COMMANDS = 8
-_MAX_FACT_FILE_BYTES = 256 * 1024
-
-_GIT_TIMEOUT = 2.5
-
-
-# Per-model edit-format steering. Matching the edit tool format to how a model
-# was trained reduces mistakes and wasted reasoning (OpenAI/Codex handle
-# patch-style diffs best; Anthropic models — and most open-weight coding
-# models, whose RL scaffolds use str_replace-style editors — do best with
-# string-replacement). Our `patch` tool exposes both: mode="patch" (V4A
-# multi-file) and mode="replace" (find-and-swap). We nudge each family toward
-# its native format. Unknown families get nothing (the brief's neutral wording
-# stands). Substrings match the model id; aligned with TOOL_USE_ENFORCEMENT_MODELS.
-_EDIT_FORMAT_GUIDANCE: dict[str, tuple[tuple[str, ...], str]] = {
-    "patch": (
-        ("gpt", "codex"),
-        "- Edit format: author new files with `write_file`; for edits to "
-        "existing code prefer `patch` with `mode='patch'` (V4A multi-file diff) "
-        "for structured or multi-file changes — it's the diff format you handle "
-        "most reliably. Use `mode='replace'` for a single small swap.",
-    ),
-    "replace": (
-        ("claude", "sonnet", "opus", "haiku",
-         "gemini", "gemma", "deepseek", "qwen", "kimi", "glm", "grok",
-         "hermes", "llama", "mistral", "devstral", "minimax"),
-        "- Edit format: author new files with `write_file`; for edits to "
-        "existing code prefer `patch` in `mode='replace'` — match a unique "
-        "snippet and swap it. Reach for `mode='patch'` (V4A) only when an edit "
-        "genuinely spans several files at once.",
-    ),
-}
-
-
-def _model_family(model: Optional[str]) -> Optional[str]:
-    """Classify a model id into an edit-format family key, or ``None``.
-
-    Used to steer the coding posture toward the edit tool format a model was
-    trained on. Family-agnostic by design: an unrecognised model gets ``None``
-    and the operating brief's neutral edit wording applies.
-    """
-    if not model:
-        return None
-    lowered = model.lower()
-    for family, (needles, _line) in _EDIT_FORMAT_GUIDANCE.items():
-        if any(n in lowered for n in needles):
-            return family
-    return None
-
-
-def _edit_format_line(model: Optional[str]) -> str:
-    """The edit-format guidance line for this model's family (``""`` if none)."""
-    family = _model_family(model)
-    if family is None:
-        return ""
-    return _EDIT_FORMAT_GUIDANCE[family][1]
-
-
-# Operating brief for the coding posture. Tool names referenced here (read_file,
-# search_files, patch, write_file, terminal, todo) are in the coding toolset and
-# in _HERMES_CORE_TOOLS, so they're present on every surface this fires on.
-CODING_AGENT_GUIDANCE = (
-    "You are a coding agent pairing with the user inside their codebase. "
-    "Operate like a careful senior engineer.\n"
-    "\n"
-    "Gather context first:\n"
-    "- Read the relevant files with `read_file` and locate code with "
-    "`search_files` before changing anything. Trace a symbol to its definition "
-    "and usages rather than guessing its shape.\n"
-    "- Batch independent lookups: when several reads/searches don't depend on "
-    "each other, issue them together in one turn instead of one at a time.\n"
-    "- Never invent files, symbols, APIs, or imports. If you haven't seen it in "
-    "the repo, go look. Don't assume a library is available — check the project "
-    "manifest (pyproject.toml / package.json / Cargo.toml / go.mod) and how "
-    "neighbouring files import it.\n"
-    "\n"
-    "Make changes through the tools, not the chat:\n"
-    "- Edit with `patch`/`write_file`. Do NOT print code blocks to the user as "
-    "a substitute for editing — apply the change, then summarise it. Only show "
-    "code when the user explicitly asks to see it.\n"
-    "- Match the project's existing style and conventions; AGENTS.md / "
-    "CLAUDE.md / .cursorrules already in context win over your defaults. Touch "
-    "only what the task needs — no drive-by refactors, renames, or reformatting "
-    "— and add any imports/dependencies your code requires.\n"
-    "- If an edit fails to apply, re-read the file to get the current exact "
-    "contents before retrying — don't repeat a stale patch. If the same region "
-    "fails twice, rewrite the enclosing function or file with `write_file` "
-    "instead of attempting a third patch.\n"
-    "\n"
-    "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"
-    "- 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 "
-    "attempts on the same file and ask the user rather than looping.\n"
-    "- Track multi-step work with `todo`. Reference code as `path:line` instead "
-    "of pasting whole files.\n"
-    "\n"
-    "Respect the user's repo: don't commit, push, or rewrite history unless "
-    "asked, and never read, print, or commit secrets — leave `.env` and "
-    "credential files alone unless the user explicitly asks. The Workspace "
-    "block below is a snapshot from session start — re-run `git status`/"
-    "`git branch` before relying on it. Be concise: lead with the change or "
-    "answer, not a preamble."
-)
-
-
-# ── Context profiles (declarative posture definitions) ──────────────────────
-
-
-@dataclass(frozen=True)
-class ContextProfile:
-    """A named operating posture. Pure data — consumers read these fields.
-
-    ``toolset``      — collapse to this toolset (+ enabled MCP) when no explicit
-                       selection is pinned; ``None`` keeps the platform default.
-    ``guidance``     — operating brief injected into the stable system prompt;
-                       ``""`` injects nothing.
-    ``model_hint``   — routing preference key for smart model routing
-                       (extension seam; not yet consumed by the router).
-    ``memory_policy``— memory namespace/weighting hint (extension seam).
-    ``hidden_skill_categories`` — skill categories pruned from the system-prompt
-                       skill index while this posture is active. Discovery-only:
-                       nothing is disabled — ``skills_list`` still returns the
-                       full catalog and ``skill_view`` loads anything. Deny-list
-                       semantics so unknown/custom categories stay visible.
-    """
-
-    name: str
-    toolset: Optional[str] = None
-    guidance: str = ""
-    model_hint: Optional[str] = None
-    memory_policy: str = "default"
-    hidden_skill_categories: tuple[str, ...] = ()
-
-
-# Skill categories that are clearly not part of a coding workflow. Hidden from
-# the prompt's skill index in the coding posture (deny-list — anything not
-# listed here, incl. custom user categories, stays visible). Coding-adjacent
-# categories (devops, github, mcp, data-science, diagramming, research,
-# security, …) are intentionally absent.
-_NON_CODING_SKILL_CATEGORIES = (
-    "apple", "communication", "cooking", "creative", "email", "finance",
-    "gaming", "gifs", "health", "media", "music", "note-taking",
-    "productivity", "shopping", "smart-home", "social-media", "travel",
-    "yuanbao",
-)
-
-
-GENERAL_PROFILE = ContextProfile(name="general")
-CODING_PROFILE = ContextProfile(
-    name="coding",
-    toolset=CODING_TOOLSET,
-    guidance=CODING_AGENT_GUIDANCE,
-    model_hint="coding",
-    memory_policy="project",
-    hidden_skill_categories=_NON_CODING_SKILL_CATEGORIES,
-)
-
-_PROFILES: dict[str, ContextProfile] = {
-    GENERAL_PROFILE.name: GENERAL_PROFILE,
-    CODING_PROFILE.name: CODING_PROFILE,
-}
-
-
-def get_profile(name: str) -> ContextProfile:
-    """Return a registered profile, falling back to ``general``."""
-    return _PROFILES.get(name, GENERAL_PROFILE)
-
-
-# ── Helpers ─────────────────────────────────────────────────────────────────
-
-
-def _coding_mode(config: Optional[dict[str, Any]]) -> str:
-    """Return the normalized ``agent.coding_context`` mode (auto/focus/on/off)."""
-    if config is None:
-        try:
-            from hermes_cli.config import load_config
-
-            config = load_config()
-        except Exception:
-            config = {}
-    raw = ((config or {}).get("agent", {}) or {}).get("coding_context", "auto")
-    mode = str(raw).strip().lower()
-    if mode in {"focus", "strict", "lean"}:
-        return "focus"
-    if mode in {"on", "true", "yes", "1", "always"}:
-        return "on"
-    if mode in {"off", "false", "no", "0", "never"}:
-        return "off"
-    return "auto"
-
-
-def _resolve_cwd(cwd: Optional[str | Path]) -> Path:
-    if cwd:
-        return Path(cwd).expanduser()
-    try:
-        from agent.runtime_cwd import resolve_agent_cwd
-
-        return resolve_agent_cwd()
-    except Exception:
-        return Path(os.getcwd())
-
-
-def _git_root(cwd: Path) -> Optional[Path]:
-    current = cwd.resolve()
-    for parent in [current, *current.parents]:
-        if (parent / ".git").exists():
-            return parent
-    return None
-
-
-def _home() -> Optional[Path]:
-    try:
-        return Path.home().resolve()
-    except (OSError, RuntimeError):
-        return None
-
-
-def _marker_root(cwd: Path) -> Optional[Path]:
-    """Nearest ancestor that looks like a project root, or ``None``.
-
-    Walks up at most a few levels so a manifest in the workspace root counts
-    even when the user is in a subdirectory. ``$HOME`` itself is skipped — a
-    Makefile or AGENTS.md sitting in the home directory is global user config,
-    not a project-root signal.
-    """
-    current = cwd.resolve()
-    home = _home()
-    for depth, parent in enumerate([current, *current.parents]):
-        if depth > 6:
-            break
-        if parent == home:
-            continue
-        for marker in _PROJECT_MARKERS:
-            if (parent / marker).exists():
-                return parent
-    return None
-
-
-def _detect_profile_name(mode: str, platform: str, cwd_str: str) -> str:
-    """Resolve which profile applies.
-
-    ``auto``/``focus``: coding when the surface is interactive AND the cwd is a
-    code workspace (a git repo or a recognised project root). ``on``: always
-    coding. ``off``: always general.
-
-    A git repo rooted at ``$HOME`` (the dotfiles pattern) is NOT a workspace
-    signal — without the guard, every session anywhere under a dotfiles-managed
-    home directory would silently flip to the coding posture.
-
-    Detection is intentionally not memoized: it's a handful of ``stat`` calls,
-    and callers resolve the mode once per session anyway. Caching here would
-    risk a stale posture if a long-lived process (gateway/TUI) serves sessions
-    from different working directories.
-    """
-    if mode == "off":
-        return GENERAL_PROFILE.name
-    if mode == "on":
-        return CODING_PROFILE.name
-    if platform and platform.strip().lower() not in INTERACTIVE_CODING_PLATFORMS:
-        return GENERAL_PROFILE.name
-    cwd = Path(cwd_str)
-    git_root = _git_root(cwd)
-    if git_root is not None and git_root == _home():
-        git_root = None  # dotfiles repo at $HOME — not a code workspace
-    if git_root is not None or _marker_root(cwd) is not None:
-        return CODING_PROFILE.name
-    return GENERAL_PROFILE.name
-
-
-# ── RuntimeMode (the seam) ──────────────────────────────────────────────────
-
-
-@dataclass(frozen=True)
-class RuntimeMode:
-    """The resolved operating posture for a session. Immutable by construction.
-
-    Built once via :func:`resolve_runtime_mode` and consumed by every domain
-    that cares about the coding/general distinction. Never mutate or re-resolve
-    mid-session — that would break the prompt cache.
-    """
-
-    profile: ContextProfile
-    surface: str
-    cwd: Path
-    # The normalized ``agent.coding_context`` mode this posture was resolved
-    # under (auto/focus/on/off). Toolset collapse is gated on ``focus``.
-    config_mode: str = "auto"
-    # The model id this session runs (e.g. "anthropic/claude-opus-4.8"). Used
-    # only to steer edit-format guidance toward the model's family — see
-    # ``_edit_format_line``. Fixed for the session, so cache-safe.
-    model: Optional[str] = None
-
-    @property
-    def kind(self) -> str:
-        return self.profile.name
-
-    @property
-    def is_coding(self) -> bool:
-        return self.profile.name == CODING_PROFILE.name
-
-    def toolset_selection(self, config: Optional[dict[str, Any]] = None) -> Optional[list[str]]:
-        """Toolset list for this posture, or ``None`` to keep the platform default.
-
-        Non-``None`` only under the opt-in ``focus`` mode. The default posture
-        is prompt-only: most strippable toolsets are off-by-default anyway, and
-        a user who explicitly enabled one (image-gen for frontend/game assets,
-        messaging for build notifications, …) keeps it while coding.
-
-        Callers apply this only when the user hasn't pinned an explicit
-        selection (``--toolsets``, ``HERMES_TUI_TOOLSETS``, …); they never
-        override a pin. Returns the profile's toolset plus enabled MCP servers.
-        """
-        if self.config_mode != "focus":
-            return None
-        if self.profile.toolset is None:
-            return None
-        return [self.profile.toolset, *_enabled_mcp_servers(config)]
-
-    def system_blocks(self) -> list[str]:
-        """Stable system-prompt blocks for this posture (brief + workspace).
-
-        The operating brief carries a model-family edit-format nudge appended
-        to it (one cached string, not a separate block) so the model is steered
-        toward the `patch` mode it handles best — see ``_edit_format_line``.
-        """
-        if not self.is_coding:
-            return []
-        blocks: list[str] = []
-        if self.profile.guidance:
-            brief = self.profile.guidance
-            edit_line = _edit_format_line(self.model)
-            if edit_line:
-                brief = f"{brief}\n{edit_line}"
-            blocks.append(brief)
-        workspace = build_coding_workspace_block(self.cwd)
-        if workspace:
-            blocks.append(workspace)
-        return blocks
-
-    def hidden_skill_categories(self) -> frozenset[str]:
-        """Skill categories to prune from the prompt's skill index (may be empty)."""
-        return frozenset(self.profile.hidden_skill_categories)
-
-
-def resolve_runtime_mode(
-    *,
-    platform: Optional[str] = None,
-    cwd: Optional[str | Path] = None,
-    config: Optional[dict[str, Any]] = None,
-    model: Optional[str] = None,
-) -> RuntimeMode:
-    """Resolve the operating posture once. Cheap — a handful of ``stat`` calls.
-
-    This is the single entry point every domain should call. The returned
-    object is immutable and safe to cache for the session. Detection itself is
-    intentionally *not* memoized (see ``_detect_profile_name``) so a long-lived
-    process can't pin a stale posture; callers resolve once per session and
-    hold the result. ``model`` is recorded only to steer edit-format guidance;
-    it never affects detection.
-    """
-    resolved_cwd = _resolve_cwd(cwd)
-    mode = _coding_mode(config)
-    name = _detect_profile_name(
-        mode, (platform or "").strip().lower(), str(resolved_cwd)
-    )
-    return RuntimeMode(
-        profile=get_profile(name),
-        surface=platform or "",
-        cwd=resolved_cwd,
-        config_mode=mode,
-        model=model,
-    )
-
-
-# ── Back-compat surface (thin wrappers over RuntimeMode) ────────────────────
-
-
-def is_coding_context(
-    *,
-    platform: Optional[str] = None,
-    cwd: Optional[str | Path] = None,
-    config: Optional[dict[str, Any]] = None,
-) -> bool:
-    """Whether Hermes should operate in its coding posture right now."""
-    return resolve_runtime_mode(platform=platform, cwd=cwd, config=config).is_coding
-
-
-def coding_selection(
-    *,
-    platform: Optional[str] = None,
-    cwd: Optional[str | Path] = None,
-    config: Optional[dict[str, Any]] = None,
-) -> Optional[list[str]]:
-    """Toolset selection for the coding posture.
-
-    ``None`` unless the user opted into ``focus`` mode AND the posture is
-    active — the default coding posture never overrides configured toolsets.
-    """
-    return resolve_runtime_mode(
-        platform=platform, cwd=cwd, config=config
-    ).toolset_selection(config)
-
-
-def coding_system_blocks(
-    *,
-    platform: Optional[str] = None,
-    cwd: Optional[str | Path] = None,
-    config: Optional[dict[str, Any]] = None,
-    model: Optional[str] = None,
-) -> list[str]:
-    """Stable system-prompt blocks for the current posture (empty when general).
-
-    ``model`` steers the brief's edit-format nudge toward the model's family.
-    """
-    return resolve_runtime_mode(
-        platform=platform, cwd=cwd, config=config, model=model
-    ).system_blocks()
-
-
-def coding_hidden_skill_categories(
-    *,
-    platform: Optional[str] = None,
-    cwd: Optional[str | Path] = None,
-    config: Optional[dict[str, Any]] = None,
-) -> frozenset[str]:
-    """Skill categories the active posture prunes from the prompt's skill index.
-
-    Empty outside the coding posture. Discovery-only: hidden skills remain
-    loadable via ``skills_list`` / ``skill_view``.
-    """
-    return resolve_runtime_mode(
-        platform=platform, cwd=cwd, config=config
-    ).hidden_skill_categories()
-
-
-def _enabled_mcp_servers(config: Optional[dict[str, Any]]) -> list[str]:
-    """Names of MCP servers the user has enabled — kept in the coding posture.
-
-    MCP servers (figma, browser, tophat, …) are explicitly configured and part
-    of the coding workflow, not noise to strip.
-    """
-    try:
-        from hermes_cli.config import read_raw_config
-        from hermes_cli.tools_config import _parse_enabled_flag
-
-        servers = read_raw_config().get("mcp_servers") or {}
-        return [
-            str(name)
-            for name, cfg in servers.items()
-            if isinstance(cfg, dict)
-            and _parse_enabled_flag(cfg.get("enabled", True), default=True)
-        ]
-    except Exception:
-        return []
-
-
-# ── git/workspace probe ─────────────────────────────────────────────────────
-
-
-def _git(cwd: Path, *args: str) -> str:
-    try:
-        out = subprocess.run(
-            ["git", "-C", str(cwd), *args],
-            capture_output=True,
-            text=True,
-            timeout=_GIT_TIMEOUT,
-        )
-    except (OSError, subprocess.SubprocessError):
-        return ""
-    return out.stdout.strip() if out.returncode == 0 else ""
-
-
-def _parse_status(porcelain: str) -> tuple[dict[str, str], dict[str, int]]:
-    """Parse ``git status --porcelain=2 --branch`` into branch + counts."""
-    branch: dict[str, str] = {}
-    counts = {"staged": 0, "modified": 0, "untracked": 0, "conflicts": 0}
-    for line in porcelain.splitlines():
-        if line.startswith("# branch.head"):
-            branch["head"] = line.split(maxsplit=2)[-1]
-        elif line.startswith("# branch.upstream"):
-            branch["upstream"] = line.split(maxsplit=2)[-1]
-        elif line.startswith("# branch.ab"):
-            parts = line.split()
-            branch["ahead"], branch["behind"] = parts[2].lstrip("+"), parts[3].lstrip("-")
-        elif line.startswith(("1 ", "2 ")):
-            xy = line.split(maxsplit=2)[1]
-            if xy[0] != ".":
-                counts["staged"] += 1
-            if xy[1] != ".":
-                counts["modified"] += 1
-        elif line.startswith("u "):
-            counts["conflicts"] += 1
-        elif line.startswith("? "):
-            counts["untracked"] += 1
-    return branch, counts
-
-
-def _read_small(path: Path) -> str:
-    """Read a small text file, or ``""`` — never raises, never reads huge files."""
-    try:
-        if not path.is_file() or path.stat().st_size > _MAX_FACT_FILE_BYTES:
-            return ""
-        return path.read_text(encoding="utf-8", errors="replace")
-    except OSError:
-        return ""
-
-
-def _project_facts(root: Path) -> list[str]:
-    """Detected project facts for the workspace snapshot.
-
-    The point is to hand the model its *verify loop* up front — which manifest,
-    which package manager, and the exact test/lint/build commands — instead of
-    making it rediscover them every session. Cheap: stat calls plus reads of a
-    couple of small files; built once at prompt-build time (cache-safe).
-    """
-    facts: list[str] = []
-
-    manifests = [m for m in _PROJECT_MARKERS if m not in _CONTEXT_FILES and (root / m).is_file()]
-    package_managers = [
-        pm for lock, pm in (*_PY_LOCKFILES, *_JS_LOCKFILES) if (root / lock).is_file()
-    ]
-    if manifests:
-        line = f"- Project: {', '.join(manifests[:6])}"
-        if package_managers:
-            line += f" ({'/'.join(dict.fromkeys(package_managers))})"
-        facts.append(line)
-
-    verify: list[str] = []
-    if (root / "scripts" / "run_tests.sh").is_file():
-        verify.append("scripts/run_tests.sh")
-    if (root / "package.json").is_file():
-        try:
-            scripts = json.loads(_read_small(root / "package.json") or "{}").get("scripts") or {}
-        except (json.JSONDecodeError, AttributeError):
-            scripts = {}
-        js_pm = next((pm for lock, pm in _JS_LOCKFILES if (root / lock).is_file()), "npm")
-        verify.extend(f"{js_pm} run {name}" for name in _VERIFY_TARGETS if name in scripts)
-    if (root / "pytest.ini").is_file() or "[tool.pytest" in _read_small(root / "pyproject.toml"):
-        verify.append("pytest")
-    makefile = _read_small(root / "Makefile")
-    if makefile:
-        verify.extend(
-            f"make {name}" for name in _VERIFY_TARGETS
-            if re.search(rf"^{re.escape(name)}\s*:", makefile, re.MULTILINE)
-        )
-    if verify:
-        deduped = list(dict.fromkeys(verify))[:_MAX_VERIFY_COMMANDS]
-        facts.append(f"- Verify: {'; '.join(deduped)}")
-
-    context_files = [c for c in _CONTEXT_FILES if (root / c).is_file()]
-    if context_files:
-        facts.append(f"- Context files: {', '.join(context_files)}")
-
-    return facts
-
-
-def build_coding_workspace_block(cwd: Optional[str | Path] = None) -> str:
-    """Workspace snapshot for the system prompt (empty outside a workspace).
-
-    Git state (branch/status/commits) when the cwd is in a repo, plus detected
-    project facts (manifest, package manager, verify commands, context files)
-    — so marker-only (non-git) projects still get a snapshot.
-    """
-    resolved = _resolve_cwd(cwd)
-    git_root = _git_root(resolved)
-    root = git_root or _marker_root(resolved)
-    if root is None:
-        return ""
-
-    lines = ["Workspace (snapshot at session start — re-check with `git` before acting on it):"]
-    lines.append(f"- Root: {root}")
-
-    if git_root is not None:
-        branch, counts = _parse_status(_git(root, "status", "--porcelain=2", "--branch"))
-        head = branch.get("head", "")
-        if head and head != "(detached)":
-            line = f"- Branch: {head}"
-            if branch.get("upstream"):
-                line += f" \u2192 {branch['upstream']}"
-                ahead, behind = branch.get("ahead", "0"), branch.get("behind", "0")
-                if ahead != "0" or behind != "0":
-                    line += f" (ahead {ahead}, behind {behind})"
-            lines.append(line)
-        elif head == "(detached)":
-            lines.append("- Branch: (detached HEAD)")
-
-        # Linked worktree: the per-worktree git dir differs from the shared common dir.
-        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():
-            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"]),
-            ("untracked", counts["untracked"]), ("conflicts", counts["conflicts"]),
-        ) if n]
-        lines.append(f"- Status: {', '.join(dirty) if dirty else 'clean'}")
-
-        recent = _git(root, "log", "-3", "--pretty=%h %s")
-        if recent:
-            lines.append("- Recent commits:")
-            lines.extend(f"    {c}" for c in recent.splitlines())
-
-    lines.extend(_project_facts(root))
-    return "\n".join(lines)

agent/context_compressor.py

@@ -553,22 +553,6 @@ class ContextCompressor(ContextEngine):
         self.last_rough_tokens_when_real_prompt_fit = 0
         self.awaiting_real_usage_after_compression = False
 
-    def on_session_end(self, session_id: str, messages: List[Dict[str, Any]]) -> None:
-        """Clear per-session compaction state at a real session boundary.
-
-        ``_previous_summary`` is per-session iterative-summary state. It is
-        cleared on ``on_session_reset()`` (/new, /reset), but session *end*
-        (CLI exit, gateway expiry, session-id rotation) goes through
-        ``on_session_end()`` instead — which inherited a no-op from
-        ``ContextEngine``. Without clearing here, a cron/background session's
-        summary could survive on a reused compressor instance and leak into the
-        next live session via the ``_generate_summary()`` iterative-update path
-        (#38788). ``compress()`` already guards the leak at the point of use;
-        this is defense-in-depth that drops the stale summary the moment the
-        owning session ends.
-        """
-        self._previous_summary = None
-
     def update_model(
         self,
         model: str,
@@ -1834,41 +1818,6 @@ The user has requested that this compaction PRIORITISE preserving all informatio
             accumulated += msg_tokens
             cut_idx = i
 
-        # If the backward walk never broke early because the entire transcript
-        # fits within soft_ceiling, accumulated now holds the total transcript
-        # size.  Without intervention _ensure_last_user_message_in_tail pushes
-        # cut_idx forward to include the last user message, and the caller's
-        # compress_start >= compress_end guard either returns unchanged (no-op)
-        # or compresses a single message — both of which trigger the infinite
-        # compaction loop described in #40803.
-        #
-        # Fix: when the whole transcript fits in soft_ceiling, compute a
-        # meaningful cut point using the raw (non-inflated) budget so that
-        # compression actually summarizes a worthwhile middle section.
-        if cut_idx <= head_end and accumulated <= soft_ceiling and accumulated > 0:
-            # The entire compressable region fits in the soft ceiling.
-            # Re-walk with the raw budget (no 1.5x multiplier) to find a
-            # split that gives the summarizer something useful.
-            raw_budget = token_budget
-            raw_accumulated = 0
-            for j in range(n - 1, head_end - 1, -1):
-                raw_msg = messages[j]
-                raw_content = raw_msg.get("content") or ""
-                raw_len = _content_length_for_budget(raw_content)
-                raw_tok = raw_len // _CHARS_PER_TOKEN + 10
-                for tc in raw_msg.get("tool_calls") or []:
-                    if isinstance(tc, dict):
-                        args = tc.get("function", {}).get("arguments", "")
-                        raw_tok += len(args) // _CHARS_PER_TOKEN
-                if raw_accumulated + raw_tok > raw_budget and (n - j) >= min_tail:
-                    cut_idx = j
-                    break
-                raw_accumulated += raw_tok
-                cut_idx = j
-            # If the raw-budget walk also consumed everything (very small
-            # transcript), fall through — the existing fallback logic below
-            # will still force a minimal cut after head_end.
-
         # Ensure we protect at least min_tail messages
         fallback_cut = n - min_tail
         cut_idx = min(cut_idx, fallback_cut)
@@ -1971,21 +1920,6 @@ The user has requested that this compaction PRIORITISE preserving all informatio
         compress_end = self._find_tail_cut_by_tokens(messages, compress_start)
 
         if compress_start >= compress_end:
-            # No compressable window — the entire transcript fits within
-            # the tail budget (soft_ceiling).  Without recording this as
-            # an ineffective compression the anti-thrashing guard in
-            # should_compress() never fires and every subsequent turn
-            # re-triggers a no-op compression loop.  (#40803)
-            self._ineffective_compression_count += 1
-            self._last_compression_savings_pct = 0.0
-            if not self.quiet_mode:
-                logger.warning(
-                    "Compression skipped: compress_start (%d) >= compress_end (%d) "
-                    "— transcript fits within tail budget, nothing to compress. "
-                    "ineffective_compression_count=%d",
-                    compress_start, compress_end,
-                    self._ineffective_compression_count,
-                )
             return messages
 
         turns_to_summarize = messages[compress_start:compress_end]
@@ -2006,13 +1940,6 @@ The user has requested that this compaction PRIORITISE preserving all informatio
             if summary_body and not self._previous_summary:
                 self._previous_summary = summary_body
             turns_to_summarize = messages[max(compress_start, summary_idx + 1):compress_end]
-        elif self._previous_summary:
-            # No handoff summary found in the current messages, but
-            # _previous_summary is non-empty — it was set by a different
-            # (now-ended) session (e.g., a cron job, a prior /new).  Discard
-            # it so _generate_summary() does not inject cross-session content
-            # into the summarizer prompt via the iterative-update path.
-            self._previous_summary = None
 
         if not self.quiet_mode:
             logger.info(

agent/context_references.py

@@ -246,14 +246,7 @@ def _expand_file_reference(
     if not path.is_file():
         return f"{ref.raw}: path is not a file", None
     if _is_binary_file(path):
-        # A binary file can't be inlined as text, but it IS on disk (the agent's
-        # tools run where this resolves — the local cwd, or the staged copy in a
-        # remote session workspace). Returning a bare "not supported" warning
-        # with no content was a dead end: the model saw a failure and gave up
-        # (told the user the file type wasn't supported). Instead, hand it an
-        # actionable block — the path, type, size, and a nudge to use its tools —
-        # so it can read/convert/view the file itself.
-        return None, _binary_reference_block(ref, path)
+        return f"{ref.raw}: binary files are not supported", None
 
     text = path.read_text(encoding="utf-8")
     if ref.line_start is not None:
@@ -297,7 +290,6 @@ def _expand_git_reference(
             capture_output=True,
             text=True,
             timeout=30,
-            stdin=subprocess.DEVNULL,
         )
     except subprocess.TimeoutExpired:
         return f"{ref.raw}: git command timed out (30s)", None
@@ -490,7 +482,6 @@ def _rg_files(path: Path, cwd: Path, limit: int) -> list[Path] | None:
             capture_output=True,
             text=True,
             timeout=10,
-            stdin=subprocess.DEVNULL,
         )
     except (FileNotFoundError, OSError, subprocess.TimeoutExpired):
         return None
@@ -500,30 +491,6 @@ def _rg_files(path: Path, cwd: Path, limit: int) -> list[Path] | None:
     return files[:limit]
 
 
-def _human_bytes(n: int) -> str:
-    size = float(n)
-    for unit in ("B", "KB", "MB", "GB"):
-        if size < 1024 or unit == "GB":
-            return f"{int(size)} {unit}" if unit == "B" else f"{size:.1f} {unit}"
-        size /= 1024
-    return f"{size:.1f} GB"
-
-
-def _binary_reference_block(ref: ContextReference, path: Path) -> str:
-    mime, _ = mimetypes.guess_type(path.name)
-    mime = mime or "application/octet-stream"
-    try:
-        size = _human_bytes(path.stat().st_size)
-    except OSError:
-        size = "unknown size"
-    return (
-        f"📎 {ref.raw} ({mime}, {size}) — binary file, not inlined as text. "
-        f"It is available on disk at `{path}`. Use your tools to work with it "
-        f"(read or convert it, extract its text, or view/render it as needed); "
-        f"do not tell the user the file type is unsupported."
-    )
-
-
 def _file_metadata(path: Path) -> str:
     if _is_binary_file(path):
         return f"{path.stat().st_size} bytes"

agent/conversation_compression.py

@@ -507,29 +507,12 @@ def compress_context(
             agent._session_db.end_session(agent.session_id, "compression")
             old_session_id = agent.session_id
             agent.session_id = f"{datetime.now().strftime('%Y%m%d_%H%M%S')}_{uuid.uuid4().hex[:6]}"
-            # Ordering contract: the agent thread updates the contextvar here;
-            # the gateway propagates to SessionEntry after run_in_executor returns.
             try:
                 from gateway.session_context import set_current_session_id
 
                 set_current_session_id(agent.session_id)
             except Exception:
                 os.environ["HERMES_SESSION_ID"] = agent.session_id
-            # The gateway/tools session context (ContextVar + env) and the
-            # logging session context are SEPARATE mechanisms. The call above
-            # moves the former; the ``[session_id]`` tag on log lines comes
-            # from ``hermes_logging._session_context`` (set once per turn in
-            # conversation_loop.py). Without this, post-rotation log lines in
-            # the same turn keep the STALE old id while the message/DB/gateway
-            # state carry the new one — breaking log correlation exactly at the
-            # compaction boundary (see #34089). Guarded separately so a logging
-            # failure can never regress the routing update above.
-            try:
-                from hermes_logging import set_session_context
-
-                set_session_context(agent.session_id)
-            except Exception:
-                pass
             agent._session_db_created = False
             agent._session_db.create_session(
                 session_id=agent.session_id,

agent/credential_pool.py

@@ -91,7 +91,6 @@ AUTH_TYPE_OAUTH = "oauth"
 AUTH_TYPE_API_KEY = "api_key"
 
 SOURCE_MANUAL = "manual"
-SOURCE_MANUAL_DEVICE_CODE = f"{SOURCE_MANUAL}:device_code"
 
 STRATEGY_FILL_FIRST = "fill_first"
 STRATEGY_ROUND_ROBIN = "round_robin"
@@ -375,7 +374,7 @@ def _iter_custom_providers(config: Optional[dict] = None):
         yield _normalize_custom_pool_name(name), entry
 
 
-def get_custom_provider_pool_key(base_url: Optional[str], provider_name: Optional[str] = None) -> Optional[str]:
+def get_custom_provider_pool_key(base_url: str, provider_name: Optional[str] = None) -> Optional[str]:
     """Look up the custom_providers list in config.yaml and return 'custom:<name>' for a matching base_url.
 
     When provider_name is given, prefer matching by name first (solving the case where

agent/credits_tracker.py

@@ -194,71 +194,17 @@ class AgentNotice:
     id: Optional[str] = None
 
 
-# ── is_free_tier_model (local-data-only free-model check) ────────────────────
-
-
-def is_free_tier_model(model: str, base_url: str = "") -> bool:
-    """Return True when *model* is a Nous free-tier model, using ONLY local data.
-
-    Two signals, both zero-network:
-
-    1. The ``:free`` suffix — the canonical Nous free SKU marker (e.g.
-       ``nvidia/nemotron-3-ultra:free``). Free by construction on the API side
-       (spend is forced to 0 for ``:free`` ids).
-    2. A peek into the in-process pricing cache in ``hermes_cli.models``
-       (populated when the model picker fetched ``/v1/models`` pricing for
-       *base_url*). PEEK ONLY — a cache miss never triggers a fetch. This is
-       CLI/TUI-session best-effort: gateway sessions never run the picker's
-       pricing fetch, so suppression there rests entirely on the ``:free``
-       suffix (which all Nous free SKUs carry).
-
-    Fail-open to False (the depleted notice still shows) on any error: wrongly
-    showing the warning is recoverable noise; wrongly hiding it on a paid model
-    would mask a real billing block.
-    """
-    if not model:
-        return False
-    if model.endswith(":free"):
-        return True
-    if not base_url:
-        return False
-    try:
-        from hermes_cli.models import _is_model_free, _pricing_cache
-
-        # Mirror get_pricing_for_provider's key normalization: the agent's
-        # Nous base_url is /v1-suffixed (https://inference-api.nousresearch.com/v1)
-        # but the picker keys _pricing_cache on the pre-/v1 root.
-        key = base_url.rstrip("/")
-        if key.endswith("/v1"):
-            key = key[:-3].rstrip("/")
-        pricing = _pricing_cache.get(key)
-        if not pricing:
-            return False
-        return _is_model_free(model, pricing)
-    except Exception:
-        return False
-
-
 # ── evaluate_credits_notices (pure reconciliation function) ──────────────────
 
 
 def evaluate_credits_notices(
     state: CreditsState,
     latch: dict,
-    *,
-    model_is_free: bool = False,
 ) -> tuple[list[AgentNotice], list[str]]:
     """Reconcile credits notices against the latch. Mutates ``latch`` IN PLACE.
 
     latch = {"active": set[str], "seen_below_90": bool, "usage_band": Optional[int]}.
 
-    ``model_is_free``: True when the session's active model is a Nous free-tier
-    model (see :func:`is_free_tier_model`). Suppresses the ``credits.depleted``
-    notice — a depleted account on a free model can keep inferencing, so the
-    error banner is noise (and confuses free-tier users who never had credits).
-    Suppression does NOT emit the "restored" success notice; that fires only on
-    a genuine ``paid_access`` flip back to True.
-
     Returns ``(to_show: list[AgentNotice], to_clear: list[str])``.
     Caller emits to_clear FIRST, then to_show.
 
@@ -338,11 +284,7 @@ def evaluate_credits_notices(
         active.discard("credits.grant_spent")
 
     # ── depleted ─────────────────────────────────────────────────────────────
-    # Suppressed while the active model is free: inference still works there,
-    # so the error banner would just alarm users (free-tier users especially,
-    # who never had paid credits to "lose").
-    show_depleted = depleted_cond and not model_is_free
-    if show_depleted and "credits.depleted" not in active:
+    if depleted_cond and "credits.depleted" not in active:
         to_show.append(
             AgentNotice(
                 text="✕ Credit access paused · run /usage for balance",
@@ -353,23 +295,20 @@ def evaluate_credits_notices(
             )
         )
         active.add("credits.depleted")
-    elif "credits.depleted" in active and not show_depleted:
+    elif "credits.depleted" in active and not depleted_cond:
         to_clear.append("credits.depleted")
         active.discard("credits.depleted")
-        if not depleted_cond:
-            # Genuine recovery (paid_access flipped back True): also emit the
-            # success notice. A clear caused by switching to a free model while
-            # still depleted must NOT claim access was restored.
-            to_show.append(
-                AgentNotice(
-                    text="✓ Credit access restored",
-                    level="success",
-                    kind="ttl",
-                    ttl_ms=CREDITS_RESTORED_TTL_MS,
-                    key="credits.restored",
-                    id="credits.restored",
-                )
+        # Recovery: also emit the success notice
+        to_show.append(
+            AgentNotice(
+                text="✓ Credit access restored",
+                level="success",
+                kind="ttl",
+                ttl_ms=CREDITS_RESTORED_TTL_MS,
+                key="credits.restored",
+                id="credits.restored",
             )
+        )
 
     return (to_show, to_clear)
 

agent/curator.py

@@ -25,6 +25,7 @@ import json
 import logging
 import os
 import re
+import tempfile
 import threading
 from datetime import datetime, timedelta, timezone
 from pathlib import Path
@@ -32,7 +33,6 @@ from typing import Any, Callable, Dict, List, NamedTuple, Optional, Set
 
 from hermes_constants import get_hermes_home
 from tools import skill_usage
-from utils import atomic_json_write
 
 logger = logging.getLogger(__name__)
 
@@ -97,7 +97,20 @@ def load_state() -> Dict[str, Any]:
 def save_state(data: Dict[str, Any]) -> None:
     path = _state_file()
     try:
-        atomic_json_write(path, data, indent=2, sort_keys=True)
+        path.parent.mkdir(parents=True, exist_ok=True)
+        fd, tmp = tempfile.mkstemp(dir=str(path.parent), prefix=".curator_state_", suffix=".tmp")
+        try:
+            with os.fdopen(fd, "w", encoding="utf-8") as f:
+                json.dump(data, f, indent=2, sort_keys=True, ensure_ascii=False)
+                f.flush()
+                os.fsync(f.fileno())
+            os.replace(tmp, path)
+        except BaseException:
+            try:
+                os.unlink(tmp)
+            except OSError:
+                pass
+            raise
     except Exception as e:
         logger.debug("Failed to save curator state: %s", e, exc_info=True)
 
@@ -362,11 +375,6 @@ CURATOR_REVIEW_PROMPT = (
     "into ~/.hermes/skills/.archive/) is the maximum destructive action. "
     "Archives are recoverable; deletion is not.\n"
     "3. DO NOT touch skills shown as pinned=yes. Skip them entirely.\n"
-    "3b. DO NOT archive, delete, consolidate, move, or otherwise modify any "
-    "skill named in the protected built-ins list (currently: plan). These "
-    "back load-bearing UX (slash-command entry points referenced in docs and "
-    "tips) and are filtered out of the candidate list below — never resurrect "
-    "one as an archive or absorb target.\n"
     "4. DO NOT use usage counters as a reason to skip consolidation. The "
     "counters are new and often mostly zero. Judge overlap on CONTENT, "
     "not on use_count. 'use=0' is not evidence a skill is valuable; it's "

agent/display.py

@@ -858,20 +858,6 @@ def _detect_tool_failure(tool_name: str, result: str | None) -> tuple[bool, str]
     return False, ""
 
 
-def _used_free_parallel(result: str | None) -> bool:
-    """True when a web result came from Parallel's free Search MCP.
-
-    Only the keyless Parallel path tags its result with ``provider="parallel"``;
-    the paid REST path and every other provider omit it. Used to label the tool
-    line "Parallel search" / "Parallel fetch" exactly when the free MCP served
-    the call.
-    """
-    if not isinstance(result, str) or '"provider"' not in result:
-        return False
-    data = safe_json_loads(result)
-    return isinstance(data, dict) and str(data.get("provider", "")).lower() == "parallel"
-
-
 def get_cute_tool_message(
     tool_name: str, args: dict, duration: float, result: str | None = None,
 ) -> str:
@@ -909,17 +895,15 @@ def get_cute_tool_message(
         return f"{line}{failure_suffix}"
 
     if tool_name == "web_search":
-        verb = "Parallel search" if _used_free_parallel(result) else "search"
-        return _wrap(f"┊ 🔍 {verb:<9} {_trunc(args.get('query', ''), 42)}  {dur}")
+        return _wrap(f"┊ 🔍 search    {_trunc(args.get('query', ''), 42)}  {dur}")
     if tool_name == "web_extract":
-        verb = "Parallel fetch" if _used_free_parallel(result) else "fetch"
         urls = args.get("urls", [])
         if urls:
             url = urls[0] if isinstance(urls, list) else str(urls)
             domain = url.replace("https://", "").replace("http://", "").split("/")[0]
             extra = f" +{len(urls)-1}" if len(urls) > 1 else ""
-            return _wrap(f"┊ 📄 {verb:<9} {_trunc(domain, 35)}{extra}  {dur}")
-        return _wrap(f"┊ 📄 {verb:<9} pages  {dur}")
+            return _wrap(f"┊ 📄 fetch     {_trunc(domain, 35)}{extra}  {dur}")
+        return _wrap(f"┊ 📄 fetch     pages  {dur}")
     if tool_name == "terminal":
         return _wrap(f"┊ 💻 $         {_trunc(args.get('command', ''), 42)}  {dur}")
     if tool_name == "process":

agent/error_classifier.py

@@ -549,32 +549,14 @@ def classify_api_error(
             should_fallback=True,
         )
 
-    # Anthropic thinking block recovery (400).  Two distinct failure modes,
-    # same recovery (strip all reasoning_details and retry without thinking
-    # blocks — see the thinking_signature handler in conversation_loop.py):
-    #   1. Signature mismatch: a thinking block is signed against the full
-    #      turn content; any upstream mutation (context compression, session
-    #      truncation, message merging) invalidates the signature.
-    #      Pattern: "signature" + "thinking".
-    #   2. Frozen-block mutation: Anthropic rejects any change to the
-    #      thinking/redacted_thinking blocks in the *latest* assistant
-    #      message — "`thinking` or `redacted_thinking` blocks in the latest
-    #      assistant message cannot be modified. These blocks must remain as
-    #      they were in the original response."  This carries no "signature"
-    #      token, so the original pattern missed it and the turn hard-aborted
-    #      as a non-retryable client error instead of self-healing.
-    #      Pattern: "thinking" + ("cannot be modified" | "must remain as they were").
+    # Anthropic thinking block signature invalid (400).
     # Don't gate on provider — OpenRouter proxies Anthropic errors, so the
     # provider may be "openrouter" even though the error is Anthropic-specific.
-    # The combined patterns are unique enough.
+    # The message pattern ("signature" + "thinking") is unique enough.
     if (
         status_code == 400
+        and "signature" in error_msg
         and "thinking" in error_msg
-        and (
-            "signature" in error_msg
-            or "cannot be modified" in error_msg
-            or "must remain as they were" in error_msg
-        )
     ):
         return _result(
             FailoverReason.thinking_signature,
@@ -984,34 +966,6 @@ def _classify_400(
             should_fallback=False,
         )
 
-    # Request-validation errors (unsupported / unknown parameter) MUST be
-    # checked BEFORE context_overflow.  A GPT-5 model rejecting max_tokens
-    # returns:
-    #   "Unsupported parameter: 'max_tokens' is not supported with this model.
-    #    Use 'max_completion_tokens' instead."
-    # That string contains the literal substring "max_tokens", which is one of
-    # the _CONTEXT_OVERFLOW_PATTERNS — so without this guard the 400 is
-    # misclassified as context_overflow, routed into the compression loop,
-    # re-sent with the same bad parameter, and ends in "Cannot compress
-    # further".  These errors are deterministic (every retry gets the identical
-    # rejection), so classify as a non-retryable format_error and fall back.
-    #
-    # NOTE: we deliberately do NOT key off the generic ``invalid_request_error``
-    # code here — OpenAI stamps that same code on genuine context-overflow 400s,
-    # so matching it would mis-route real overflows away from compression. The
-    # unambiguous signals are the explicit "unsupported/unknown parameter"
-    # message text and the specific parameter-level error codes.
-    if (
-        any(p in error_msg for p in _REQUEST_VALIDATION_PATTERNS
-            if p != "invalid_request_error")
-        or error_code_lower in {"unknown_parameter", "unsupported_parameter"}
-    ):
-        return result_fn(
-            FailoverReason.format_error,
-            retryable=False,
-            should_fallback=True,
-        )
-
     # Context overflow from 400
     if any(p in error_msg for p in _CONTEXT_OVERFLOW_PATTERNS):
         return result_fn(

agent/image_routing.py

@@ -219,35 +219,6 @@ def _supports_vision_override(
         coerced = _coerce_capability_bool(per_model.get("supports_vision"))
         if coerced is not None:
             return coerced
-
-    # 2b. Legacy list-style custom_providers. Entries are dicts with a
-    # "name" key and a nested "models" dict. Match by provider name (which
-    # may appear as the raw name or "custom:<name>" at runtime).
-    custom_providers = cfg.get("custom_providers")
-    if isinstance(custom_providers, list):
-        # Build candidate names: the provider value and the config provider
-        # value, both raw and with "custom:" prefix stripped/added.
-        candidate_names: set = set()
-        for p in filter(None, (provider, config_provider)):
-            candidate_names.add(p)
-            if p.startswith("custom:"):
-                candidate_names.add(p[len("custom:"):])
-            else:
-                candidate_names.add(f"custom:{p}")
-        for entry_raw in custom_providers:
-            if not isinstance(entry_raw, dict):
-                continue
-            entry_name = str(entry_raw.get("name") or "").strip()
-            if entry_name not in candidate_names:
-                continue
-            models_raw = entry_raw.get("models")
-            models_cfg = models_raw if isinstance(models_raw, dict) else {}
-            per_model_raw = models_cfg.get(model)
-            per_model = per_model_raw if isinstance(per_model_raw, dict) else {}
-            coerced = _coerce_capability_bool(per_model.get("supports_vision"))
-            if coerced is not None:
-                return coerced
-
     return None
 
 

agent/lsp/install.py

@@ -262,7 +262,6 @@ def _install_npm(
             capture_output=True,
             text=True,
             timeout=300,
-            stdin=subprocess.DEVNULL,
         )
         if proc.returncode != 0:
             logger.warning(
@@ -311,7 +310,6 @@ def _install_go(pkg: str, bin_name: str) -> Optional[str]:
             text=True,
             timeout=600,
             env=env,
-            stdin=subprocess.DEVNULL,
         )
         if proc.returncode != 0:
             logger.warning(
@@ -349,7 +347,6 @@ def _install_pip(pkg: str, bin_name: str) -> Optional[str]:
             capture_output=True,
             text=True,
             timeout=300,
-            stdin=subprocess.DEVNULL,
         )
         if proc.returncode != 0:
             logger.warning(

agent/memory_manager.py

@@ -28,8 +28,6 @@ from __future__ import annotations
 import logging
 import re
 import inspect
-import threading
-from concurrent.futures import ThreadPoolExecutor
 from typing import Any, Dict, List, Optional
 
 from agent.memory_provider import MemoryProvider
@@ -37,12 +35,6 @@ from tools.registry import tool_error
 
 logger = logging.getLogger(__name__)
 
-# How long shutdown_all() waits for in-flight background sync/prefetch work
-# to drain before abandoning it. A wedged provider must never block process
-# teardown indefinitely — the worker threads are daemon, so anything still
-# running past this window dies with the interpreter.
-_SYNC_DRAIN_TIMEOUT_S = 5.0
-
 
 # ---------------------------------------------------------------------------
 # Context fencing helpers
@@ -260,13 +252,6 @@ class MemoryManager:
         self._providers: List[MemoryProvider] = []
         self._tool_to_provider: Dict[str, MemoryProvider] = {}
         self._has_external: bool = False  # True once a non-builtin provider is added
-        # Background executor for end-of-turn sync/prefetch. Lazily created on
-        # first use so the common builtin-only path spawns no extra threads.
-        # A single worker serializes a provider's writes (turn N must land
-        # before turn N+1) and caps thread growth at one per manager. See
-        # _submit_background() and the sync_all/queue_prefetch_all rationale.
-        self._sync_executor: Optional[ThreadPoolExecutor] = None
-        self._sync_executor_lock = threading.Lock()
 
     # -- Registration --------------------------------------------------------
 
@@ -390,27 +375,15 @@ class MemoryManager:
         return "\n\n".join(parts)
 
     def queue_prefetch_all(self, query: str, *, session_id: str = "") -> None:
-        """Queue background prefetch on all providers for the next turn.
-
-        Provider work is dispatched to a background worker so a slow or
-        wedged provider can never block the caller. See ``sync_all`` for
-        the full rationale (agent stuck "running" minutes after a turn).
-        """
-        providers = list(self._providers)
-        if not providers:
-            return
-
-        def _run() -> None:
-            for provider in providers:
-                try:
-                    provider.queue_prefetch(query, session_id=session_id)
-                except Exception as e:
-                    logger.debug(
-                        "Memory provider '%s' queue_prefetch failed (non-fatal): %s",
-                        provider.name, e,
-                    )
-
-        self._submit_background(_run)
+        """Queue background prefetch on all providers for the next turn."""
+        for provider in self._providers:
+            try:
+                provider.queue_prefetch(query, session_id=session_id)
+            except Exception as e:
+                logger.debug(
+                    "Memory provider '%s' queue_prefetch failed (non-fatal): %s",
+                    provider.name, e,
+                )
 
     # -- Sync ----------------------------------------------------------------
 
@@ -434,120 +407,27 @@ class MemoryManager:
         session_id: str = "",
         messages: Optional[List[Dict[str, Any]]] = None,
     ) -> None:
-        """Sync a completed turn to all providers.
-
-        Runs on a background worker thread, NOT inline on the
-        turn-completion path. A provider's ``sync_turn`` may make a
-        blocking network/daemon call (a misconfigured Hindsight daemon
-        was observed blocking ~298s before failing); doing that inline
-        held ``run_conversation`` open long after the user saw their
-        response, so every interface (CLI, TUI, gateway) kept the agent
-        marked "running" for minutes and any follow-up message triggered
-        an aggressive interrupt. Dispatching off-thread means a slow or
-        broken provider can never stall the turn — the sync simply
-        completes (or fails, logged) in the background.
-
-        Writes are serialized through a single worker so turn N lands
-        before turn N+1; provider implementations don't need their own
-        ordering guarantees.
-        """
-        providers = list(self._providers)
-        if not providers:
-            return
-
-        def _run() -> None:
-            for provider in providers:
-                try:
-                    if messages is not None and self._provider_sync_accepts_messages(provider):
-                        provider.sync_turn(
-                            user_content,
-                            assistant_content,
-                            session_id=session_id,
-                            messages=messages,
-                        )
-                    else:
-                        provider.sync_turn(
-                            user_content,
-                            assistant_content,
-                            session_id=session_id,
-                        )
-                except Exception as e:
-                    logger.warning(
-                        "Memory provider '%s' sync_turn failed: %s",
-                        provider.name, e,
-                    )
-
-        self._submit_background(_run)
-
-    # -- Background dispatch -------------------------------------------------
-
-    def _submit_background(self, fn) -> None:
-        """Run ``fn`` on the manager's background worker.
-
-        The executor is created lazily and shared across calls. If the
-        executor can't be created or has already been shut down, ``fn``
-        runs inline as a last-resort fallback — losing the async benefit
-        but never losing the write itself. ``fn`` must do its own
-        per-provider error handling; this wrapper only guards executor
-        plumbing.
-        """
-        executor = self._get_sync_executor()
-        if executor is None:
-            # Executor unavailable (shut down / creation failed) — run
-            # inline rather than drop the work. Slow, but correct.
+        """Sync a completed turn to all providers."""
+        for provider in self._providers:
             try:
-                fn()
-            except Exception as e:  # pragma: no cover - fn guards internally
-                logger.debug("Inline memory background task failed: %s", e)
-            return
-        try:
-            executor.submit(fn)
-        except RuntimeError:
-            # Executor was shut down between the get and the submit
-            # (teardown race). Fall back to inline.
-            try:
-                fn()
-            except Exception as e:  # pragma: no cover - fn guards internally
-                logger.debug("Inline memory background task failed: %s", e)
-
-    def _get_sync_executor(self) -> Optional[ThreadPoolExecutor]:
-        """Lazily create the single-worker background executor."""
-        if self._sync_executor is not None:
-            return self._sync_executor
-        with self._sync_executor_lock:
-            if self._sync_executor is None:
-                try:
-                    self._sync_executor = ThreadPoolExecutor(
-                        max_workers=1,
-                        thread_name_prefix="mem-sync",
+                if messages is not None and self._provider_sync_accepts_messages(provider):
+                    provider.sync_turn(
+                        user_content,
+                        assistant_content,
+                        session_id=session_id,
+                        messages=messages,
                     )
-                except Exception as e:  # pragma: no cover - resource exhaustion
-                    logger.warning("Failed to create memory sync executor: %s", e)
-                    return None
-            return self._sync_executor
-
-    def flush_pending(self, timeout: Optional[float] = None) -> bool:
-        """Block until queued sync/prefetch work has drained.
-
-        Single-worker executor means submitting a sentinel and waiting on
-        it guarantees every previously-submitted task has run. Returns
-        True if the barrier completed within ``timeout`` (or no executor
-        exists), False on timeout. Used at real session boundaries and by
-        tests that need to assert provider state deterministically.
-        """
-        executor = self._sync_executor
-        if executor is None:
-            return True
-        try:
-            fut = executor.submit(lambda: None)
-        except RuntimeError:
-            # Executor already shut down — nothing pending.
-            return True
-        try:
-            fut.result(timeout=timeout)
-            return True
-        except Exception:
-            return False
+                else:
+                    provider.sync_turn(
+                        user_content,
+                        assistant_content,
+                        session_id=session_id,
+                    )
+            except Exception as e:
+                logger.warning(
+                    "Memory provider '%s' sync_turn failed: %s",
+                    provider.name, e,
+                )
 
     # -- Tools ---------------------------------------------------------------
 
@@ -773,15 +653,7 @@ class MemoryManager:
                 )
 
     def shutdown_all(self) -> None:
-        """Shut down all providers (reverse order for clean teardown).
-
-        Drains the background sync/prefetch executor first (bounded by
-        ``_SYNC_DRAIN_TIMEOUT_S``) so a turn's final sync has a chance to
-        land before providers are torn down. The worker threads are
-        daemon, so anything still wedged past the drain window dies with
-        the interpreter rather than blocking exit.
-        """
-        self._drain_sync_executor()
+        """Shut down all providers (reverse order for clean teardown)."""
         for provider in reversed(self._providers):
             try:
                 provider.shutdown()
@@ -791,52 +663,6 @@ class MemoryManager:
                     provider.name, e,
                 )
 
-    def _drain_sync_executor(self) -> None:
-        """Shut down the background executor, waiting briefly for drain.
-
-        Bounded by ``_SYNC_DRAIN_TIMEOUT_S``: a wedged provider must never
-        hang process/session teardown. We stop accepting new work and
-        cancel anything still queued, then wait at most the drain timeout
-        for the currently-running task on a watcher thread. The worker is
-        daemon, so an over-running task dies with the interpreter.
-        """
-        with self._sync_executor_lock:
-            executor = self._sync_executor
-            self._sync_executor = None
-        if executor is None:
-            return
-        try:
-            # Stop accepting new work and drop anything still queued, but
-            # do NOT block here — cancel_futures cancels not-yet-started
-            # tasks; the in-flight one keeps running on its daemon thread.
-            executor.shutdown(wait=False, cancel_futures=True)
-        except TypeError:
-            # Older Python without cancel_futures kwarg.
-            try:
-                executor.shutdown(wait=False)
-            except Exception as e:  # pragma: no cover
-                logger.debug("Memory sync executor shutdown failed: %s", e)
-            return
-        except Exception as e:  # pragma: no cover
-            logger.debug("Memory sync executor shutdown failed: %s", e)
-            return
-        # Give an in-flight sync a bounded chance to finish on a watcher
-        # thread so we don't block the caller past the drain timeout.
-        drainer = threading.Thread(
-            target=lambda: self._bounded_executor_wait(executor),
-            daemon=True,
-            name="mem-sync-drain",
-        )
-        drainer.start()
-        drainer.join(timeout=_SYNC_DRAIN_TIMEOUT_S)
-
-    @staticmethod
-    def _bounded_executor_wait(executor: ThreadPoolExecutor) -> None:
-        try:
-            executor.shutdown(wait=True)
-        except Exception as e:  # pragma: no cover
-            logger.debug("Memory sync executor drain wait failed: %s", e)
-
     def initialize_all(self, session_id: str, **kwargs) -> None:
         """Initialize all providers.
 

agent/model_metadata.py

@@ -141,8 +141,6 @@ DEFAULT_CONTEXT_LENGTHS = {
     # fuzzy-match collisions (e.g. "anthropic/claude-sonnet-4" is a
     # substring of "anthropic/claude-sonnet-4.6").
     # OpenRouter-prefixed models resolve via OpenRouter live API or models.dev.
-    "claude-fable-5": 1000000,
-    "claude-fable": 1000000,
     "claude-opus-4-8": 1000000,
     "claude-opus-4.8": 1000000,
     "claude-opus-4-7": 1000000,
@@ -970,16 +968,6 @@ def parse_available_output_tokens_from_error(error_msg: str) -> Optional[int]:
         # OpenRouter/Nous phrasing of the same condition.
         "in the output" in error_lower
         and "maximum context length" in error_lower
-    ) or (
-        # LM Studio / llama.cpp / some OpenAI-compatible servers:
-        #   "This model's maximum context length is 65536 tokens. However, you
-        #    requested 65536 output tokens and your prompt contains 77409
-        #    characters ..."
-        # The "requested N output tokens" phrasing means the OUTPUT cap is the
-        # problem (the input itself fits) — reduce max_tokens, don't compress.
-        "maximum context length" in error_lower
-        and "requested" in error_lower
-        and "output tokens" in error_lower
     )
     if not is_output_cap_error:
         return None
@@ -1011,22 +999,6 @@ def parse_available_output_tokens_from_error(error_msg: str) -> Optional[int]:
         if _available >= 1:
             return _available
 
-    # LM Studio / llama.cpp style: context window is reported in tokens but the
-    # prompt size is reported in CHARACTERS, e.g.
-    #   "maximum context length is 65536 tokens ... your prompt contains 77409
-    #    characters ...".
-    # Estimate the input tokens conservatively (~3 chars/token, which
-    # over-reserves the input so the retried output cap stays safely inside the
-    # window) and leave the remainder of the window for output.
-    _m_ctx_tok = re.search(r'maximum context length is (\d+)\s*token', error_lower)
-    _m_chars = re.search(r'prompt contains (\d+)\s*character', error_lower)
-    if _m_ctx_tok and _m_chars:
-        _ctx = int(_m_ctx_tok.group(1))
-        _est_input = (int(_m_chars.group(1)) + 2) // 3
-        _available = _ctx - _est_input
-        if _available >= 1:
-            return _available
-
     return None
 
 
@@ -1712,26 +1684,6 @@ def get_model_context_length(
                 "in config.yaml to override.",
                 model, base_url, f"{DEFAULT_FALLBACK_CONTEXT:,}",
             )
-            # 3b. Before falling back to the hard 256K default, consult the
-            # hardcoded catalog as a last resort.  A proxied/custom Anthropic
-            # gateway (e.g. corporate proxy) fails the Ollama/local probes
-            # above, but the model name may still match an entry in
-            # DEFAULT_CONTEXT_LENGTHS (e.g. "claude-opus-4-8" → 1M).
-            # Without this, the early return here short-circuits the catalog
-            # lookup at step 8 and silently caps context at 256K.
-            model_lower = model.lower()
-            for default_model, length in sorted(
-                DEFAULT_CONTEXT_LENGTHS.items(),
-                key=lambda x: len(x[0]),
-                reverse=True,
-            ):
-                if default_model in model_lower:
-                    logger.info(
-                        "Using hardcoded context length %s for model %r "
-                        "(custom endpoint, catalog match on %r)",
-                        f"{length:,}", model, default_model,
-                    )
-                    return length
             return DEFAULT_FALLBACK_CONTEXT
 
     # 4. Anthropic /v1/models API (only for regular API keys, not OAuth)
@@ -1812,43 +1764,10 @@ def get_model_context_length(
         if ctx is not None:
             save_context_length(model, base_url, ctx)
             return ctx
-    # 5f. OpenRouter live /models metadata — authoritative for OpenRouter-routed
-    # models. OpenRouter's catalog carries per-model context_length (e.g.
-    # anthropic/claude-fable-5 -> 1M) and refreshes as new slugs ship, so it
-    # must win over both models.dev (step 5g) and the hardcoded family catch-all
-    # (step 8). Before this branch, an OpenRouter selection set
-    # effective_provider="openrouter", which (a) made the models.dev lookup miss
-    # brand-new slugs and (b) skipped the step-6 OR fallback (gated on `not
-    # effective_provider`), so a fresh slug like claude-fable-5 fell through to
-    # the generic "claude": 200K entry and under-reported a 1M window. Mirrors
-    # the dedicated Nous/Copilot/GMI branches above.
-    if effective_provider == "openrouter":
-        metadata = fetch_model_metadata()
-        entry = metadata.get(model)
-        if entry:
-            or_ctx = entry.get("context_length")
-            # Guard against the known OpenRouter Kimi-family 32k underreport
-            # (same class the hardcoded overrides exist to mitigate).
-            if isinstance(or_ctx, int) and or_ctx > 0 and not (
-                or_ctx == 32768 and _model_name_suggests_kimi(model)
-            ):
-                return or_ctx
-
     if effective_provider:
         from agent.models_dev import lookup_models_dev_context
         ctx = lookup_models_dev_context(effective_provider, model)
         if ctx:
-            # MiniMax M3: models.dev reports 512K but actual context is 1M.
-            # Prefer hardcoded catalog over stale probe value.
-            if _model_name_suggests_minimax_m3(model):
-                catalog = DEFAULT_CONTEXT_LENGTHS.get("minimax-m3")
-                if catalog and ctx < catalog:
-                    logger.info(
-                        "Rejecting models.dev context=%s for %r "
-                        "(MiniMax-M3 underreport); using hardcoded default %s",
-                        ctx, model, f"{catalog:,}",
-                    )
-                    ctx = catalog
             return ctx
 
     # 6. OpenRouter live API metadata — provider-unaware fallback.

agent/prompt_builder.py

@@ -885,22 +885,6 @@ def build_environment_hints() -> str:
                 f"`uname -a && whoami && pwd`."
             )
 
-    # Hermes desktop GUI — any agent running under the desktop app should know
-    # it. HERMES_DESKTOP marks the backend powering the chat; HERMES_DESKTOP_TERMINAL
-    # marks a hermes launched in the embedded terminal pane. Both set by main.cjs.
-    _truthy = ("1", "true", "yes")
-    _in_desktop = (os.getenv("HERMES_DESKTOP") or "").strip().lower() in _truthy
-    _in_desktop_term = (os.getenv("HERMES_DESKTOP_TERMINAL") or "").strip().lower() in _truthy
-    if _in_desktop or _in_desktop_term:
-        _desktop_hint = "Runtime surface: you're running inside the Hermes desktop GUI app."
-        if _in_desktop_term:
-            _desktop_hint += (
-                " You're in its embedded terminal pane, beside the GUI chat — the user can "
-                "select your output (⌥-drag on macOS, Shift-drag elsewhere) and press "
-                "⌘/Ctrl+L to send it to the chat composer."
-            )
-        hints.append(_desktop_hint)
-
     if is_wsl():
         hints.append(WSL_ENVIRONMENT_HINT)
 
@@ -1101,12 +1085,11 @@ def _skill_should_show(
 def build_skills_system_prompt(
     available_tools: "set[str] | None" = None,
     available_toolsets: "set[str] | None" = None,
-    hidden_categories: "frozenset[str] | None" = None,
 ) -> str:
     """Build a compact skill index for the system prompt.
 
     Two-layer cache:
-      1. In-process LRU dict keyed by (skills_dir, tools, toolsets, hidden)
+      1. In-process LRU dict keyed by (skills_dir, tools, toolsets)
       2. Disk snapshot (``.skills_prompt_snapshot.json``) validated by
          mtime/size manifest — survives process restarts
 
@@ -1116,12 +1099,6 @@ def build_skills_system_prompt(
     scanned alongside the local ``~/.hermes/skills/`` directory.  External dirs
     are read-only — they appear in the index but new skills are always created
     in the local dir.  Local skills take precedence when names collide.
-
-    ``hidden_categories`` (e.g. from the coding posture — see
-    agent/coding_context.py) prunes whole categories from the rendered index.
-    Discovery-only: the snapshot stores everything, ``skills_list`` /
-    ``skill_view`` still reach every skill, and a footer note tells the model
-    the full catalog exists.
     """
     skills_dir = get_skills_dir()
     external_dirs = get_all_skills_dirs()[1:]  # skip local (index 0)
@@ -1146,7 +1123,6 @@ def build_skills_system_prompt(
         tuple(sorted(str(ts) for ts in (available_toolsets or set()))),
         _platform_hint,
         tuple(sorted(disabled)),
-        tuple(sorted(hidden_categories or ())),
     )
     with _SKILLS_PROMPT_CACHE_LOCK:
         cached = _SKILLS_PROMPT_CACHE.get(cache_key)
@@ -1280,26 +1256,6 @@ def build_skills_system_prompt(
             except Exception as e:
                 logger.debug("Could not read external skill description %s: %s", desc_file, e)
 
-    # Posture-driven category pruning (e.g. non-coding skills while pairing on
-    # code). Match on the top-level category segment so nested categories
-    # ("social-media/twitter") are pruned with their parent.
-    hidden_note = ""
-    if hidden_categories:
-        before = sum(len(v) for v in skills_by_category.values())
-        skills_by_category = {
-            cat: entries
-            for cat, entries in skills_by_category.items()
-            if cat.split("/", 1)[0] not in hidden_categories
-        }
-        pruned = before - sum(len(v) for v in skills_by_category.values())
-        if pruned:
-            hidden_note = (
-                f"\n(Note: {pruned} skill(s) in categories unrelated to the "
-                "current coding context are not listed here. The full catalog "
-                "is available via skills_list if the user asks for something "
-                "outside this list.)"
-            )
-
     if not skills_by_category:
         result = ""
     else:
@@ -1348,7 +1304,6 @@ def build_skills_system_prompt(
             "</available_skills>\n"
             "\n"
             "Only proceed without loading a skill if genuinely none are relevant to the task."
-            + hidden_note
         )
 
     # ── Store in LRU cache ────────────────────────────────────────────

agent/secret_sources/bitwarden.py

@@ -274,7 +274,6 @@ def _platform_asset_name() -> str:
                 capture_output=True,
                 text=True,
                 timeout=2,
-                stdin=subprocess.DEVNULL,
             )
             if "musl" in (res.stdout + res.stderr).lower():
                 libc = "musl"
@@ -526,7 +525,6 @@ def _run_bws_list(
             capture_output=True,
             text=True,
             timeout=_BWS_RUN_TIMEOUT,
-            stdin=subprocess.DEVNULL,
         )
     except subprocess.TimeoutExpired as exc:
         raise RuntimeError(

agent/skill_preprocessing.py

@@ -74,7 +74,6 @@ def run_inline_shell(command: str, cwd: Path | None, timeout: int) -> str:
             text=True,
             timeout=max(1, int(timeout)),
             check=False,
-            stdin=subprocess.DEVNULL,
         )
     except subprocess.TimeoutExpired:
         return f"[inline-shell timeout after {timeout}s: {command}]"

agent/system_prompt.py

@@ -191,21 +191,9 @@ def build_system_prompt_parts(agent: Any, system_message: Optional[str] = None)
             )
             if toolset
         }
-        # Coding posture prunes non-coding skill categories from the index
-        # (discovery-only — skills_list/skill_view still reach everything).
-        _hidden_cats = frozenset()
-        try:
-            from agent.coding_context import coding_hidden_skill_categories
-
-            _hidden_cats = coding_hidden_skill_categories(
-                platform=agent.platform, cwd=resolve_context_cwd()
-            )
-        except Exception:
-            _hidden_cats = frozenset()
         skills_prompt = _r.build_skills_system_prompt(
             available_tools=agent.valid_tool_names,
             available_toolsets=avail_toolsets,
-            hidden_categories=_hidden_cats or None,
         )
     else:
         skills_prompt = ""
@@ -233,26 +221,6 @@ def build_system_prompt_parts(agent: Any, system_message: Optional[str] = None)
     if _env_hints:
         stable_parts.append(_env_hints)
 
-    # Coding posture (base Hermes, any interactive coding surface in a code
-    # workspace — see agent/coding_context.py). The operating brief + the live
-    # git/workspace snapshot are built once here and cached for the session;
-    # the snapshot is never re-probed per turn (that would break the prompt
-    # cache), so the brief tells the model to re-check git before relying on it.
-    if agent.valid_tool_names:
-        try:
-            from agent.coding_context import coding_system_blocks
-
-            stable_parts.extend(
-                coding_system_blocks(
-                    platform=agent.platform,
-                    cwd=resolve_context_cwd(),
-                    model=agent.model,
-                )
-            )
-        except Exception:
-            # Coding-context probing must never block prompt build.
-            pass
-
     # Local Python toolchain probe — names python/pip/uv/PEP-668 state when
     # something is non-default so the model can pick the right install
     # strategy without discovering by failure.  Emits a single line; emits

agent/tool_executor.py

@@ -417,7 +417,7 @@ def execute_tool_calls_concurrent(agent, assistant_message, messages: list, effe
 
     # ── Logging / callbacks ──────────────────────────────────────────
     tool_names_str = ", ".join(name for _, name, _, _, _, _ in parsed_calls)
-    if not agent.quiet_mode and getattr(agent, "tool_progress_mode", "all") != "off":
+    if not agent.quiet_mode:
         print(f"  ⚡ Concurrent: {num_tools} tool calls — {tool_names_str}")
         for i, (tc, name, args, middleware_trace, block_result, blocked_by_guardrail) in enumerate(parsed_calls, 1):
             args_str = json.dumps(args, ensure_ascii=False)
@@ -702,7 +702,7 @@ def execute_tool_calls_concurrent(agent, assistant_message, messages: list, effe
         if agent._should_emit_quiet_tool_messages():
             cute_msg = _get_cute_tool_message_impl(name, args, tool_duration, result=function_result)
             agent._safe_print(f"  {cute_msg}")
-        elif not agent.quiet_mode and getattr(agent, "tool_progress_mode", "all") != "off":
+        elif not agent.quiet_mode:
             _preview_str = _multimodal_text_summary(function_result)
             if agent.verbose_logging:
                 print(f"  ✅ Tool {i+1} completed in {tool_duration:.2f}s")
@@ -866,7 +866,7 @@ def execute_tool_calls_sequential(agent, assistant_message, messages: list, effe
         elif function_name == "skill_manage":
             agent._iters_since_skill = 0
 
-        if not agent.quiet_mode and getattr(agent, "tool_progress_mode", "all") != "off":
+        if not agent.quiet_mode:
             args_str = json.dumps(function_args, ensure_ascii=False)
             if agent.verbose_logging:
                 print(f"  📞 Tool {i}: {function_name}({list(function_args.keys())})")
@@ -1065,25 +1065,6 @@ def execute_tool_calls_sequential(agent, assistant_message, messages: list, effe
             tool_duration = time.time() - tool_start_time
             if agent._should_emit_quiet_tool_messages():
                 agent._vprint(f"  {_get_cute_tool_message_impl('clarify', function_args, tool_duration, result=function_result)}")
-        elif function_name == "read_terminal":
-            def _execute(next_args: dict) -> Any:
-                from tools.read_terminal_tool import read_terminal_tool as _read_terminal_tool
-                return _read_terminal_tool(
-                    start_line=next_args.get("start_line"),
-                    count=next_args.get("count"),
-                    callback=getattr(agent, "read_terminal_callback", None),
-                )
-            function_result, function_args = _run_agent_tool_execution_middleware(
-                agent,
-                function_name=function_name,
-                function_args=function_args,
-                effective_task_id=effective_task_id,
-                tool_call_id=getattr(tool_call, "id", "") or "",
-                execute=_execute,
-            )
-            tool_duration = time.time() - tool_start_time
-            if agent._should_emit_quiet_tool_messages():
-                agent._vprint(f"  {_get_cute_tool_message_impl('read_terminal', function_args, tool_duration, result=function_result)}")
         elif function_name == "delegate_task":
             tasks_arg = function_args.get("tasks")
             if tasks_arg and isinstance(tasks_arg, list):
@@ -1384,7 +1365,7 @@ def execute_tool_calls_sequential(agent, assistant_message, messages: list, effe
         # entire batch.  The model sees it on the next API iteration.
         agent._apply_pending_steer_to_tool_results(messages, 1)
 
-        if not agent.quiet_mode and getattr(agent, "tool_progress_mode", "all") != "off":
+        if not agent.quiet_mode:
             if agent.verbose_logging:
                 print(f"  ✅ Tool {i} completed in {tool_duration:.2f}s")
                 print(agent._wrap_verbose("Result: ", function_result))

agent/transports/anthropic.py

@@ -84,7 +84,7 @@ class AnthropicTransport(ProviderTransport):
         to OpenAI finish_reason, and collects reasoning_details in provider_data.
         """
         import json
-        from agent.anthropic_adapter import _to_plain_data, _sanitize_replay_block
+        from agent.anthropic_adapter import _to_plain_data
         from agent.transports.types import ToolCall
 
         strip_tool_prefix = kwargs.get("strip_tool_prefix", False)
@@ -94,40 +94,14 @@ class AnthropicTransport(ProviderTransport):
         reasoning_parts = []
         reasoning_details = []
         tool_calls = []
-        # Verbatim, order-preserving copy of every content block in the turn.
-        # Anthropic signs each thinking block against the turn content that
-        # PRECEDES it at its position; when a turn interleaves thinking and
-        # tool_use (adaptive/interleaved thinking, Claude 4.6+), the parallel
-        # reasoning_details + tool_calls lists below lose that cross-type
-        # ordering. Replaying the latest assistant message in the wrong order
-        # invalidates the signatures -> HTTP 400 "thinking ... blocks in the
-        # latest assistant message cannot be modified". Preserve the exact
-        # block sequence here so the adapter can replay it unchanged. See
-        # tests/agent/test_anthropic_thinking_block_order.py.
-        ordered_blocks = []
 
         for block in response.content:
-            block_dict = _to_plain_data(block)
-            clean_block = None
-            if isinstance(block_dict, dict):
-                # Sanitize at capture so output-only SDK fields (parsed_output,
-                # caller, citations=None, …) never persist to state.db and leak
-                # back as request input on replay → HTTP 400 "Extra inputs are
-                # not permitted". Defence-in-depth with the replay-side sanitize.
-                clean_block = _sanitize_replay_block(block_dict)
-                if clean_block is not None:
-                    ordered_blocks.append(clean_block)
             if block.type == "text":
                 text_parts.append(block.text)
-            elif block.type in ("thinking", "redacted_thinking"):
-                if block.type == "thinking":
-                    reasoning_parts.append(block.thinking)
-                # Use the sanitized block (clean_block) for reasoning_details too,
-                # since _extract_preserved_thinking_blocks replays these on the
-                # non-ordered path. Falls back to raw only if sanitize dropped it.
-                if isinstance(clean_block, dict):
-                    reasoning_details.append(clean_block)
-                elif isinstance(block_dict, dict):
+            elif block.type == "thinking":
+                reasoning_parts.append(block.thinking)
+                block_dict = _to_plain_data(block)
+                if isinstance(block_dict, dict):
                     reasoning_details.append(block_dict)
             elif block.type == "tool_use":
                 name = block.name
@@ -156,23 +130,6 @@ class AnthropicTransport(ProviderTransport):
         provider_data = {}
         if reasoning_details:
             provider_data["reasoning_details"] = reasoning_details
-        # Only worth carrying the ordered-blocks channel when the turn
-        # actually interleaves signed thinking with tool_use — that's the
-        # only shape the parallel lists reconstruct incorrectly. A turn that
-        # is purely text, or thinking-then-tools with a single leading
-        # thinking block, replays correctly without it.
-        _has_signed_thinking = any(
-            isinstance(b, dict)
-            and b.get("type") in ("thinking", "redacted_thinking")
-            and (b.get("signature") or b.get("data"))
-            for b in ordered_blocks
-        )
-        _has_tool_use = any(
-            isinstance(b, dict) and b.get("type") == "tool_use"
-            for b in ordered_blocks
-        )
-        if _has_signed_thinking and _has_tool_use:
-            provider_data["anthropic_content_blocks"] = ordered_blocks
 
         return NormalizedResponse(
             content="\n".join(text_parts) if text_parts else None,

agent/transports/codex_app_server.py

@@ -378,7 +378,6 @@ def check_codex_binary(
             capture_output=True,
             text=True,
             timeout=10,
-            stdin=subprocess.DEVNULL,
         )
     except FileNotFoundError:
         return False, (

agent/transports/codex_app_server_session.py

@@ -72,9 +72,6 @@ class TurnResult:
     error: Optional[str] = None  # Set if turn ended in a non-recoverable error
     turn_id: Optional[str] = None
     thread_id: Optional[str] = None
-    token_usage_last: Optional[dict[str, Any]] = None
-    token_usage_total: Optional[dict[str, Any]] = None
-    model_context_window: Optional[int] = None
     # Hint to the caller that the underlying codex subprocess is likely
     # wedged (turn-level timeout fired, post-tool watchdog tripped, or
     # token-refresh failure killed the child). The caller should retire
@@ -504,7 +501,6 @@ class CodexAppServerSession:
                     pending = self._client.take_notification(timeout=0)
                     if pending is None:
                         break
-                    _apply_token_usage_notification(result, pending)
                     self._track_pending_file_change(pending)
                     proj = projector.project(pending)
                     if proj.messages:
@@ -540,8 +536,6 @@ class CodexAppServerSession:
                 except Exception:  # pragma: no cover - display callback
                     logger.debug("on_event callback raised", exc_info=True)
 
-            _apply_token_usage_notification(result, note)
-
             # Track in-progress fileChange items so the approval bridge
             # can surface a real change summary when codex requests
             # approval (the approval params themselves don't carry the
@@ -808,30 +802,6 @@ class CodexAppServerSession:
         return cached
 
 
-def _apply_token_usage_notification(result: TurnResult, note: dict) -> None:
-    """Capture Codex app-server token usage updates for caller accounting.
-
-    Codex does not put token usage on turn/completed. It emits a separate
-    thread/tokenUsage/updated notification containing cumulative totals and
-    the latest turn breakdown.
-    """
-    if not isinstance(note, dict) or note.get("method") != "thread/tokenUsage/updated":
-        return
-    params = note.get("params") or {}
-    token_usage = params.get("tokenUsage") or {}
-    if not isinstance(token_usage, dict):
-        return
-    last = token_usage.get("last")
-    total = token_usage.get("total")
-    if isinstance(last, dict):
-        result.token_usage_last = dict(last)
-    if isinstance(total, dict):
-        result.token_usage_total = dict(total)
-    window = token_usage.get("modelContextWindow")
-    if isinstance(window, int) and window > 0:
-        result.model_context_window = window
-
-
 def _approval_choice_to_codex_decision(choice: str) -> str:
     """Map Hermes approval choices onto codex's CommandExecutionApprovalDecision
     / FileChangeApprovalDecision wire values.

agent/transports/types.py

@@ -121,18 +121,6 @@ class NormalizedResponse:
         pd = self.provider_data or {}
         return pd.get("reasoning_details")
 
-    @property
-    def anthropic_content_blocks(self):
-        """Verbatim, order-preserving Anthropic content blocks for a turn.
-
-        Present only when an Anthropic turn interleaves signed thinking with
-        tool_use — the one shape the parallel reasoning_details + tool_calls
-        lists reconstruct in the wrong order, invalidating thinking-block
-        signatures on replay. See agent/transports/anthropic.py.
-        """
-        pd = self.provider_data or {}
-        return pd.get("anthropic_content_blocks")
-
     @property
     def codex_reasoning_items(self):
         pd = self.provider_data or {}

agent/turn_context.py

@@ -1,388 +0,0 @@
-"""Per-turn setup for ``run_conversation`` (the turn prologue).
-
-``run_conversation`` opened with ~470 lines of straight-line setup before the
-tool-calling loop ever started: stdio guarding, runtime-main wiring, retry-counter
-resets, user-message sanitization, todo/nudge-counter hydration, system-prompt
-restore-or-build, crash-resilience persistence, preflight context compression, the
-``pre_llm_call`` plugin hook, and external-memory prefetch.
-
-All of that is *prologue* — it runs once per turn, has no back-references into the
-loop, and produces a fixed set of values the loop then consumes. ``TurnContext``
-captures those produced values; ``build_turn_context`` performs the setup work and
-returns one. ``run_conversation`` is left to unpack the context and run the loop,
-shrinking the orchestrator by the full prologue.
-
-The builder still mutates ``agent`` heavily (counters, thread id, cached prompt,
-session DB) exactly as the inline code did — those side effects are the point. The
-``TurnContext`` it returns carries only the *locals* the loop reads back.
-
-Behavior is identical to the original inline prologue; this is a pure
-move-and-name refactor with no semantic change.
-"""
-
-from __future__ import annotations
-
-import logging
-import threading
-import uuid
-from dataclasses import dataclass
-from typing import Any, Dict, List, Optional
-
-from agent.iteration_budget import IterationBudget
-from agent.model_metadata import estimate_request_tokens_rough
-
-logger = logging.getLogger(__name__)
-
-
-@dataclass
-class TurnContext:
-    """Values produced by the turn prologue and consumed by the turn loop."""
-
-    # Sanitized inbound message (surrogates stripped).
-    user_message: str
-    # Clean message preserved for transcripts / memory queries (no nudge injection).
-    original_user_message: Any
-    # Working message list for this turn (loop appends to it).
-    messages: List[Dict[str, Any]]
-    # May be reset to None by preflight compression (new session created).
-    conversation_history: Optional[List[Dict[str, Any]]]
-    # Cached system prompt active for this turn (may be rebuilt by compression).
-    active_system_prompt: Optional[str]
-    # Task / turn identifiers.
-    effective_task_id: str
-    turn_id: str
-    # Index of the current user turn within ``messages``.
-    current_turn_user_idx: int
-    # Whether the post-turn memory review should fire.
-    should_review_memory: bool = False
-    # Context contributed by ``pre_llm_call`` plugins (appended to user message).
-    plugin_user_context: str = ""
-    # External-memory prefetch result, reused across loop iterations.
-    ext_prefetch_cache: str = ""
-
-
-def build_turn_context(
-    agent,
-    user_message: str,
-    system_message: Optional[str],
-    conversation_history: Optional[List[Dict[str, Any]]],
-    task_id: Optional[str],
-    stream_callback,
-    persist_user_message: Optional[str],
-    *,
-    restore_or_build_system_prompt,
-    install_safe_stdio,
-    sanitize_surrogates,
-    summarize_user_message_for_log,
-    set_session_context,
-    set_current_write_origin,
-    ra,
-) -> TurnContext:
-    """Run the once-per-turn setup and return the loop's input context.
-
-    The callables/helpers the original prologue referenced from the
-    ``conversation_loop`` module are passed in explicitly to keep this module
-    free of an import cycle with ``agent.conversation_loop``.
-    """
-    # Guard stdio against OSError from broken pipes (systemd/headless/daemon).
-    install_safe_stdio()
-
-    agent._ensure_db_session()
-
-    # Tell auxiliary_client what the live main provider/model are for this turn.
-    try:
-        from agent.auxiliary_client import set_runtime_main
-        set_runtime_main(
-            getattr(agent, "provider", "") or "",
-            getattr(agent, "model", "") or "",
-            base_url=getattr(agent, "base_url", "") or "",
-            api_key=getattr(agent, "api_key", "") or "",
-            api_mode=getattr(agent, "api_mode", "") or "",
-        )
-    except Exception:
-        pass
-
-    # Tag log records on this thread with the session ID for ``hermes logs``.
-    set_session_context(agent.session_id)
-
-    # Bind the skill write-origin ContextVar for this thread.
-    set_current_write_origin(getattr(agent, "_memory_write_origin", "assistant_tool"))
-
-    # Restore the primary runtime if the previous turn activated fallback.
-    agent._restore_primary_runtime()
-
-    # Sanitize surrogate characters from user input.
-    if isinstance(user_message, str):
-        user_message = sanitize_surrogates(user_message)
-    if isinstance(persist_user_message, str):
-        persist_user_message = sanitize_surrogates(persist_user_message)
-
-    # Store stream callback for _interruptible_api_call to pick up.
-    agent._stream_callback = stream_callback
-    agent._persist_user_message_idx = None
-    agent._persist_user_message_override = persist_user_message
-    # Generate unique task_id if not provided to isolate VMs between tasks.
-    effective_task_id = task_id or str(uuid.uuid4())
-    agent._current_task_id = effective_task_id
-    turn_id = f"{agent.session_id or 'session'}:{effective_task_id}:{uuid.uuid4().hex[:8]}"
-    agent._current_turn_id = turn_id
-    agent._current_api_request_id = ""
-
-    # Reset retry counters and iteration budget at the start of each turn.
-    agent._invalid_tool_retries = 0
-    agent._invalid_json_retries = 0
-    agent._empty_content_retries = 0
-    agent._incomplete_scratchpad_retries = 0
-    agent._codex_incomplete_retries = 0
-    agent._thinking_prefill_retries = 0
-    agent._post_tool_empty_retried = False
-    agent._last_content_with_tools = None
-    agent._last_content_tools_all_housekeeping = False
-    agent._mute_post_response = False
-    agent._unicode_sanitization_passes = 0
-    agent._tool_guardrails.reset_for_turn()
-    agent._tool_guardrail_halt_decision = None
-    agent._vision_supported = True
-
-    # Pre-turn connection health check: clean up dead TCP connections.
-    if agent.api_mode != "anthropic_messages":
-        try:
-            if agent._cleanup_dead_connections():
-                agent._emit_status(
-                    "🔌 Detected stale connections from a previous provider "
-                    "issue — cleaned up automatically. Proceeding with fresh "
-                    "connection."
-                )
-        except Exception:
-            pass
-    # Replay compression warning through status_callback for gateway platforms.
-    if agent._compression_warning:
-        agent._replay_compression_warning()
-        agent._compression_warning = None  # send once
-
-    # NOTE: _turns_since_memory and _iters_since_skill are NOT reset here.
-    agent.iteration_budget = IterationBudget(agent.max_iterations)
-
-    # Log conversation turn start for debugging/observability.
-    _preview_text = summarize_user_message_for_log(user_message)
-    _msg_preview = (_preview_text[:80] + "...") if len(_preview_text) > 80 else _preview_text
-    _msg_preview = _msg_preview.replace("\n", " ")
-    logger.info(
-        "conversation turn: session=%s model=%s provider=%s platform=%s history=%d msg=%r",
-        agent.session_id or "none", agent.model, agent.provider or "unknown",
-        agent.platform or "unknown", len(conversation_history or []),
-        _msg_preview,
-    )
-
-    # Initialize conversation (copy to avoid mutating the caller's list).
-    messages = list(conversation_history) if conversation_history else []
-
-    # Hydrate todo store from conversation history.
-    if conversation_history and not agent._todo_store.has_items():
-        agent._hydrate_todo_store(conversation_history)
-
-    # Hydrate per-session nudge counters from persisted history (issue #22357).
-    if conversation_history and agent._user_turn_count == 0:
-        prior_user_turns = sum(
-            1 for m in conversation_history if m.get("role") == "user"
-        )
-        if prior_user_turns > 0:
-            agent._user_turn_count = prior_user_turns
-            if agent._memory_nudge_interval > 0 and agent._turns_since_memory == 0:
-                agent._turns_since_memory = prior_user_turns % agent._memory_nudge_interval
-
-    # Track user turns for memory flush and periodic nudge logic.
-    agent._user_turn_count += 1
-
-    # Reset the streaming context scrubber at the top of each turn.
-    scrubber = getattr(agent, "_stream_context_scrubber", None)
-    if scrubber is not None:
-        scrubber.reset()
-    # Reset the think scrubber for the same reason.
-    think_scrubber = getattr(agent, "_stream_think_scrubber", None)
-    if think_scrubber is not None:
-        think_scrubber.reset()
-
-    # Preserve the original user message (no nudge injection).
-    original_user_message = persist_user_message if persist_user_message is not None else user_message
-
-    # Track memory nudge trigger (turn-based, checked here).
-    should_review_memory = False
-    if (agent._memory_nudge_interval > 0
-            and "memory" in agent.valid_tool_names
-            and agent._memory_store):
-        agent._turns_since_memory += 1
-        if agent._turns_since_memory >= agent._memory_nudge_interval:
-            should_review_memory = True
-            agent._turns_since_memory = 0
-
-    # Add user message.
-    user_msg = {"role": "user", "content": user_message}
-    messages.append(user_msg)
-    current_turn_user_idx = len(messages) - 1
-    agent._persist_user_message_idx = current_turn_user_idx
-
-    if not agent.quiet_mode:
-        _print_preview = summarize_user_message_for_log(user_message)
-        agent._safe_print(
-            f"💬 Starting conversation: '{_print_preview[:60]}"
-            f"{'...' if len(_print_preview) > 60 else ''}'"
-        )
-
-    # ── System prompt (cached per session for prefix caching) ──
-    if agent._cached_system_prompt is None:
-        restore_or_build_system_prompt(agent, system_message, conversation_history)
-
-    active_system_prompt = agent._cached_system_prompt
-
-    # Crash-resilience: persist the inbound user turn as soon as the session row exists.
-    try:
-        agent._persist_session(messages, conversation_history)
-    except Exception:
-        logger.warning(
-            "Early turn-start session persistence failed for session=%s",
-            agent.session_id or "none",
-            exc_info=True,
-        )
-
-    # ── Preflight context compression ──
-    if (
-        agent.compression_enabled
-        and len(messages) > agent.context_compressor.protect_first_n
-                            + agent.context_compressor.protect_last_n + 1
-    ):
-        _preflight_tokens = estimate_request_tokens_rough(
-            messages,
-            system_prompt=active_system_prompt or "",
-            tools=agent.tools or None,
-        )
-        _compressor = agent.context_compressor
-        _defer_preflight = getattr(
-            _compressor,
-            "should_defer_preflight_to_real_usage",
-            lambda _tokens: False,
-        )
-        _preflight_deferred = _defer_preflight(_preflight_tokens)
-
-        if not _preflight_deferred:
-            _last = _compressor.last_prompt_tokens
-            # Do NOT overwrite the -1 sentinel (#36718).
-            if _last >= 0 and _preflight_tokens > _last:
-                _compressor.last_prompt_tokens = _preflight_tokens
-
-        if _preflight_deferred:
-            logger.info(
-                "Skipping preflight compression: rough estimate ~%s >= %s, "
-                "but last real provider prompt was %s after compression",
-                f"{_preflight_tokens:,}",
-                f"{_compressor.threshold_tokens:,}",
-                f"{_compressor.last_real_prompt_tokens:,}",
-            )
-        elif _compressor.should_compress(_preflight_tokens):
-            logger.info(
-                "Preflight compression: ~%s tokens >= %s threshold (model %s, ctx %s)",
-                f"{_preflight_tokens:,}",
-                f"{_compressor.threshold_tokens:,}",
-                agent.model,
-                f"{_compressor.context_length:,}",
-            )
-            agent._emit_status(
-                f"📦 Preflight compression: ~{_preflight_tokens:,} tokens "
-                f">= {_compressor.threshold_tokens:,} threshold. "
-                "This may take a moment."
-            )
-            for _pass in range(3):
-                _orig_len = len(messages)
-                messages, active_system_prompt = agent._compress_context(
-                    messages, system_message, approx_tokens=_preflight_tokens,
-                    task_id=effective_task_id,
-                )
-                if len(messages) >= _orig_len:
-                    break  # Cannot compress further
-                conversation_history = None
-                agent._empty_content_retries = 0
-                agent._thinking_prefill_retries = 0
-                agent._last_content_with_tools = None
-                agent._last_content_tools_all_housekeeping = False
-                agent._mute_post_response = False
-                _preflight_tokens = estimate_request_tokens_rough(
-                    messages,
-                    system_prompt=active_system_prompt or "",
-                    tools=agent.tools or None,
-                )
-                if not _compressor.should_compress(_preflight_tokens):
-                    break
-
-    # Plugin hook: pre_llm_call (context injected into user message, not system prompt).
-    plugin_user_context = ""
-    try:
-        from hermes_cli.plugins import invoke_hook as _invoke_hook
-        _pre_results = _invoke_hook(
-            "pre_llm_call",
-            session_id=agent.session_id,
-            task_id=effective_task_id,
-            turn_id=turn_id,
-            user_message=original_user_message,
-            conversation_history=list(messages),
-            is_first_turn=(not bool(conversation_history)),
-            model=agent.model,
-            platform=getattr(agent, "platform", None) or "",
-            sender_id=getattr(agent, "_user_id", None) or "",
-        )
-        _ctx_parts: list[str] = []
-        for r in _pre_results:
-            if isinstance(r, dict) and r.get("context"):
-                _ctx_parts.append(str(r["context"]))
-            elif isinstance(r, str) and r.strip():
-                _ctx_parts.append(r)
-        if _ctx_parts:
-            plugin_user_context = "\n\n".join(_ctx_parts)
-    except Exception as exc:
-        logger.warning("pre_llm_call hook failed: %s", exc)
-
-    # Per-turn file-mutation verifier state.
-    agent._turn_failed_file_mutations = {}
-
-    # Record the execution thread so interrupt()/clear_interrupt() can scope
-    # the tool-level interrupt signal to THIS agent's thread only.
-    agent._execution_thread_id = threading.current_thread().ident
-
-    # Clear stale per-thread interrupt state, preserving a pending interrupt.
-    ra()._set_interrupt(False, agent._execution_thread_id)
-    if agent._interrupt_requested:
-        ra()._set_interrupt(True, agent._execution_thread_id)
-        agent._interrupt_thread_signal_pending = False
-    else:
-        agent._interrupt_message = None
-        agent._interrupt_thread_signal_pending = False
-
-    # Notify memory providers of the new turn (BEFORE prefetch_all).
-    if agent._memory_manager:
-        try:
-            _turn_msg = original_user_message if isinstance(original_user_message, str) else ""
-            agent._memory_manager.on_turn_start(agent._user_turn_count, _turn_msg)
-        except Exception:
-            pass
-
-    # External memory provider: prefetch once before the tool loop.
-    ext_prefetch_cache = ""
-    if agent._memory_manager:
-        try:
-            _query = original_user_message if isinstance(original_user_message, str) else ""
-            ext_prefetch_cache = agent._memory_manager.prefetch_all(_query) or ""
-        except Exception:
-            pass
-
-    return TurnContext(
-        user_message=user_message,
-        original_user_message=original_user_message,
-        messages=messages,
-        conversation_history=conversation_history,
-        active_system_prompt=active_system_prompt,
-        effective_task_id=effective_task_id,
-        turn_id=turn_id,
-        current_turn_user_idx=current_turn_user_idx,
-        should_review_memory=should_review_memory,
-        plugin_user_context=plugin_user_context,
-        ext_prefetch_cache=ext_prefetch_cache,
-    )

agent/turn_finalizer.py

@@ -1,428 +0,0 @@
-"""Post-loop turn finalization for ``run_conversation``.
-
-Extracted from ``agent/conversation_loop.py`` as part of the god-file
-decomposition campaign (``~/.hermes/plans/god-file-decomposition.md``, Phase 1
-step 4 — the post-loop ``TurnFinalizer`` seam). ``run_conversation``'s tail
-(everything after the main tool-calling ``while`` loop) is lifted here verbatim:
-budget-exhaustion summary, trajectory save, session persist, turn diagnostics,
-response transforms, result-dict assembly, steer drain, and the memory/skill
-review trigger.
-
-Behavior-neutral: the body is moved unchanged. All ``agent.*`` side effects fire
-exactly as before; only the post-loop *locals* are passed in as keyword args, and
-the assembled ``result`` dict is returned to ``run_conversation`` which returns it
-to the caller. The function is synchronous with a single return — mirroring the
-region it replaces (no awaits, no early returns).
-
-Module ``logger`` is imported lazily inside the body (``from
-agent.conversation_loop import logger``) so this module never imports
-``agent.conversation_loop`` at import time -> no import cycle, and the log records
-keep the exact logger name (``"agent.conversation_loop"``).
-"""
-
-from __future__ import annotations
-
-import os
-
-from agent.codex_responses_adapter import _summarize_user_message_for_log
-
-
-def finalize_turn(
-    agent,
-    *,
-    final_response,
-    api_call_count,
-    interrupted,
-    failed,
-    messages,
-    conversation_history,
-    effective_task_id,
-    turn_id,
-    user_message,
-    original_user_message,
-    _should_review_memory,
-    _turn_exit_reason,
-):
-    """Run the post-loop finalization and return the turn ``result`` dict.
-
-    Lifted verbatim from ``run_conversation`` (the region after the main agent
-    loop). See module docstring.
-    """
-    from agent.conversation_loop import logger
-
-    if final_response is None and (
-        api_call_count >= agent.max_iterations
-        or agent.iteration_budget.remaining <= 0
-    ):
-        # Budget exhausted — ask the model for a summary via one extra
-        # API call with tools stripped.  _handle_max_iterations injects a
-        # user message and makes a single toolless request.
-        _turn_exit_reason = f"max_iterations_reached({api_call_count}/{agent.max_iterations})"
-        agent._emit_status(
-            f"⚠️ Iteration budget exhausted ({api_call_count}/{agent.max_iterations}) "
-            "— asking model to summarise"
-        )
-        if not agent.quiet_mode:
-            agent._safe_print(
-                f"\n⚠️  Iteration budget exhausted ({api_call_count}/{agent.max_iterations}) "
-                "— requesting summary..."
-            )
-        final_response = agent._handle_max_iterations(messages, api_call_count)
-
-        # If running as a kanban worker, signal the dispatcher that the
-        # worker could not complete (rather than treating it as a
-        # protocol violation).  The agent loop strips tools before calling
-        # _handle_max_iterations, so the model cannot call kanban_block
-        # itself — we must do it on its behalf.
-        #
-        # We route through ``_record_task_failure(outcome="timed_out")``
-        # rather than ``kanban_block`` so this counts toward the
-        # ``consecutive_failures`` counter and the dispatcher's
-        # ``failure_limit`` circuit breaker (#29747 gap 2).  Without this,
-        # a task whose worker keeps exhausting its budget would block
-        # silently each run, get auto-promoted by the operator (or never
-        # surface), and re-block in an endless loop with no signal.
-        _kanban_task = os.environ.get("HERMES_KANBAN_TASK")
-        if _kanban_task:
-            try:
-                from hermes_cli import kanban_db as _kb
-                _conn = _kb.connect()
-                try:
-                    _kb._record_task_failure(
-                        _conn,
-                        _kanban_task,
-                        error=(
-                            f"Iteration budget exhausted "
-                            f"({api_call_count}/{agent.max_iterations}) — "
-                            "task could not complete within the allowed "
-                            "iterations"
-                        ),
-                        outcome="timed_out",
-                        release_claim=True,
-                        end_run=True,
-                        event_payload_extra={
-                            "budget_used": api_call_count,
-                            "budget_max": agent.max_iterations,
-                        },
-                    )
-                    logger.info(
-                        "recorded budget-exhausted failure for task %s (%d/%d)",
-                        _kanban_task, api_call_count, agent.max_iterations,
-                    )
-                finally:
-                    try:
-                        _conn.close()
-                    except Exception:
-                        pass
-            except Exception:
-                logger.warning(
-                    "Failed to record budget-exhausted failure for task %s",
-                    _kanban_task,
-                    exc_info=True,
-                )
-
-    # Determine if conversation completed successfully
-    completed = (
-        final_response is not None
-        and api_call_count < agent.max_iterations
-        and not failed
-    )
-
-    # Save trajectory if enabled.  ``user_message`` may be a multimodal
-    # list of parts; the trajectory format wants a plain string.
-    agent._save_trajectory(messages, _summarize_user_message_for_log(user_message), completed)
-
-    # Clean up VM and browser for this task after conversation completes
-    agent._cleanup_task_resources(effective_task_id)
-
-    # Persist session to both JSON log and SQLite only after private retry
-    # scaffolding has been removed. Otherwise a later user "continue" turn
-    # can replay assistant("(empty)") / recovery nudges and fall into the
-    # same empty-response loop again.
-    agent._drop_trailing_empty_response_scaffolding(messages)
-    agent._persist_session(messages, conversation_history)
-
-    # ── Turn-exit diagnostic log ─────────────────────────────────────
-    # Always logged at INFO so agent.log captures WHY every turn ended.
-    # When the last message is a tool result (agent was mid-work), log
-    # at WARNING — this is the "just stops" scenario users report.
-    _last_msg_role = messages[-1].get("role") if messages else None
-    _last_tool_name = None
-    if _last_msg_role == "tool":
-        # Walk back to find the assistant message with the tool call
-        for _m in reversed(messages):
-            if _m.get("role") == "assistant" and _m.get("tool_calls"):
-                _tcs = _m["tool_calls"]
-                if _tcs and isinstance(_tcs[0], dict):
-                    _last_tool_name = _tcs[-1].get("function", {}).get("name")
-                break
-
-    _turn_tool_count = sum(
-        1 for m in messages
-        if isinstance(m, dict) and m.get("role") == "assistant" and m.get("tool_calls")
-    )
-    _resp_len = len(final_response) if final_response else 0
-    _budget_used = agent.iteration_budget.used if agent.iteration_budget else 0
-    _budget_max = agent.iteration_budget.max_total if agent.iteration_budget else 0
-
-    _diag_msg = (
-        "Turn ended: reason=%s model=%s api_calls=%d/%d budget=%d/%d "
-        "tool_turns=%d last_msg_role=%s response_len=%d session=%s"
-    )
-    _diag_args = (
-        _turn_exit_reason, agent.model, api_call_count, agent.max_iterations,
-        _budget_used, _budget_max,
-        _turn_tool_count, _last_msg_role, _resp_len,
-        agent.session_id or "none",
-    )
-
-    if _last_msg_role == "tool" and not interrupted:
-        # Agent was mid-work — this is the "just stops" case.
-        logger.warning(
-            "Turn ended with pending tool result (agent may appear stuck). "
-            + _diag_msg + " last_tool=%s",
-            *_diag_args, _last_tool_name,
-        )
-    else:
-        logger.info(_diag_msg, *_diag_args)
-
-    # File-mutation verifier footer.
-    # If one or more ``write_file`` / ``patch`` calls failed during this
-    # turn and were never superseded by a successful write to the same
-    # path, append an advisory footer to the assistant response.  This
-    # catches the specific case — reported by Ben Eng (#15524-adjacent)
-    # — where a model issues a batch of parallel patches, half of them
-    # fail with "Could not find old_string", and the model summarises
-    # the turn claiming every file was edited.  The user then has to
-    # manually run ``git status`` to catch the lie.  With this footer
-    # the truth is surfaced on every turn, so over-claiming is
-    # structurally impossible past the model.
-    #
-    # Gate: only applied when a real text response exists for this
-    # turn and the user didn't interrupt.  Empty/interrupted turns
-    # already have other surface text that shouldn't be augmented.
-    if final_response and not interrupted:
-        try:
-            _failed = getattr(agent, "_turn_failed_file_mutations", None) or {}
-            if _failed and agent._file_mutation_verifier_enabled():
-                footer = agent._format_file_mutation_failure_footer(_failed)
-                if footer:
-                    final_response = final_response.rstrip() + "\n\n" + footer
-        except Exception as _ver_err:
-            logger.debug("file-mutation verifier footer failed: %s", _ver_err)
-
-    # Turn-completion explainer.
-    # When a turn ends abnormally after substantive work — empty content
-    # after retries, a partial/truncated stream, a still-pending tool
-    # result, or an iteration/budget limit — the user otherwise gets a
-    # blank or fragmentary response box with no consolidated reason why
-    # the agent stopped (#34452).  Surface a single user-visible
-    # explanation derived from ``_turn_exit_reason``, mirroring the
-    # file-mutation verifier footer pattern above.
-    #
-    # Gate carefully so healthy turns stay quiet:
-    #   - ``text_response(...)`` exits never produce an explanation
-    #     (handled inside the formatter), so a terse ``Done.`` is silent.
-    #   - We only ACT when there is no genuinely usable reply this turn:
-    #     an empty response, the "(empty)" terminal sentinel, or a
-    #     suspiciously short partial fragment with no terminating
-    #     punctuation (e.g. "The").  A real short answer keeps its text.
-    if not interrupted:
-        try:
-            if agent._turn_completion_explainer_enabled():
-                _stripped = (final_response or "").strip()
-                _is_empty_terminal = _stripped == "" or _stripped == "(empty)"
-                # A short fragment that is not a normal text_response exit
-                # and lacks sentence-ending punctuation is treated as a
-                # truncated partial (the "The" case from #34452).
-                _is_partial_fragment = (
-                    not _is_empty_terminal
-                    and not str(_turn_exit_reason).startswith("text_response")
-                    and len(_stripped) <= 24
-                    and _stripped[-1:] not in {".", "!", "?", "。", "！", "？", "`", ")"}
-                )
-                if _is_empty_terminal or _is_partial_fragment:
-                    _explanation = agent._format_turn_completion_explanation(
-                        _turn_exit_reason
-                    )
-                    if _explanation:
-                        if _is_empty_terminal:
-                            # Replace the bare "(empty)"/blank sentinel with
-                            # the actionable explanation.
-                            final_response = _explanation
-                        else:
-                            # Keep the partial fragment, append the reason so
-                            # the user sees both what arrived and why it
-                            # stopped.
-                            final_response = (
-                                _stripped + "\n\n" + _explanation
-                            )
-        except Exception as _exp_err:
-            logger.debug("turn-completion explainer failed: %s", _exp_err)
-
-    _response_transformed = False
-
-    # Plugin hook: transform_llm_output
-    # Fired once per turn after the tool-calling loop completes.
-    # Plugins can transform the LLM's output text before it's returned.
-    # First hook to return a string wins; None/empty return leaves text unchanged.
-    if final_response and not interrupted:
-        try:
-            from hermes_cli.plugins import invoke_hook as _invoke_hook
-            _transform_results = _invoke_hook(
-                "transform_llm_output",
-                response_text=final_response,
-                session_id=agent.session_id or "",
-                model=agent.model,
-                platform=getattr(agent, "platform", None) or "",
-            )
-            for _hook_result in _transform_results:
-                if isinstance(_hook_result, str) and _hook_result:
-                    final_response = _hook_result
-                    _response_transformed = True
-                    break  # First non-empty string wins
-        except Exception as exc:
-            logger.warning("transform_llm_output hook failed: %s", exc)
-
-    # Plugin hook: post_llm_call
-    # Fired once per turn after the tool-calling loop completes.
-    # Plugins can use this to persist conversation data (e.g. sync
-    # to an external memory system).
-    if final_response and not interrupted:
-        try:
-            from hermes_cli.plugins import invoke_hook as _invoke_hook
-            _invoke_hook(
-                "post_llm_call",
-                session_id=agent.session_id,
-                task_id=effective_task_id,
-                turn_id=turn_id,
-                user_message=original_user_message,
-                assistant_response=final_response,
-                conversation_history=list(messages),
-                model=agent.model,
-                platform=getattr(agent, "platform", None) or "",
-            )
-        except Exception as exc:
-            logger.warning("post_llm_call hook failed: %s", exc)
-
-    # Extract reasoning from the CURRENT turn only.  Walk backwards
-    # but stop at the user message that started this turn — anything
-    # earlier is from a prior turn and must not leak into the reasoning
-    # box (confusing stale display; #17055).  Within the current turn
-    # we still want the *most recent* non-empty reasoning: many
-    # providers (Claude thinking, DeepSeek v4, Codex Responses) emit
-    # reasoning on the tool-call step and leave the final-answer step
-    # with reasoning=None, so picking only the last assistant would
-    # silently drop legitimate same-turn reasoning.
-    last_reasoning = None
-    for msg in reversed(messages):
-        if msg.get("role") == "user":
-            break  # turn boundary — don't cross into prior turns
-        if msg.get("role") == "assistant" and msg.get("reasoning"):
-            last_reasoning = msg["reasoning"]
-            break
-
-    # Build result with interrupt info if applicable
-    result = {
-        "final_response": final_response,
-        "last_reasoning": last_reasoning,
-        "messages": messages,
-        "api_calls": api_call_count,
-        "completed": completed,
-        "turn_exit_reason": _turn_exit_reason,
-        "failed": failed,
-        "partial": False,  # True only when stopped due to invalid tool calls
-        "interrupted": interrupted,
-        "response_transformed": _response_transformed,
-        "response_previewed": getattr(agent, "_response_was_previewed", False),
-        "model": agent.model,
-        "provider": agent.provider,
-        "base_url": agent.base_url,
-        "input_tokens": agent.session_input_tokens,
-        "output_tokens": agent.session_output_tokens,
-        "cache_read_tokens": agent.session_cache_read_tokens,
-        "cache_write_tokens": agent.session_cache_write_tokens,
-        "reasoning_tokens": agent.session_reasoning_tokens,
-        "prompt_tokens": agent.session_prompt_tokens,
-        "completion_tokens": agent.session_completion_tokens,
-        "total_tokens": agent.session_total_tokens,
-        "last_prompt_tokens": getattr(agent.context_compressor, "last_prompt_tokens", 0) or 0,
-        "estimated_cost_usd": agent.session_estimated_cost_usd,
-        "cost_status": agent.session_cost_status,
-        "cost_source": agent.session_cost_source,
-        "session_id": agent.session_id,
-    }
-    if agent._tool_guardrail_halt_decision is not None:
-        result["guardrail"] = agent._tool_guardrail_halt_decision.to_metadata()
-    # If a /steer landed after the final assistant turn (no more tool
-    # batches to drain into), hand it back to the caller so it can be
-    # delivered as the next user turn instead of being silently lost.
-    _leftover_steer = agent._drain_pending_steer()
-    if _leftover_steer:
-        result["pending_steer"] = _leftover_steer
-    agent._response_was_previewed = False
-
-    # Include interrupt message if one triggered the interrupt
-    if interrupted and agent._interrupt_message:
-        result["interrupt_message"] = agent._interrupt_message
-
-    # Clear interrupt state after handling
-    agent.clear_interrupt()
-
-    # Clear stream callback so it doesn't leak into future calls
-    agent._stream_callback = None
-
-    # Check skill trigger NOW — based on how many tool iterations THIS turn used.
-    _should_review_skills = False
-    if (agent._skill_nudge_interval > 0
-            and agent._iters_since_skill >= agent._skill_nudge_interval
-            and "skill_manage" in agent.valid_tool_names):
-        _should_review_skills = True
-        agent._iters_since_skill = 0
-
-    # External memory provider: sync the completed turn + queue next prefetch.
-    agent._sync_external_memory_for_turn(
-        original_user_message=original_user_message,
-        final_response=final_response,
-        interrupted=interrupted,
-        messages=messages,
-    )
-
-    # Background memory/skill review — runs AFTER the response is delivered
-    # so it never competes with the user's task for model attention.
-    if final_response and not interrupted and (_should_review_memory or _should_review_skills):
-        try:
-            agent._spawn_background_review(
-                messages_snapshot=list(messages),
-                review_memory=_should_review_memory,
-                review_skills=_should_review_skills,
-            )
-        except Exception:
-            pass  # Background review is best-effort
-
-    # Note: Memory provider on_session_end() + shutdown_all() are NOT
-    # called here — run_conversation() is called once per user message in
-    # multi-turn sessions. Shutting down after every turn would kill the
-    # provider before the second message. Actual session-end cleanup is
-    # handled by the CLI (atexit / /reset) and gateway (session expiry /
-    # _reset_session).
-
-    # Plugin hook: on_session_end
-    # Fired at the very end of every run_conversation call.
-    # Plugins can use this for cleanup, flushing buffers, etc.
-    try:
-        from hermes_cli.plugins import invoke_hook as _invoke_hook
-        _invoke_hook(
-            "on_session_end",
-            session_id=agent.session_id,
-            task_id=effective_task_id,
-            turn_id=turn_id,
-            completed=completed,
-            interrupted=interrupted,
-            model=agent.model,
-            platform=getattr(agent, "platform", None) or "",
-        )
-    except Exception as exc:
-        logger.warning("on_session_end hook failed: %s", exc)
-
-    return result

agent/turn_retry_state.py

@@ -1,68 +0,0 @@
-"""Per-attempt recovery bookkeeping for the conversation turn loop.
-
-The inner retry loop in ``run_conversation`` (``while retry_count <
-max_retries``) makes several distinct recovery attempts on a single model API
-call: a credential-pool 429 retry, a per-provider OAuth refresh (codex,
-anthropic, nous, copilot), a long-context compression restart, a length-
-continuation restart, and a handful of format-recovery branches (thinking-
-signature stripping, multimodal-tool-content stripping, llama.cpp grammar
-fallback, image shrink, invalid-encrypted-content, 1M-beta header).
-
-Each of those branches is guarded by a one-shot boolean so it fires at most
-once per attempt. They used to be ~16 bare ``*_attempted`` / ``has_retried_*``
-/ ``restart_with_*`` locals declared inline before the loop and threaded
-through its 2,400-line body. ``TurnRetryState`` collapses them into one object
-the loop mutates in place (``state.codex_auth_retry_attempted = True``), giving
-the recovery bookkeeping a single named, testable home.
-
-Loop-control variables (``retry_count``, ``max_retries``,
-``max_compression_attempts``) intentionally stay as plain locals — they are the
-``while`` mechanics, not recovery bookkeeping, and putting them on the object
-would add indirection without clarifying anything.
-
-This module is dependency-free so it can be unit-tested in isolation and
-imported by the turn loop without an import cycle.
-"""
-
-from __future__ import annotations
-
-from dataclasses import dataclass, fields
-
-
-@dataclass
-class TurnRetryState:
-    """One-shot recovery guards + restart signals for a single API-call attempt.
-
-    A fresh instance is created for each iteration of the outer turn loop
-    (once per ``api_call_count``). Each guard fires its recovery branch at most
-    once; the ``restart_with_*`` signals are read by the loop after the attempt
-    to decide whether to rebuild the request and retry.
-    """
-
-    # ── Per-provider OAuth / credential refresh guards ───────────────────
-    codex_auth_retry_attempted: bool = False
-    anthropic_auth_retry_attempted: bool = False
-    nous_auth_retry_attempted: bool = False
-    nous_paid_entitlement_refresh_attempted: bool = False
-    copilot_auth_retry_attempted: bool = False
-
-    # ── Format / payload recovery guards ─────────────────────────────────
-    thinking_sig_retry_attempted: bool = False
-    invalid_encrypted_content_retry_attempted: bool = False
-    image_shrink_retry_attempted: bool = False
-    multimodal_tool_content_retry_attempted: bool = False
-    oauth_1m_beta_retry_attempted: bool = False
-    llama_cpp_grammar_retry_attempted: bool = False
-
-    # ── Transport / rate-limit recovery ──────────────────────────────────
-    primary_recovery_attempted: bool = False
-    has_retried_429: bool = False
-
-    # ── Restart signals (read by the outer loop after the attempt) ───────
-    restart_with_compressed_messages: bool = False
-    restart_with_length_continuation: bool = False
-
-    def __iter__(self):
-        # Convenience for debugging / tests: iterate (name, value) pairs.
-        for f in fields(self):
-            yield f.name, getattr(self, f.name)

agent/usage_pricing.py

@@ -13,7 +13,6 @@ DEFAULT_PRICING = {"input": 0.0, "output": 0.0}
 
 _ZERO = Decimal("0")
 _ONE_MILLION = Decimal("1000000")
-_NOUS_DEFAULT_BASE_URL = "https://inference-api.nousresearch.com/v1"
 
 CostStatus = Literal["actual", "estimated", "included", "unknown"]
 CostSource = Literal[
@@ -571,8 +570,6 @@ def resolve_billing_route(
         return BillingRoute(provider="openai-codex", model=model, base_url=base_url or "", billing_mode="subscription_included")
     if provider_name == "openrouter" or base_url_host_matches(base_url or "", "openrouter.ai"):
         return BillingRoute(provider="openrouter", model=model, base_url=base_url or "", billing_mode="official_models_api")
-    if provider_name == "nous" or base_url_host_matches(base_url or "", "inference-api.nousresearch.com"):
-        return BillingRoute(provider="nous", model=model, base_url=base_url or _NOUS_DEFAULT_BASE_URL, billing_mode="official_models_api")
     if provider_name == "anthropic":
         return BillingRoute(provider="anthropic", model=model.split("/")[-1], base_url=base_url or "", billing_mode="official_docs_snapshot")
     if provider_name == "openai":

apps/bootstrap-installer/package.json

@@ -11,8 +11,7 @@
     "tauri": "tauri",
     "tauri:dev": "tauri dev",
     "tauri:build": "tauri build",
-    "tauri:build:debug": "tauri build --debug",
-    "typecheck": "tsc -p . --noEmit"
+    "tauri:build:debug": "tauri build --debug"
   },
   "dependencies": {
     "@nous-research/ui": "0.16.0",
@@ -41,7 +40,7 @@
     "@types/react": "^19.2.14",
     "@types/react-dom": "^19.2.3",
     "@vitejs/plugin-react": "^5.2.0",
-    "typescript": "^6.0.3",
+    "typescript": "~5.9.3",
     "vite": "^7.3.1"
   }
 }

apps/bootstrap-installer/tsconfig.json

@@ -16,8 +16,9 @@
     "noUnusedParameters": true,
     "esModuleInterop": true,
     "noFallthroughCasesInSwitch": true,
+    "baseUrl": ".",
     "paths": {
-      "@/*": ["./src/*"]
+      "@/*": ["src/*"]
     }
   },
   "include": ["src"],

apps/desktop/README.md

@@ -93,7 +93,7 @@ Run before opening a PR (lint may surface pre-existing warnings but must exit cl
 
 ```bash
 npm run fix
-npm run typecheck
+npm run type-check
 npm run lint
 npm run test:desktop:all
 ```

apps/desktop/electron/bootstrap-runner.cjs

@@ -40,15 +40,6 @@ const path = require('node:path')
 const https = require('node:https')
 const { spawn } = require('node:child_process')
 
-const IS_WINDOWS = process.platform === 'win32'
-
-function hiddenWindowsChildOptions(options = {}) {
-  if (!IS_WINDOWS || Object.prototype.hasOwnProperty.call(options, 'windowsHide')) {
-    return options
-  }
-  return { ...options, windowsHide: true }
-}
-
 const STAMP_COMMIT_RE = /^[0-9a-f]{7,40}$/i
 
 // Stages flagged needs_user_input=true in the manifest are skipped by the
@@ -293,7 +284,7 @@ function spawnPowerShell(scriptPath, args, { emit, stageName, abortSignal, herme
     const ps = process.platform === 'win32' ? resolveWindowsPowerShell() : 'pwsh'
     const fullArgs = ['-NoProfile', '-ExecutionPolicy', 'Bypass', '-File', scriptPath, ...args]
 
-    const child = spawn(ps, fullArgs, hiddenWindowsChildOptions({
+    const child = spawn(ps, fullArgs, {
       stdio: ['ignore', 'pipe', 'pipe'],
       env: {
         ...process.env,
@@ -301,7 +292,7 @@ function spawnPowerShell(scriptPath, args, { emit, stageName, abortSignal, herme
         // choice rather than re-computing the default.
         HERMES_HOME: hermesHome || process.env.HERMES_HOME || ''
       }
-    }))
+    })
 
     let stdout = ''
     let stderr = ''

apps/desktop/electron/main.cjs

@@ -26,15 +26,9 @@ const { fileURLToPath, 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 } = require('./session-windows.cjs')
 const { canImportHermesCli, verifyHermesCli } = require('./backend-probes.cjs')
 const { probeGatewayWebSocket } = require('./gateway-ws-probe.cjs')
 const { serializeJsonBody, setJsonRequestHeaders } = require('./oauth-net-request.cjs')
-const { fetchMarketplaceThemes, searchMarketplaceThemes } = require('./vscode-marketplace.cjs')
-const {
-  OFFICIAL_REPO_HTTPS_URL,
-  isOfficialSshRemote
-} = require('./update-remote.cjs')
 const {
   buildPosixCleanupScript,
   buildWindowsCleanupScript,
@@ -44,7 +38,6 @@ const {
   shouldRemoveAppBundle,
   uninstallArgsForMode
 } = require('./desktop-uninstall.cjs')
-const { isPackagedInstallPath: isPackagedInstallPathUnderRoots } = require('./workspace-cwd.cjs')
 const {
   authModeFromStatus,
   buildGatewayWsUrl,
@@ -69,11 +62,9 @@ const {
 } = require('./hardening.cjs')
 
 let nodePty = null
-let nodePtyDir = null
 
 try {
   nodePty = require('node-pty')
-  nodePtyDir = path.dirname(require.resolve('node-pty/package.json'))
 } catch {
   // Packaged builds set `files:` in package.json, which excludes node_modules
   // from the asar.  Workspace dedup also hoists this native dep to the repo
@@ -86,12 +77,10 @@ try {
     const path = require('node:path')
     const resourcesPath = process.resourcesPath
     if (resourcesPath) {
-      nodePtyDir = path.join(resourcesPath, 'native-deps', 'node-pty')
-      nodePty = require(nodePtyDir)
+      nodePty = require(path.join(resourcesPath, 'native-deps', 'node-pty'))
     }
   } catch {
     nodePty = null
-    nodePtyDir = null
   }
 }
 
@@ -111,13 +100,6 @@ const IS_WINDOWS = process.platform === 'win32'
 const IS_WSL = isWslEnvironment()
 const APP_ROOT = app.getAppPath()
 
-function hiddenWindowsChildOptions(options = {}) {
-  if (!IS_WINDOWS || Object.prototype.hasOwnProperty.call(options, 'windowsHide')) {
-    return options
-  }
-  return { ...options, windowsHide: true }
-}
-
 // Remote displays (SSH X11 forwarding, VNC, RDP) make Chromium's GPU
 // compositor flicker — accelerated layers can't be presented cleanly over the
 // wire, so the window flashes during scroll/streaming/animation. Local
@@ -137,20 +119,6 @@ if (REMOTE_DISPLAY_REASON) {
     `[hermes] remote display detected (${REMOTE_DISPLAY_REASON}); disabling GPU hardware acceleration to prevent flicker`
   )
 }
-
-// Keep the renderer running at full speed while the window is in the background
-// or occluded. The chat transcript streams to screen through a
-// requestAnimationFrame-gated flush; Chromium pauses rAF (and clamps timers)
-// for backgrounded/occluded renderers, so without these the live answer stalls
-// whenever the window loses focus (switching to your editor mid-turn, detached
-// devtools, another window covering it) and only paints on refocus or refresh.
-// `backgroundThrottling: false` on the BrowserWindow covers the blurred case;
-// these process-level switches additionally stop Chromium from backgrounding or
-// occlusion-throttling the renderer. Must run before app `ready`.
-app.commandLine.appendSwitch('disable-renderer-backgrounding')
-app.commandLine.appendSwitch('disable-backgrounding-occluded-windows')
-app.commandLine.appendSwitch('disable-background-timer-throttling')
-
 const SOURCE_REPO_ROOT = path.resolve(APP_ROOT, '../..')
 
 // Build-time install stamp -- the git ref this .exe was built against.
@@ -1117,7 +1085,7 @@ function findSystemPython() {
         const out = execFileSync(
           'reg',
           ['query', `${hive}\\SOFTWARE\\Python\\PythonCore\\${version}\\InstallPath`, '/ve', '/reg:64'],
-          hiddenWindowsChildOptions({ encoding: 'utf8', stdio: ['ignore', 'pipe', 'ignore'] })
+          { encoding: 'utf8', stdio: ['ignore', 'pipe', 'ignore'] }
         )
         // Output format: "    (Default)    REG_SZ    C:\Path\To\Python\"
         const match = out.match(/REG_SZ\s+(.+?)\s*$/m)
@@ -1153,10 +1121,10 @@ function findSystemPython() {
   if (pyExe) {
     for (const version of SUPPORTED_VERSIONS) {
       try {
-        const out = execFileSync(pyExe, [`-${version}`, '-c', 'import sys; print(sys.executable)'], hiddenWindowsChildOptions({
+        const out = execFileSync(pyExe, [`-${version}`, '-c', 'import sys; print(sys.executable)'], {
           encoding: 'utf8',
           stdio: ['ignore', 'pipe', 'ignore']
-        }))
+        })
         const candidate = out.trim()
         if (candidate && fileExists(candidate)) return candidate
       } catch {
@@ -1291,11 +1259,11 @@ function resolveUpdateRoot() {
 
 function runGit(args, options = {}) {
   return new Promise((resolve, reject) => {
-    const child = spawn(resolveGitBinary(), IS_WINDOWS ? ['-c', 'windows.appendAtomically=false', ...args] : args, hiddenWindowsChildOptions({
+    const child = spawn(resolveGitBinary(), IS_WINDOWS ? ['-c', 'windows.appendAtomically=false', ...args] : args, {
       cwd: options.cwd,
       env: { ...process.env, ...(options.env || {}), GIT_TERMINAL_PROMPT: '0' },
       stdio: ['ignore', 'pipe', 'pipe']
-    }))
+    })
 
     let stdout = ''
     let stderr = ''
@@ -1316,11 +1284,6 @@ function runGit(args, options = {}) {
 
 const firstLine = text => (text || '').split('\n').find(Boolean) || ''
 
-async function getOriginUrl(updateRoot) {
-  const origin = await runGit(['remote', 'get-url', 'origin'], { cwd: updateRoot })
-  return origin.code === 0 ? origin.stdout.trim() : ''
-}
-
 function emitUpdateProgress(payload) {
   const merged = { stage: 'idle', message: '', percent: null, error: null, ...payload, at: Date.now() }
   rememberLog(`[updates] ${merged.stage}: ${merged.message || merged.error || ''}`)
@@ -1340,9 +1303,7 @@ async function resolveHealedBranch(updateRoot, branch) {
     return branch || 'main'
   }
 
-  const originUrl = await getOriginUrl(updateRoot)
-  const remote = isOfficialSshRemote(originUrl) ? OFFICIAL_REPO_HTTPS_URL : 'origin'
-  const probe = await runGit(['ls-remote', '--exit-code', '--heads', remote, branch], { cwd: updateRoot })
+  const probe = await runGit(['ls-remote', '--exit-code', '--heads', 'origin', branch], { cwd: updateRoot })
   if (probe.code !== 2) {
     return branch
   }
@@ -1370,40 +1331,6 @@ async function checkUpdates() {
   }
 
   branch = await resolveHealedBranch(updateRoot, branch)
-  const originUrl = await getOriginUrl(updateRoot)
-  if (isOfficialSshRemote(originUrl)) {
-    const git = args => runGit(args, { cwd: updateRoot }).then(r => r.stdout.trim())
-    const [currentSha, target, dirtyStr, currentBranch] = await Promise.all([
-      git(['rev-parse', 'HEAD']),
-      runGit(['ls-remote', OFFICIAL_REPO_HTTPS_URL, `refs/heads/${branch}`], { cwd: updateRoot }),
-      git(['status', '--porcelain']),
-      git(['rev-parse', '--abbrev-ref', 'HEAD'])
-    ])
-    const targetSha = firstLine(target.stdout).split(/\s+/)[0] || ''
-    if (target.code !== 0 || !targetSha) {
-      return {
-        supported: true,
-        branch,
-        error: 'fetch-failed',
-        message: firstLine(target.stderr) || 'git ls-remote failed.',
-        hermesRoot: updateRoot,
-        fetchedAt: Date.now()
-      }
-    }
-    return {
-      supported: true,
-      branch,
-      currentBranch,
-      behind: currentSha && currentSha === targetSha ? 0 : 1,
-      currentSha,
-      targetSha,
-      commits: [],
-      dirty: dirtyStr.length > 0,
-      hermesRoot: updateRoot,
-      fetchedAt: Date.now()
-    }
-  }
-
   const fetched = await runGit(['fetch', '--quiet', 'origin', branch], { cwd: updateRoot })
   if (fetched.code !== 0) {
     return {
@@ -1546,7 +1473,7 @@ function forceKillProcessTree(pid) {
   if (!IS_WINDOWS) return
   if (!Number.isInteger(pid) || pid <= 0) return
   try {
-    execFileSync('taskkill', ['/PID', String(pid), '/T', '/F'], hiddenWindowsChildOptions({ stdio: 'ignore' }))
+    execFileSync('taskkill', ['/PID', String(pid), '/T', '/F'], { stdio: 'ignore' })
   } catch {
     // Already gone, or no permission — best effort; the unlock wait below is
     // the real gate.
@@ -1732,11 +1659,11 @@ function runStreamedUpdate(command, args, { cwd, env, stage } = {}) {
   return new Promise(resolve => {
     let child
     try {
-      child = spawn(command, args, hiddenWindowsChildOptions({
+      child = spawn(command, args, {
         cwd,
         env: { ...process.env, ...(env || {}) },
         stdio: ['ignore', 'pipe', 'pipe']
-      }))
+      })
     } catch (err) {
       resolve({ code: 1, error: err.message })
       return
@@ -1975,51 +1902,12 @@ function resolveWebDist() {
   const unpackedDist = path.join(unpackedPathFor(APP_ROOT), 'dist')
   if (directoryExists(unpackedDist)) return unpackedDist
 
-  // Final fallback: APP_ROOT/dist. When packaged with asar:true this lives
-  // INSIDE app.asar — not a servable filesystem directory — so the embedded
-  // dashboard backend 404s on static routes (see #41327, #39472). The durable
-  // fix is unpacking dist/ (PR #41411 adds dist/** to asarUnpack so the tier-2
-  // unpackedDist above resolves). If we still land here while packaged, log it
-  // so the cause isn't silent.
-  const fallback = path.join(APP_ROOT, 'dist')
-  if (IS_PACKAGED && /app\.asar(?=$|[\\/])/.test(fallback) && !directoryExists(fallback)) {
-    rememberLog(
-      `[web-dist] dashboard frontend dir resolved to an asar-internal path that ` +
-        `is not a real directory: ${fallback}. Static routes will 404. ` +
-        `Ensure dist/** is unpacked (asarUnpack) or set HERMES_DESKTOP_WEB_DIST.`
-    )
-  }
-  return fallback
+  return path.join(APP_ROOT, 'dist')
 }
 
 function resolveRendererIndex() {
   const candidates = [path.join(APP_ROOT, 'dist', 'index.html'), path.join(resolveWebDist(), 'index.html')]
-  const found = candidates.find(fileExists)
-  if (found) return found
-  // Nothing on disk. A packaged build with no renderer bundle blank-pages with
-  // a bare ERR_FILE_NOT_FOUND and no clue why (see #39484). Surface the cause
-  // and the fix before Electron loads the missing file.
-  rememberLog(
-    `[renderer] index.html not found — the desktop app was packaged without a ` +
-      `renderer bundle. Tried: ${candidates.join(', ')}. ` +
-      `Rebuild with: hermes desktop --force-build`
-  )
-  return candidates[0]
-}
-
-// True when `dir` lives inside the packaged app bundle / install tree.
-// Packaged Electron's process.cwd() (and npm's INIT_CWD when dev tooling
-// leaked into a release build) often resolve here — e.g. win-unpacked on
-// Windows — which is exactly where PR #37536 item 16 said we must NOT run.
-function isPackagedInstallPath(dir) {
-  return isPackagedInstallPathUnderRoots(dir, {
-    isPackaged: IS_PACKAGED,
-    installRoots: [
-      APP_ROOT,
-      path.dirname(process.execPath),
-      resolveRemovableAppPath(process.execPath, process.platform, process.env)
-    ]
-  })
+  return candidates.find(fileExists) || candidates[0]
 }
 
 function resolveHermesCwd() {
@@ -2033,7 +1921,7 @@ function resolveHermesCwd() {
   const candidates = [
     readDefaultProjectDir(),
     process.env.HERMES_DESKTOP_CWD,
-    IS_PACKAGED ? null : process.env.INIT_CWD,
+    process.env.INIT_CWD,
     IS_PACKAGED ? null : process.cwd(),
     !IS_PACKAGED ? SOURCE_REPO_ROOT : null,
     app.getPath('home')
@@ -2042,37 +1930,12 @@ function resolveHermesCwd() {
   for (const candidate of candidates) {
     if (!candidate) continue
     const resolved = path.resolve(String(candidate))
-
-    if (isPackagedInstallPath(resolved)) {
-      continue
-    }
-
     if (directoryExists(resolved)) return resolved
   }
 
   return app.getPath('home')
 }
 
-function sanitizeWorkspaceCwd(cwd) {
-  const trimmed = typeof cwd === 'string' ? cwd.trim() : ''
-
-  if (!trimmed || isPackagedInstallPath(trimmed)) {
-    return { cwd: resolveHermesCwd(), sanitized: Boolean(trimmed) }
-  }
-
-  try {
-    const resolved = path.resolve(trimmed)
-
-    if (directoryExists(resolved)) {
-      return { cwd: resolved, sanitized: false }
-    }
-  } catch {
-    // Fall through to the resolved default.
-  }
-
-  return { cwd: resolveHermesCwd(), sanitized: Boolean(trimmed) }
-}
-
 // Persisted "Default project directory" — surfaced as a setting in the
 // renderer (see app/settings/sessions-settings.tsx). Stored as JSON in
 // userData so it survives self-updates without bleeding into the new
@@ -2723,7 +2586,7 @@ function fetchHtmlTitleWithCurl(rawUrl) {
       '--raw',
       url
     ]
-    const child = spawn('curl', args, hiddenWindowsChildOptions({ stdio: ['ignore', 'pipe', 'ignore'] }))
+    const child = spawn('curl', args, { stdio: ['ignore', 'pipe', 'ignore'] })
     const chunks = []
     let bytes = 0
 
@@ -3274,7 +3137,7 @@ function buildApplicationMenu() {
         label: 'Actual Size',
         accelerator: 'CommandOrControl+0',
         click: () => {
-          setAndPersistZoomLevel(mainWindow, 0)
+          if (mainWindow && !mainWindow.isDestroyed()) mainWindow.webContents.setZoomLevel(0)
         }
       },
       {
@@ -3282,7 +3145,8 @@ function buildApplicationMenu() {
         accelerator: 'CommandOrControl+Plus',
         click: () => {
           if (mainWindow && !mainWindow.isDestroyed()) {
-            setAndPersistZoomLevel(mainWindow, mainWindow.webContents.getZoomLevel() + 0.1)
+            const next = Math.min(mainWindow.webContents.getZoomLevel() + 0.1, 9)
+            mainWindow.webContents.setZoomLevel(next)
           }
         }
       },
@@ -3291,7 +3155,8 @@ function buildApplicationMenu() {
         accelerator: 'CommandOrControl+-',
         click: () => {
           if (mainWindow && !mainWindow.isDestroyed()) {
-            setAndPersistZoomLevel(mainWindow, mainWindow.webContents.getZoomLevel() - 0.1)
+            const next = Math.max(mainWindow.webContents.getZoomLevel() - 0.1, -9)
+            mainWindow.webContents.setZoomLevel(next)
           }
         }
       },
@@ -3353,42 +3218,6 @@ function installPreviewShortcut(window) {
   })
 }
 
-// Zoom level is persisted in the renderer's own localStorage (per-origin,
-// survives reloads/restarts) rather than a main-process JSON file. The main
-// process owns setZoomLevel, so we mirror each change into localStorage and
-// read it back on did-finish-load to re-apply after reloads or crash recovery.
-const ZOOM_STORAGE_KEY = 'hermes:desktop:zoomLevel'
-
-function clampZoomLevel(value) {
-  if (!Number.isFinite(value)) return 0
-  return Math.min(Math.max(value, -9), 9)
-}
-
-function setAndPersistZoomLevel(window, zoomLevel) {
-  if (!window || window.isDestroyed()) return
-  const next = clampZoomLevel(zoomLevel)
-  window.webContents.setZoomLevel(next)
-  window.webContents
-    .executeJavaScript(
-      `try { localStorage.setItem(${JSON.stringify(ZOOM_STORAGE_KEY)}, ${JSON.stringify(String(next))}) } catch {}`
-    )
-    .catch(error => rememberLog(`[zoom] persist failed: ${error?.message || error}`))
-}
-
-function restorePersistedZoomLevel(window) {
-  if (!window || window.isDestroyed()) return
-  window.webContents
-    .executeJavaScript(
-      `(() => { try { return localStorage.getItem(${JSON.stringify(ZOOM_STORAGE_KEY)}) } catch { return null } })()`
-    )
-    .then(stored => {
-      if (stored == null || !window || window.isDestroyed()) return
-      const level = clampZoomLevel(Number(stored))
-      window.webContents.setZoomLevel(level)
-    })
-    .catch(error => rememberLog(`[zoom] restore failed: ${error?.message || error}`))
-}
-
 function installZoomShortcuts(window) {
   // Override Ctrl/Cmd + +/-/0 with half the default zoom step (0.1 vs 0.2).
   // The menu items handle this on macOS (where the menu is always present),
@@ -3402,13 +3231,15 @@ function installZoomShortcuts(window) {
     const key = input.key
     if (key === '0') {
       event.preventDefault()
-      setAndPersistZoomLevel(window, 0)
+      window.webContents.setZoomLevel(0)
     } else if (key === '=' || key === '+') {
       event.preventDefault()
-      setAndPersistZoomLevel(window, window.webContents.getZoomLevel() + ZOOM_STEP)
+      const next = Math.min(window.webContents.getZoomLevel() + ZOOM_STEP, 9)
+      window.webContents.setZoomLevel(next)
     } else if (key === '-') {
       event.preventDefault()
-      setAndPersistZoomLevel(window, window.webContents.getZoomLevel() - ZOOM_STEP)
+      const next = Math.max(window.webContents.getZoomLevel() - ZOOM_STEP, -9)
+      window.webContents.setZoomLevel(next)
     }
   })
 }
@@ -4016,12 +3847,10 @@ async function sanitizeDesktopConnectionConfig(config = readDesktopConnectionCon
   const scoped = key ? config.profiles?.[key] || null : null
   const block = key ? scoped || {} : config.remote || {}
 
-  const envOverride = key ? false : Boolean(process.env.HERMES_DESKTOP_REMOTE_URL)
-
   const remoteToken = decryptDesktopSecret(block.token)
   const authMode = normAuthMode(block.authMode)
-  const remoteUrl = envOverride ? String(process.env.HERMES_DESKTOP_REMOTE_URL || '') : String(block.url || '')
-  const mode = envOverride || (key ? scoped?.mode : config.mode) === 'remote' ? 'remote' : 'local'
+  const remoteUrl = String(block.url || '')
+  const mode = (key ? scoped?.mode : config.mode) === 'remote' ? 'remote' : 'local'
 
   let remoteOauthConnected = false
   if (authMode === 'oauth' && remoteUrl) {
@@ -4047,7 +3876,7 @@ async function sanitizeDesktopConnectionConfig(config = readDesktopConnectionCon
     remoteTokenSet: Boolean(remoteToken),
     // The env override only forces the global/primary connection; a per-profile
     // scope is never overridden by HERMES_DESKTOP_REMOTE_URL.
-    envOverride
+    envOverride: key ? false : Boolean(process.env.HERMES_DESKTOP_REMOTE_URL)
   }
 }
 
@@ -4239,7 +4068,9 @@ async function requestJsonForProfile(profile, path, method, body) {
   const conn = await ensureBackend(profile)
   const url = `${conn.baseUrl}${path}`
   const opts = { method, body, timeoutMs: DEFAULT_FETCH_TIMEOUT_MS }
-  return conn.authMode === 'oauth' ? fetchJsonViaOauthSession(url, opts) : fetchJson(url, conn.token, opts)
+  return conn.authMode === 'oauth'
+    ? fetchJsonViaOauthSession(url, opts)
+    : fetchJson(url, conn.token, opts)
 }
 
 async function probeRemoteAuthMode(rawUrl) {
@@ -4313,8 +4144,7 @@ async function testDesktopConnectionConfig(input = {}) {
   // The block under test: a per-profile entry or the global remote. Coerce has
   // already normalized the URL and resolved token inheritance for the scope.
   const block = key ? config.profiles?.[key] || null : config.remote
-  const wantRemote =
-    block?.mode === 'remote' || (!key && config.mode === 'remote') || (input.mode === 'remote' && block)
+  const wantRemote = block?.mode === 'remote' || (!key && config.mode === 'remote') || (input.mode === 'remote' && block)
   // ``/api/status`` is public on every gateway (no creds needed), so a
   // reachability test works for local, token, and oauth modes alike — we only
   // need a base URL. For a remote config we normalize the URL from the input;
@@ -4397,31 +4227,20 @@ async function teardownPrimaryBackendAndWait() {
   const dying = hermesProcess && !hermesProcess.killed ? hermesProcess : null
   resetHermesConnection()
 
-  await waitForBackendExit(dying)
-}
-
-async function waitForBackendExit(child, timeoutMs = 5000) {
-  if (!child) {
-    return
-  }
-  if (child.exitCode !== null || child.signalCode !== null) {
+  if (!dying) {
     return
   }
 
   await new Promise(resolve => {
     const timer = setTimeout(() => {
       try {
-        if (IS_WINDOWS && Number.isInteger(child.pid)) {
-          forceKillProcessTree(child.pid)
-        } else {
-          child.kill('SIGKILL')
-        }
+        dying.kill('SIGKILL')
       } catch {
         // Already gone.
       }
       resolve()
-    }, timeoutMs)
-    child.once('exit', () => {
+    }, 5000)
+    dying.once('exit', () => {
       clearTimeout(timer)
       resolve()
     })
@@ -4543,16 +4362,12 @@ async function spawnPoolBackend(profile, entry) {
 
   rememberLog(`Starting Hermes backend for profile "${profile}" via ${backend.label}`)
 
-  const child = spawn(backend.command, backend.args, hiddenWindowsChildOptions({
+  const child = spawn(backend.command, backend.args, {
     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).
@@ -4561,7 +4376,7 @@ async function spawnPoolBackend(profile, entry) {
     },
     shell: backend.shell,
     stdio: ['ignore', 'pipe', 'pipe']
-  }))
+  })
   entry.process = child
   entry.port = port
   entry.token = token
@@ -4583,9 +4398,7 @@ async function spawnPoolBackend(profile, entry) {
     rememberLog(`Hermes backend for profile "${profile}" exited (${signal || code})`)
     backendPool.delete(profile)
     if (!ready) {
-      rejectStart?.(
-        new Error(`Hermes backend for profile "${profile}" exited before it became ready (${signal || code}).`)
-      )
+      rejectStart?.(new Error(`Hermes backend for profile "${profile}" exited before it became ready (${signal || code}).`))
     }
   })
 
@@ -4619,70 +4432,12 @@ function stopPoolBackend(profile) {
   }
 }
 
-async function teardownPoolBackendAndWait(profile) {
-  const entry = backendPool.get(profile)
-  if (!entry) return
-  backendPool.delete(profile)
-
-  if (entry.process && !entry.process.killed) {
-    try {
-      entry.process.kill('SIGTERM')
-    } catch {
-      // Already gone.
-    }
-  }
-
-  await waitForBackendExit(entry.process)
-}
-
 function stopAllPoolBackends() {
   for (const profile of [...backendPool.keys()]) {
     stopPoolBackend(profile)
   }
 }
 
-function profileNameFromDeleteRequest(request) {
-  if (!request || String(request.method || 'GET').toUpperCase() !== 'DELETE') {
-    return null
-  }
-
-  const match = String(request.path || '').match(/^\/api\/profiles\/([^/?#]+)(?:[?#].*)?$/)
-  if (!match) {
-    return null
-  }
-
-  let raw = ''
-  try {
-    raw = decodeURIComponent(match[1])
-  } catch {
-    return null
-  }
-
-  const name = raw.trim()
-  if (!name) {
-    return null
-  }
-  if (name.toLowerCase() === 'default') {
-    return 'default'
-  }
-  return name.toLowerCase()
-}
-
-async function prepareProfileDeleteRequest(request) {
-  const profile = profileNameFromDeleteRequest(request)
-  if (!profile || profile === 'default' || !PROFILE_NAME_RE.test(profile)) {
-    return
-  }
-
-  if (profile === primaryProfileKey()) {
-    writeActiveDesktopProfile('default')
-    await teardownPrimaryBackendAndWait()
-    return
-  }
-
-  await teardownPoolBackendAndWait(profile)
-}
-
 async function startHermes() {
   // Latched-failure short-circuit: once bootstrap has failed in this
   // process, every subsequent startHermes() call re-throws the same error
@@ -4743,7 +4498,7 @@ 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({
+    hermesProcess = spawn(backend.command, backend.args, {
       cwd: hermesCwd,
       env: {
         ...process.env,
@@ -4757,7 +4512,6 @@ async function startHermes() {
         // 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).
@@ -4766,7 +4520,7 @@ async function startHermes() {
       },
       shell: backend.shell,
       stdio: ['ignore', 'pipe', 'pipe']
-    }))
+    })
 
     hermesProcess.stdout.on('data', rememberLog)
     hermesProcess.stderr.on('data', rememberLog)
@@ -4855,100 +4609,12 @@ async function startHermes() {
   return connectionPromise
 }
 
-// Shared navigation guards + window chrome wiring applied to every window
-// (the primary plus any secondary session windows). Factored out of
-// createWindow() so secondary windows can't drift from the main window's
-// security posture: external links open in the OS browser, in-app navigation
-// stays confined to the dev server / packaged file URL, and the preview /
-// devtools / zoom / context-menu affordances behave identically everywhere.
-function wireCommonWindowHandlers(win) {
-  installPreviewShortcut(win)
-  installDevToolsShortcut(win)
-  installZoomShortcuts(win)
-  installContextMenu(win)
-  win.webContents.setWindowOpenHandler(details => {
-    openExternalUrl(details.url)
-
-    return { action: 'deny' }
-  })
-  win.webContents.on('will-navigate', (event, url) => {
-    if ((DEV_SERVER && url.startsWith(DEV_SERVER)) || (!DEV_SERVER && url.startsWith('file:'))) {
-      return
-    }
-
-    event.preventDefault()
-    openExternalUrl(url)
-  })
-}
-
-// Secondary "session windows" — one extra OS window per chat so a user can
-// work with multiple chats side by side. The registry guarantees one window
-// per sessionId (re-opening focuses the existing window) and self-cleans on
-// close. The primary mainWindow is never tracked here. Pure logic + the URL
-// builder live in session-windows.cjs so they stay unit-testable.
-const sessionWindows = createSessionWindowRegistry()
-
-function focusWindow(win) {
-  if (!win || win.isDestroyed()) return
-  if (win.isMinimized()) win.restore()
-  if (!win.isVisible()) win.show()
-  win.focus()
-}
-
-// Open (or focus) a standalone window for a single chat session.
-function createSessionWindow(sessionId) {
-  return sessionWindows.openOrFocus(sessionId, () => {
-    const icon = getAppIconPath()
-    const win = new BrowserWindow({
-      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,
-      icon,
-      backgroundColor: '#f7f7f7',
-      webPreferences: {
-        preload: path.join(__dirname, 'preload.cjs'),
-        contextIsolation: true,
-        webviewTag: true,
-        sandbox: true,
-        nodeIntegration: false,
-        devTools: true
-      }
-    })
-
-    if (IS_MAC) {
-      win.setWindowButtonPosition?.(WINDOW_BUTTON_POSITION)
-    }
-
-    win.on('will-enter-full-screen', () => sendWindowStateChanged(true))
-    win.on('enter-full-screen', () => sendWindowStateChanged(true))
-    win.on('will-leave-full-screen', () => sendWindowStateChanged(false))
-    win.on('leave-full-screen', () => sendWindowStateChanged(false))
-
-    wireCommonWindowHandlers(win)
-
-    win.loadURL(
-      buildSessionWindowUrl(sessionId, {
-        devServer: DEV_SERVER,
-        rendererIndexPath: DEV_SERVER ? undefined : resolveRendererIndex()
-      })
-    )
-
-    return win
-  })
-}
-
 function createWindow() {
   const icon = getAppIconPath()
   mainWindow = new BrowserWindow({
     width: 1220,
     height: 800,
-    minWidth: 400,
+    minWidth: 900,
     minHeight: 620,
     title: 'Hermes',
     // Frameless title bar on every platform so the renderer can paint the
@@ -4969,16 +4635,7 @@ function createWindow() {
       webviewTag: true,
       sandbox: true,
       nodeIntegration: false,
-      devTools: true,
-      // Keep timers + requestAnimationFrame running at full speed when the
-      // window is blurred/occluded. The chat transcript streams to the screen
-      // through a requestAnimationFrame-gated flush (useSessionStateCache),
-      // so with Chromium's default background throttling the live answer
-      // stalls whenever this window isn't focused (e.g. you switch to your
-      // editor mid-turn, or open detached devtools) and only appears once you
-      // refocus or refresh. A streaming chat app must render in the
-      // background, so opt out — matching the secondary windows above.
-      backgroundThrottling: false
+      devTools: true
     }
   })
 
@@ -5003,7 +4660,23 @@ function createWindow() {
   mainWindow.on('will-leave-full-screen', () => sendWindowStateChanged(false))
   mainWindow.on('leave-full-screen', () => sendWindowStateChanged(false))
 
-  wireCommonWindowHandlers(mainWindow)
+  installPreviewShortcut(mainWindow)
+  installDevToolsShortcut(mainWindow)
+  installZoomShortcuts(mainWindow)
+  installContextMenu(mainWindow)
+  mainWindow.webContents.setWindowOpenHandler(details => {
+    openExternalUrl(details.url)
+
+    return { action: 'deny' }
+  })
+  mainWindow.webContents.on('will-navigate', (event, url) => {
+    if ((DEV_SERVER && url.startsWith(DEV_SERVER)) || (!DEV_SERVER && url.startsWith('file:'))) {
+      return
+    }
+
+    event.preventDefault()
+    openExternalUrl(url)
+  })
 
   mainWindow.webContents.on('render-process-gone', (_event, details) => {
     rememberLog(`[renderer] render-process-gone reason=${details?.reason} exitCode=${details?.exitCode}`)
@@ -5057,7 +4730,6 @@ function createWindow() {
   }
 
   mainWindow.webContents.once('did-finish-load', () => {
-    restorePersistedZoomLevel(mainWindow)
     broadcastBootProgress()
     sendWindowStateChanged()
     startHermes().catch(error => rememberLog(error.stack || error.message))
@@ -5109,15 +4781,6 @@ 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) => {
-  if (typeof sessionId !== 'string' || !sessionId.trim()) {
-    return { ok: false, error: 'invalid-session-id' }
-  }
-
-  createSessionWindow(sessionId.trim())
-
-  return { ok: true }
-})
 ipcMain.handle('hermes:bootstrap:reset', async () => {
   // Renderer's "Reload and retry" path. Clear the latched failure and
   // reset connection state so the next startHermes() call restarts the
@@ -5356,19 +5019,17 @@ async function mergeRemoteProfileSessions(searchParams, remoteProfiles) {
   let total = (Number(base.total) || 0) - remoteProfiles.reduce((n, p) => n + (profileTotals[p] || 0), 0)
 
   // Swap each remote profile's stale local rows/total for the remote's real ones.
-  await Promise.all(
-    remoteProfiles.map(async name => {
-      const list = await remoteSessionList(name, remoteParams).catch(() => null)
-      if (!list) {
-        delete profileTotals[name] // dead remote → drop its stale local total too
-        return
-      }
-      const rows = rowsOf(list)
-      merged.push(...rows)
-      profileTotals[name] = Number(list.total) || rows.length
-      total += profileTotals[name]
-    })
-  )
+  await Promise.all(remoteProfiles.map(async name => {
+    const list = await remoteSessionList(name, remoteParams).catch(() => null)
+    if (!list) {
+      delete profileTotals[name] // dead remote → drop its stale local total too
+      return
+    }
+    const rows = rowsOf(list)
+    merged.push(...rows)
+    profileTotals[name] = Number(list.total) || rows.length
+    total += profileTotals[name]
+  }))
 
   const recency = s => s?.[order] ?? s?.started_at ?? 0
   merged.sort((a, b) => recency(b) - recency(a))
@@ -5385,8 +5046,6 @@ ipcMain.handle('hermes:api', async (_event, request) => {
     return rerouted
   }
 
-  await prepareProfileDeleteRequest(request)
-
   const connection = await ensureBackend(request?.profile)
   const timeoutMs = resolveTimeoutMs(request?.timeoutMs, DEFAULT_FETCH_TIMEOUT_MS)
   const url = `${connection.baseUrl}${request.path}`
@@ -5534,12 +5193,9 @@ ipcMain.handle('hermes:openExternal', (_event, url) => {
 // session spawn (no app restart needed).
 ipcMain.handle('hermes:setting:defaultProjectDir:get', async () => ({
   dir: readDefaultProjectDir(),
-  defaultLabel: app.getPath('home'),
-  resolvedCwd: resolveHermesCwd()
+  defaultLabel: path.join(app.getPath('home'), 'hermes-projects')
 }))
 
-ipcMain.handle('hermes:workspace:sanitize', async (_event, cwd) => sanitizeWorkspaceCwd(cwd))
-
 ipcMain.handle('hermes:setting:defaultProjectDir:set', async (_event, dir) => {
   const next = typeof dir === 'string' && dir.trim() ? dir.trim() : null
 
@@ -5629,121 +5285,22 @@ function findGitRoot(start) {
   return null
 }
 
-function isExecutableFile(filePath) {
-  if (!filePath || !path.isAbsolute(filePath)) {
-    return false
+function terminalShellCommand() {
+  if (IS_WINDOWS) {
+    return { args: [], command: process.env.COMSPEC || 'cmd.exe' }
   }
 
-  try {
-    fs.accessSync(filePath, fs.constants.X_OK)
-    return true
-  } catch {
-    return false
-  }
-}
-
-function posixShellSpec(shellPath) {
+  const configuredShell = process.env.SHELL || ''
+  const shellPath =
+    (path.isAbsolute(configuredShell) && fs.existsSync(configuredShell) && configuredShell) ||
+    ['/bin/zsh', '/bin/bash', '/bin/sh'].find(candidate => fs.existsSync(candidate)) ||
+    '/bin/sh'
   const shellName = path.basename(shellPath)
   const interactiveArgs = shellName.includes('zsh') || shellName.includes('bash') ? ['-il'] : ['-i']
 
   return { args: interactiveArgs, command: shellPath, name: shellName }
 }
 
-let spawnHelperChecked = false
-
-// node-pty execs a `spawn-helper` binary on macOS/Linux to launch the shell in a
-// fresh session. The prebuilt that ships in node-pty's `prebuilds/` (and the
-// staged copy under resources/native-deps) loses its execute bit through npm
-// pack / electron-builder file collection, so every nodePty.spawn() dies with
-// "posix_spawnp failed". Restore +x once, lazily, before the first spawn.
-function ensureSpawnHelperExecutable() {
-  if (spawnHelperChecked || IS_WINDOWS || !nodePtyDir) {
-    return
-  }
-
-  spawnHelperChecked = true
-
-  const arch = process.arch
-  const candidates = [
-    path.join(nodePtyDir, 'build', 'Release', 'spawn-helper'),
-    path.join(nodePtyDir, 'prebuilds', `${process.platform}-${arch}`, 'spawn-helper')
-  ]
-
-  for (const helper of candidates) {
-    try {
-      const mode = fs.statSync(helper).mode
-
-      if ((mode & 0o111) !== 0o111) {
-        fs.chmodSync(helper, mode | 0o755)
-      }
-    } catch {
-      // Not present in this layout (e.g. compiled build vs prebuild); skip.
-    }
-  }
-}
-
-// Windows PowerShell 5.1 ships at a fixed System32 path on every Windows box;
-// prefer it only after PowerShell 7+ (`pwsh`).
-function windowsPowerShellPath() {
-  const systemRoot = process.env.SystemRoot || process.env.windir || 'C:\\Windows'
-  const builtin = path.join(systemRoot, 'System32', 'WindowsPowerShell', 'v1.0', 'powershell.exe')
-
-  return isExecutableFile(builtin) ? builtin : findOnPath('powershell.exe')
-}
-
-// Map a resolved shell path to its spawn spec, picking interactive flags by
-// family: PowerShell drops its logo banner (so the prompt sits flush like the
-// POSIX shells), cmd needs nothing, and everything else (zsh/bash/fish/sh…)
-// gets POSIX interactive-login flags.
-function shellSpecFor(shellPath) {
-  const name = path.basename(shellPath).toLowerCase()
-
-  if (name.startsWith('pwsh') || name.startsWith('powershell')) {
-    return { args: ['-NoLogo'], command: shellPath, name }
-  }
-
-  if (name.startsWith('cmd')) {
-    return { args: [], command: shellPath, name }
-  }
-
-  return posixShellSpec(shellPath)
-}
-
-// Best installed Windows shell: PowerShell 7+ (`pwsh`), then Windows PowerShell
-// 5.1, then comspec/cmd.exe as the universal fallback.
-function windowsShellSpec() {
-  const command =
-    findOnPath('pwsh.exe') || findOnPath('pwsh') || windowsPowerShellPath() || process.env.COMSPEC || 'cmd.exe'
-
-  return shellSpecFor(command)
-}
-
-// Resolve the interactive shell for the embedded terminal: an explicit user
-// override wins, otherwise auto-detect the best one installed for the platform.
-function terminalShellCommand() {
-  // HERMES_DESKTOP_SHELL is the cross-platform escape hatch (a path or a bare
-  // name on PATH); $SHELL is honored on POSIX, where it's the user's canonical
-  // choice, but ignored on Windows, where it's usually a stray MSYS/Git path
-  // node-pty can't spawn natively.
-  const override = (process.env.HERMES_DESKTOP_SHELL || (IS_WINDOWS ? '' : process.env.SHELL) || '').trim()
-
-  if (override) {
-    const resolved = isExecutableFile(override) ? override : findOnPath(override)
-
-    if (resolved) {
-      return shellSpecFor(resolved)
-    }
-  }
-
-  if (IS_WINDOWS) {
-    return windowsShellSpec()
-  }
-
-  const shellPath = ['/bin/zsh', '/bin/bash', '/bin/sh'].find(candidate => isExecutableFile(candidate))
-
-  return posixShellSpec(shellPath || '/bin/sh')
-}
-
 function safeTerminalCwd(cwd) {
   const candidate = path.resolve(String(cwd || app.getPath('home')))
 
@@ -5781,11 +5338,6 @@ function terminalShellEnv() {
   env.TERM_PROGRAM = 'Hermes'
   env.TERM_PROGRAM_VERSION = app.getVersion()
 
-  // Let a hermes/--tui launched in this pane know it's embedded in the desktop
-  // GUI (build_environment_hints surfaces this). Distinct from HERMES_DESKTOP,
-  // which marks the agent *backend* and gates cron/gateway behavior.
-  env.HERMES_DESKTOP_TERMINAL = '1'
-
   return env
 }
 
@@ -5857,8 +5409,6 @@ ipcMain.handle('hermes:terminal:start', async (event, payload = {}) => {
     throw new Error('PTY support is unavailable. Reinstall desktop dependencies and restart Hermes.')
   }
 
-  ensureSpawnHelperExecutable()
-
   const id = crypto.randomUUID()
   const { args, command, name } = terminalShellCommand()
   const cwd = safeTerminalCwd(payload?.cwd)
@@ -6038,11 +5588,11 @@ async function getUninstallSummary() {
       resolve(value)
     }
     try {
-      const child = spawn(py, ['-m', 'hermes_cli.main', 'uninstall', '--gui-summary'], hiddenWindowsChildOptions({
+      const child = spawn(py, ['-m', 'hermes_cli.main', 'uninstall', '--gui-summary'], {
         cwd: agentRoot,
         env: { ...process.env, HERMES_HOME, NO_COLOR: '1' },
         stdio: ['ignore', 'pipe', 'ignore']
-      }))
+      })
       child.stdout.on('data', chunk => {
         stdout += chunk.toString()
       })
@@ -6181,12 +5731,6 @@ ipcMain.handle('hermes:uninstall:run', async (_event, payload) => {
   return runDesktopUninstall(String(mode || ''))
 })
 
-// Download a VS Code Marketplace extension and return the raw color-theme JSON
-// it contributes. No theme code is executed — we only read JSON from the .vsix.
-ipcMain.handle('hermes:vscode-theme:fetch', async (_event, id) => fetchMarketplaceThemes(String(id || '')))
-
-// Search the Marketplace for color-theme extensions (empty query = top installs).
-ipcMain.handle('hermes:vscode-theme:search', async (_event, query) => searchMarketplaceThemes(String(query || ''), 20))
 
 app.whenReady().then(() => {
   if (IS_MAC) {
@@ -6202,14 +5746,7 @@ app.whenReady().then(() => {
   createWindow()
 
   app.on('activate', () => {
-    // Recreate the primary window if it's gone. Guard on mainWindow directly
-    // (not just total window count) so a dock click still restores the main
-    // window when only secondary session windows remain open.
-    if (!mainWindow || mainWindow.isDestroyed()) {
-      createWindow()
-    } else {
-      focusWindow(mainWindow)
-    }
+    if (BrowserWindow.getAllWindows().length === 0) createWindow()
   })
 })
 

apps/desktop/electron/preload.cjs

@@ -5,7 +5,6 @@ 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 => 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),
@@ -42,7 +41,6 @@ contextBridge.exposeInMainWorld('hermesDesktop', {
   setPreviewShortcutActive: active => ipcRenderer.send('hermes:previewShortcutActive', Boolean(active)),
   openExternal: url => ipcRenderer.invoke('hermes:openExternal', url),
   fetchLinkTitle: url => ipcRenderer.invoke('hermes:fetchLinkTitle', url),
-  sanitizeWorkspaceCwd: cwd => ipcRenderer.invoke('hermes:workspace:sanitize', cwd),
   settings: {
     getDefaultProjectDir: () => ipcRenderer.invoke('hermes:setting:defaultProjectDir:get'),
     setDefaultProjectDir: dir => ipcRenderer.invoke('hermes:setting:defaultProjectDir:set', dir),
@@ -134,9 +132,5 @@ contextBridge.exposeInMainWorld('hermesDesktop', {
       ipcRenderer.on('hermes:updates:progress', listener)
       return () => ipcRenderer.removeListener('hermes:updates:progress', listener)
     }
-  },
-  themes: {
-    fetchMarketplace: id => ipcRenderer.invoke('hermes:vscode-theme:fetch', id),
-    searchMarketplace: query => ipcRenderer.invoke('hermes:vscode-theme:search', query)
   }
 })

apps/desktop/electron/session-windows.cjs

@@ -1,86 +0,0 @@
-// Secondary "session windows" — one extra OS window per chat so a user can
-// work with multiple chats side by side. The pure, Electron-free pieces live
-// here so they can be unit-tested with node --test (mirroring how the rest of
-// electron/*.cjs splits testable logic out of the main.cjs monolith).
-
-const { pathToFileURL } = require('node:url')
-
-// 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.
-function buildSessionWindowUrl(sessionId, { devServer, rendererIndexPath } = {}) {
-  const route = `#/${encodeURIComponent(sessionId)}`
-
-  if (devServer) {
-    const base = devServer.endsWith('/') ? devServer.slice(0, -1) : devServer
-
-    return `${base}/?win=secondary${route}`
-  }
-
-  return `${pathToFileURL(rendererIndexPath).toString()}?win=secondary${route}`
-}
-
-// A small registry keyed by sessionId that guarantees one window per chat:
-// opening a session that already has a live window focuses it instead of
-// spawning a duplicate, and a window removes itself from the registry when it
-// closes. The actual BrowserWindow construction is injected (the `factory`) so
-// this module stays free of Electron and is unit-testable.
-function createSessionWindowRegistry() {
-  const windows = new Map()
-
-  function openOrFocus(sessionId, factory) {
-    const key = typeof sessionId === 'string' ? sessionId.trim() : ''
-
-    if (!key) {
-      return null
-    }
-
-    const existing = windows.get(key)
-
-    if (existing && !existing.isDestroyed()) {
-      // Focus-or-create: never duplicate a window for the same chat.
-      if (typeof existing.isMinimized === 'function' && existing.isMinimized()) {
-        existing.restore?.()
-      }
-
-      if (typeof existing.isVisible === 'function' && !existing.isVisible()) {
-        existing.show?.()
-      }
-
-      existing.focus?.()
-
-      return existing
-    }
-
-    const win = factory(key)
-
-    if (!win) {
-      return null
-    }
-
-    windows.set(key, win)
-
-    // Self-cleanup on close so the registry never holds a destroyed window.
-    win.on?.('closed', () => {
-      if (windows.get(key) === win) {
-        windows.delete(key)
-      }
-    })
-
-    return win
-  }
-
-  return {
-    openOrFocus,
-    get: key => windows.get(key),
-    has: key => windows.has(key),
-    get size() {
-      return windows.size
-    }
-  }
-}
-
-module.exports = { buildSessionWindowUrl, createSessionWindowRegistry }

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

@@ -1,165 +0,0 @@
-const assert = require('node:assert/strict')
-const test = require('node:test')
-
-const { buildSessionWindowUrl, createSessionWindowRegistry } = require('./session-windows.cjs')
-
-// A minimal fake BrowserWindow: tracks listeners + destroyed state and lets a
-// test fire the 'closed' event, mirroring the slice of the Electron API the
-// registry actually touches.
-function makeFakeWindow() {
-  const listeners = {}
-  const calls = { focus: 0, show: 0, restore: 0 }
-  let destroyed = false
-  let minimized = false
-  let visible = true
-
-  return {
-    on(event, handler) {
-      listeners[event] = handler
-
-      return this
-    },
-    emit(event) {
-      listeners[event]?.()
-    },
-    isDestroyed: () => destroyed,
-    destroy() {
-      destroyed = true
-    },
-    isMinimized: () => minimized,
-    setMinimized(value) {
-      minimized = value
-    },
-    isVisible: () => visible,
-    setVisible(value) {
-      visible = value
-    },
-    restore() {
-      calls.restore += 1
-      minimized = false
-    },
-    show() {
-      calls.show += 1
-      visible = true
-    },
-    focus() {
-      calls.focus += 1
-    },
-    calls
-  }
-}
-
-test('buildSessionWindowUrl puts the secondary flag before the hash route (dev server)', () => {
-  const url = buildSessionWindowUrl('abc123', { devServer: 'http://localhost:5173' })
-
-  assert.equal(url, 'http://localhost:5173/?win=secondary#/abc123')
-})
-
-test('buildSessionWindowUrl avoids a double slash when the dev server has a trailing slash', () => {
-  const url = buildSessionWindowUrl('abc123', { devServer: 'http://localhost:5173/' })
-
-  assert.equal(url, 'http://localhost:5173/?win=secondary#/abc123')
-})
-
-test('buildSessionWindowUrl encodes the session id in the hash route', () => {
-  const url = buildSessionWindowUrl('a b/c', { devServer: 'http://localhost:5173' })
-
-  // The query flag must precede the '#' or HashRouter would swallow it as the
-  // route; the id is URL-encoded so slashes/spaces survive routeSessionId().
-  assert.equal(url, 'http://localhost:5173/?win=secondary#/a%20b%2Fc')
-  assert.ok(url.indexOf('?win=secondary') < url.indexOf('#'))
-})
-
-test('buildSessionWindowUrl builds a packaged file URL with the flag before the hash', () => {
-  const url = buildSessionWindowUrl('abc', { rendererIndexPath: '/opt/app/index.html' })
-
-  assert.match(url, /^file:\/\/.*index\.html\?win=secondary#\/abc$/)
-})
-
-test('registry opens one window per session and focuses on re-open', () => {
-  const registry = createSessionWindowRegistry()
-  let built = 0
-  const win = makeFakeWindow()
-  const factory = () => {
-    built += 1
-
-    return win
-  }
-
-  const first = registry.openOrFocus('s1', factory)
-  const second = registry.openOrFocus('s1', factory)
-
-  assert.equal(built, 1, 'factory runs once for the same session')
-  assert.equal(first, second)
-  assert.equal(registry.size, 1)
-  assert.equal(win.calls.focus, 1, 'second open focuses the existing window')
-})
-
-test('registry restores + shows a minimized/hidden window on re-open', () => {
-  const registry = createSessionWindowRegistry()
-  const win = makeFakeWindow()
-  registry.openOrFocus('s1', () => win)
-
-  win.setMinimized(true)
-  win.setVisible(false)
-  registry.openOrFocus('s1', () => win)
-
-  assert.equal(win.calls.restore, 1)
-  assert.equal(win.calls.show, 1)
-  assert.equal(win.calls.focus, 1)
-})
-
-test('registry drops the entry when the window closes', () => {
-  const registry = createSessionWindowRegistry()
-  const win = makeFakeWindow()
-  registry.openOrFocus('s1', () => win)
-  assert.equal(registry.size, 1)
-
-  win.emit('closed')
-
-  assert.equal(registry.size, 0)
-  assert.equal(registry.has('s1'), false)
-})
-
-test('registry rebuilds a fresh window after the previous one was destroyed', () => {
-  const registry = createSessionWindowRegistry()
-  const first = makeFakeWindow()
-  registry.openOrFocus('s1', () => first)
-  first.destroy()
-
-  let built = 0
-  const second = makeFakeWindow()
-  const result = registry.openOrFocus('s1', () => {
-    built += 1
-
-    return second
-  })
-
-  assert.equal(built, 1, 'a destroyed window is replaced, not focused')
-  assert.equal(result, second)
-})
-
-test('registry ignores empty / non-string session ids', () => {
-  const registry = createSessionWindowRegistry()
-  let built = 0
-  const factory = () => {
-    built += 1
-
-    return makeFakeWindow()
-  }
-
-  assert.equal(registry.openOrFocus('', factory), null)
-  assert.equal(registry.openOrFocus('   ', factory), null)
-  assert.equal(registry.openOrFocus(null, factory), null)
-  assert.equal(registry.openOrFocus(42, factory), null)
-  assert.equal(built, 0)
-  assert.equal(registry.size, 0)
-})
-
-test('registry trims the session id before keying', () => {
-  const registry = createSessionWindowRegistry()
-  const win = makeFakeWindow()
-  registry.openOrFocus('  s1  ', () => win)
-
-  assert.equal(registry.has('s1'), true)
-})

apps/desktop/electron/update-remote.cjs

@@ -1,56 +0,0 @@
-/**
- * Pure helpers for choosing a remote URL during passive update checks.
- *
- * A public install can end up with `origin=git@github.com:NousResearch/hermes-agent.git`.
- * If the user's GitHub SSH key is FIDO2/passkey-backed, a background `git fetch
- * origin` triggers an unexplained hardware-touch prompt. For passive checks
- * against the official repo we substitute the public HTTPS `ls-remote` path,
- * which needs no auth and cannot prompt. Active update/apply flows are left
- * unchanged.
- *
- * Extracted from main.cjs so the security-critical remote detection is unit
- * testable without booting Electron (main.cjs requires('electron') at load).
- */
-
-const OFFICIAL_REPO_HTTPS_URL = 'https://github.com/NousResearch/hermes-agent.git'
-const OFFICIAL_REPO_CANONICAL = 'github.com/nousresearch/hermes-agent'
-
-// Normalize common GitHub remote URL forms to `host/owner/repo` (lowercased,
-// no trailing slash, no .git suffix) so SSH and HTTPS forms of the same repo
-// compare equal.
-function canonicalGitHubRemote(url) {
-  if (!url) return ''
-  let value = String(url).trim()
-  if (value.startsWith('git@github.com:')) {
-    value = `github.com/${value.slice('git@github.com:'.length)}`
-  } else if (value.startsWith('ssh://git@github.com/')) {
-    value = `github.com/${value.slice('ssh://git@github.com/'.length)}`
-  } else {
-    try {
-      const parsed = new URL(value)
-      if (parsed.hostname && parsed.pathname) value = `${parsed.hostname}${parsed.pathname}`
-    } catch {
-      // Leave non-URL forms unchanged.
-    }
-  }
-  value = value.trim().replace(/\/+$/, '')
-  if (value.endsWith('.git')) value = value.slice(0, -4)
-  return value.toLowerCase()
-}
-
-function isSshRemote(url) {
-  const value = String(url || '').trim().toLowerCase()
-  return value.startsWith('git@') || value.startsWith('ssh://')
-}
-
-function isOfficialSshRemote(url) {
-  return isSshRemote(url) && canonicalGitHubRemote(url) === OFFICIAL_REPO_CANONICAL
-}
-
-module.exports = {
-  OFFICIAL_REPO_HTTPS_URL,
-  OFFICIAL_REPO_CANONICAL,
-  canonicalGitHubRemote,
-  isSshRemote,
-  isOfficialSshRemote
-}

apps/desktop/electron/update-remote.test.cjs

@@ -1,78 +0,0 @@
-/**
- * Tests for electron/update-remote.cjs — the remote-detection helpers that
- * keep passive update checks off the SSH origin for official installs.
- *
- * Run with: node --test electron/update-remote.test.cjs
- * (Wired into npm test:desktop:platforms in package.json.)
- *
- * Why this matters: a public install can carry
- * origin=git@github.com:NousResearch/hermes-agent.git. A background
- * `git fetch origin` then authenticates over SSH and, with a FIDO2/passkey
- * key, triggers an unexplained hardware-touch prompt. isOfficialSshRemote
- * must reliably recognize the official SSH remote (in every URL form,
- * case-insensitively) so the caller can swap in the anonymous HTTPS path —
- * while NOT misclassifying forks, other hosts, or the HTTPS remote (which
- * never prompts and should keep the normal fetch path).
- */
-
-const test = require('node:test')
-const assert = require('node:assert/strict')
-
-const {
-  OFFICIAL_REPO_HTTPS_URL,
-  OFFICIAL_REPO_CANONICAL,
-  canonicalGitHubRemote,
-  isSshRemote,
-  isOfficialSshRemote
-} = require('./update-remote.cjs')
-
-test('canonicalGitHubRemote normalizes SSH and HTTPS forms to the same value', () => {
-  assert.equal(canonicalGitHubRemote('git@github.com:NousResearch/hermes-agent.git'), OFFICIAL_REPO_CANONICAL)
-  assert.equal(canonicalGitHubRemote('git@github.com:NousResearch/hermes-agent'), OFFICIAL_REPO_CANONICAL)
-  assert.equal(canonicalGitHubRemote('ssh://git@github.com/NousResearch/hermes-agent.git'), OFFICIAL_REPO_CANONICAL)
-  assert.equal(canonicalGitHubRemote('https://github.com/NousResearch/hermes-agent.git'), OFFICIAL_REPO_CANONICAL)
-  // Case-insensitive: an uppercased owner still canonicalizes to the same repo.
-  assert.equal(canonicalGitHubRemote('git@github.com:nousresearch/hermes-agent.git'), OFFICIAL_REPO_CANONICAL)
-  // Trailing slashes are stripped.
-  assert.equal(canonicalGitHubRemote('https://github.com/NousResearch/hermes-agent/'), OFFICIAL_REPO_CANONICAL)
-})
-
-test('canonicalGitHubRemote is empty for falsy input', () => {
-  assert.equal(canonicalGitHubRemote(''), '')
-  assert.equal(canonicalGitHubRemote(null), '')
-  assert.equal(canonicalGitHubRemote(undefined), '')
-})
-
-test('isSshRemote detects scp-like and ssh:// forms only', () => {
-  assert.equal(isSshRemote('git@github.com:NousResearch/hermes-agent.git'), true)
-  assert.equal(isSshRemote('ssh://git@github.com/NousResearch/hermes-agent.git'), true)
-  assert.equal(isSshRemote('https://github.com/NousResearch/hermes-agent.git'), false)
-  assert.equal(isSshRemote(''), false)
-  assert.equal(isSshRemote(null), false)
-})
-
-test('isOfficialSshRemote is true only for the official repo over SSH', () => {
-  assert.equal(isOfficialSshRemote('git@github.com:NousResearch/hermes-agent.git'), true)
-  assert.equal(isOfficialSshRemote('git@github.com:NousResearch/hermes-agent'), true)
-  assert.equal(isOfficialSshRemote('ssh://git@github.com/NousResearch/hermes-agent.git'), true)
-  // Case-insensitive owner/repo match.
-  assert.equal(isOfficialSshRemote('git@github.com:nousresearch/hermes-agent.git'), true)
-})
-
-test('isOfficialSshRemote does NOT match forks, other hosts, or HTTPS', () => {
-  // A fork over SSH belongs to the user — fetching it is their own remote,
-  // not the official upstream, so the SSH-avoidance swap must not apply.
-  assert.equal(isOfficialSshRemote('git@github.com:someuser/hermes-agent.git'), false)
-  // Same repo name on a different host is not the official repo.
-  assert.equal(isOfficialSshRemote('git@gitlab.com:NousResearch/hermes-agent.git'), false)
-  // HTTPS to the official repo never prompts for SSH/FIDO2, so it keeps the
-  // normal fetch path — must not be flagged as an official SSH remote.
-  assert.equal(isOfficialSshRemote('https://github.com/NousResearch/hermes-agent.git'), false)
-  assert.equal(isOfficialSshRemote(''), false)
-  assert.equal(isOfficialSshRemote(null), false)
-})
-
-test('OFFICIAL_REPO_HTTPS_URL canonicalizes to OFFICIAL_REPO_CANONICAL', () => {
-  // Invariant: the URL we substitute in must be the same repo we detect.
-  assert.equal(canonicalGitHubRemote(OFFICIAL_REPO_HTTPS_URL), OFFICIAL_REPO_CANONICAL)
-})

apps/desktop/electron/vscode-marketplace.cjs

@@ -1,331 +0,0 @@
-'use strict'
-
-/**
- * VS Code Marketplace color-theme fetcher (main process).
- *
- * Resolves an extension's latest version via the (undocumented but stable)
- * gallery ExtensionQuery API, downloads the `.vsix` (a zip), and extracts the
- * color-theme JSON files it contributes. No theme code is ever executed — we
- * only read `package.json` + the referenced `*.json` theme files out of the
- * archive and hand their text back to the renderer to convert.
- *
- * Dependency-free on purpose: a `.vsix` is a plain zip, so we parse the central
- * directory and inflate just the entries we need with `zlib`. Avoids pulling a
- * zip library into the desktop bundle for a feature this small.
- */
-
-const https = require('node:https')
-const zlib = require('node:zlib')
-
-const GALLERY_QUERY_URL = 'https://marketplace.visualstudio.com/_apis/public/gallery/extensionquery'
-const VSIX_ASSET_TYPE = 'Microsoft.VisualStudio.Services.VSIXPackage'
-const MAX_VSIX_BYTES = 40 * 1024 * 1024 // 40 MB — themes are tiny; this is paranoia.
-const MAX_REDIRECTS = 5
-const REQUEST_TIMEOUT_MS = 20_000
-
-const ID_RE = /^[\w-]+\.[\w-]+$/
-
-/** Minimal HTTPS helper with redirect-following, timeout, and a size cap. */
-function request(url, { method = 'GET', headers = {}, body = null, maxBytes = MAX_VSIX_BYTES } = {}, redirectsLeft = MAX_REDIRECTS) {
-  return new Promise((resolve, reject) => {
-    const req = https.request(url, { method, headers }, res => {
-      const status = res.statusCode ?? 0
-
-      if (status >= 300 && status < 400 && res.headers.location) {
-        if (redirectsLeft <= 0) {
-          res.resume()
-          reject(new Error('Too many redirects.'))
-
-          return
-        }
-
-        const next = new URL(res.headers.location, url).toString()
-        res.resume()
-        // Redirects to the CDN are plain GETs (drop the POST body).
-        resolve(request(next, { method: 'GET', headers: { 'User-Agent': headers['User-Agent'] }, maxBytes }, redirectsLeft - 1))
-
-        return
-      }
-
-      if (status < 200 || status >= 300) {
-        res.resume()
-        reject(new Error(`Request failed (${status}) for ${url}`))
-
-        return
-      }
-
-      const chunks = []
-      let total = 0
-
-      res.on('data', chunk => {
-        total += chunk.length
-
-        if (total > maxBytes) {
-          req.destroy()
-          reject(new Error('Response exceeded the size limit.'))
-
-          return
-        }
-
-        chunks.push(chunk)
-      })
-      res.on('end', () => resolve(Buffer.concat(chunks)))
-    })
-
-    req.on('error', reject)
-    req.setTimeout(REQUEST_TIMEOUT_MS, () => req.destroy(new Error('Request timed out.')))
-
-    if (body) {
-      req.write(body)
-    }
-
-    req.end()
-  })
-}
-
-/** Resolve `{ displayName, vsixUrl }` for the latest version of `id`. */
-async function resolveExtension(id) {
-  const json = await queryGallery({
-    // FilterType 7 = ExtensionName (the full publisher.extension id).
-    filters: [{ criteria: [{ filterType: 7, value: id }], pageNumber: 1, pageSize: 1 }],
-    // Flags: IncludeFiles | IncludeVersionProperties | IncludeAssetUri |
-    // IncludeCategoryAndTags | IncludeLatestVersionOnly = 914.
-    flags: 914
-  })
-  const extension = json?.results?.[0]?.extensions?.[0]
-
-  if (!extension) {
-    throw new Error(`Extension "${id}" was not found on the Marketplace.`)
-  }
-
-  const version = extension.versions?.[0]
-
-  if (!version) {
-    throw new Error(`Extension "${id}" has no published versions.`)
-  }
-
-  const asset = (version.files ?? []).find(file => file.assetType === VSIX_ASSET_TYPE)
-  const vsixUrl = asset?.source
-
-  if (!vsixUrl) {
-    throw new Error(`Could not find a downloadable package for "${id}".`)
-  }
-
-  return { displayName: extension.displayName || id, vsixUrl }
-}
-
-/** POST an ExtensionQuery payload and return the parsed gallery response. */
-async function queryGallery(payload, { maxBytes = 4 * 1024 * 1024 } = {}) {
-  const body = JSON.stringify(payload)
-  const raw = await request(GALLERY_QUERY_URL, {
-    method: 'POST',
-    headers: {
-      Accept: 'application/json;api-version=3.0-preview.1',
-      'Content-Type': 'application/json',
-      'Content-Length': Buffer.byteLength(body),
-      'User-Agent': 'Hermes-Desktop'
-    },
-    body,
-    maxBytes
-  })
-
-  return JSON.parse(raw.toString('utf8'))
-}
-
-/**
- * Search the Marketplace for color-theme extensions. With an empty query this
- * returns the most-installed themes; with a query it's a full-text search
- * scoped to the Themes category. Returns lightweight cards (no download).
- */
-/**
- * The "Themes" category also contains file-icon and product-icon themes (the
- * gallery has no color-only category). We can't see an extension's actual
- * contributions without downloading it, so filter the obvious icon packs out by
- * tag + name/description. Color themes that also ship icons are rare; worst case
- * a user installs them by exact id from settings.
- */
-function looksLikeIconTheme(extension) {
-  const tags = (extension.tags ?? []).map(tag => String(tag).toLowerCase())
-
-  if (tags.includes('icon-theme') || tags.includes('product-icon-theme')) {
-    return true
-  }
-
-  const text = `${extension.displayName ?? ''} ${extension.shortDescription ?? ''}`.toLowerCase()
-
-  return /\b(icon theme|file icons?|product icons?|icon pack|fileicons)\b/.test(text)
-}
-
-async function searchMarketplaceThemes(query, limit = 20) {
-  const text = String(query || '').trim()
-  const pageSize = Math.min(Math.max(Number(limit) || 20, 1), 50)
-
-  // FilterType: 8=Target, 5=Category, 10=SearchText, 12=ExcludeWithFlags.
-  const criteria = [
-    { filterType: 8, value: 'Microsoft.VisualStudio.Code' },
-    { filterType: 5, value: 'Themes' },
-    { filterType: 12, value: '4096' } // Exclude unpublished (Unpublished = 0x1000).
-  ]
-
-  if (text) {
-    criteria.push({ filterType: 10, value: text })
-  }
-
-  const json = await queryGallery({
-    // Over-fetch so the icon-theme filter below still leaves a full page.
-    filters: [{ criteria, pageNumber: 1, pageSize: Math.min(pageSize * 2, 50), sortBy: 4, sortOrder: 0 }],
-    // IncludeStatistics (0x100) | IncludeLatestVersionOnly (0x200) | IncludeCategoryAndTags (0x4).
-    flags: 772
-  })
-
-  const extensions = json?.results?.[0]?.extensions ?? []
-
-  return extensions
-    .filter(extension => !looksLikeIconTheme(extension))
-    .slice(0, pageSize)
-    .map(extension => {
-      const publisherName = extension.publisher?.publisherName ?? ''
-      const installStat = (extension.statistics ?? []).find(stat => stat.statisticName === 'install')
-
-      return {
-        extensionId: `${publisherName}.${extension.extensionName}`,
-        displayName: extension.displayName || extension.extensionName,
-        publisher: extension.publisher?.displayName || publisherName,
-        description: extension.shortDescription || '',
-        installs: Math.round(installStat?.value ?? 0)
-      }
-    })
-}
-
-// ─── Minimal zip reader ─────────────────────────────────────────────────────
-
-function findEndOfCentralDirectory(buf) {
-  // EOCD signature 0x06054b50, scanning back from the end (comment is rare).
-  for (let i = buf.length - 22; i >= 0; i--) {
-    if (buf.readUInt32LE(i) === 0x06054b50) {
-      return i
-    }
-  }
-
-  throw new Error('Not a valid zip archive (no end-of-central-directory).')
-}
-
-/** Parse the central directory into a name → record map. */
-function readCentralDirectory(buf) {
-  const eocd = findEndOfCentralDirectory(buf)
-  const count = buf.readUInt16LE(eocd + 10)
-  let offset = buf.readUInt32LE(eocd + 16)
-  const records = new Map()
-
-  for (let i = 0; i < count; i++) {
-    if (buf.readUInt32LE(offset) !== 0x02014b50) {
-      break
-    }
-
-    const method = buf.readUInt16LE(offset + 10)
-    const compressedSize = buf.readUInt32LE(offset + 20)
-    const nameLen = buf.readUInt16LE(offset + 28)
-    const extraLen = buf.readUInt16LE(offset + 30)
-    const commentLen = buf.readUInt16LE(offset + 32)
-    const localOffset = buf.readUInt32LE(offset + 42)
-    const name = buf.toString('utf8', offset + 46, offset + 46 + nameLen)
-
-    records.set(name, { method, compressedSize, localOffset })
-    offset += 46 + nameLen + extraLen + commentLen
-  }
-
-  return records
-}
-
-/** Inflate a single entry to a string. */
-function extractEntry(buf, record) {
-  // The local header's name/extra lengths can differ from the central record,
-  // so re-read them here to locate the compressed payload.
-  if (buf.readUInt32LE(record.localOffset) !== 0x04034b50) {
-    throw new Error('Corrupt zip: bad local file header.')
-  }
-
-  const nameLen = buf.readUInt16LE(record.localOffset + 26)
-  const extraLen = buf.readUInt16LE(record.localOffset + 28)
-  const dataStart = record.localOffset + 30 + nameLen + extraLen
-  const data = buf.subarray(dataStart, dataStart + record.compressedSize)
-
-  // 0 = stored, 8 = deflate. Theme files are one or the other.
-  return record.method === 0 ? data.toString('utf8') : zlib.inflateRawSync(data).toString('utf8')
-}
-
-/** Normalize a package.json theme path to its zip entry name. */
-function themeEntryName(themePath) {
-  const clean = String(themePath).replace(/^\.\//, '').replace(/^\//, '')
-
-  return `extension/${clean}`
-}
-
-/** Extract every contributed color theme from a `.vsix` buffer. */
-function extractThemes(vsixBuffer) {
-  const records = readCentralDirectory(vsixBuffer)
-  const pkgRecord = records.get('extension/package.json')
-
-  if (!pkgRecord) {
-    throw new Error('Package manifest missing from the extension.')
-  }
-
-  const pkg = JSON.parse(extractEntry(vsixBuffer, pkgRecord))
-  const contributed = pkg?.contributes?.themes
-
-  if (!Array.isArray(contributed) || contributed.length === 0) {
-    return []
-  }
-
-  const themes = []
-
-  for (const entry of contributed) {
-    if (!entry?.path) {
-      continue
-    }
-
-    const record = records.get(themeEntryName(entry.path))
-
-    if (!record) {
-      continue
-    }
-
-    try {
-      themes.push({
-        label: entry.label || entry.id || pkg.displayName || pkg.name || 'VS Code Theme',
-        uiTheme: entry.uiTheme,
-        contents: extractEntry(vsixBuffer, record)
-      })
-    } catch {
-      // Skip an entry we can't inflate rather than failing the whole install.
-    }
-  }
-
-  return themes
-}
-
-/**
- * Public entry: resolve, download, and extract color themes for `id`
- * (`publisher.extension`). Returns `{ extensionId, displayName, themes }`.
- */
-async function fetchMarketplaceThemes(id) {
-  const trimmed = String(id || '').trim()
-
-  if (!ID_RE.test(trimmed)) {
-    throw new Error('Expected a Marketplace id like "publisher.extension".')
-  }
-
-  const { displayName, vsixUrl } = await resolveExtension(trimmed)
-  const vsix = await request(vsixUrl, { headers: { 'User-Agent': 'Hermes-Desktop' } })
-  const themes = extractThemes(vsix)
-
-  return { extensionId: trimmed, displayName, themes }
-}
-
-module.exports = {
-  fetchMarketplaceThemes,
-  searchMarketplaceThemes,
-  extractThemes,
-  readCentralDirectory,
-  __testing: { themeEntryName, looksLikeIconTheme }
-}

apps/desktop/electron/vscode-marketplace.test.cjs

@@ -1,113 +0,0 @@
-'use strict'
-
-const assert = require('node:assert')
-const test = require('node:test')
-
-const { __testing, extractThemes, readCentralDirectory } = require('./vscode-marketplace.cjs')
-
-// Build a minimal zip with stored (uncompressed) entries so the test controls
-// the bytes exactly — exercises the central-directory reader + theme extraction
-// without a deflate dependency.
-function makeZip(entries) {
-  const locals = []
-  const centrals = []
-  let offset = 0
-
-  for (const { name, data } of entries) {
-    const nameBuf = Buffer.from(name, 'utf8')
-    const body = Buffer.from(data, 'utf8')
-
-    const local = Buffer.alloc(30 + nameBuf.length)
-    local.writeUInt32LE(0x04034b50, 0)
-    local.writeUInt16LE(0, 8) // method: stored
-    local.writeUInt32LE(body.length, 18) // compressed size
-    local.writeUInt32LE(body.length, 22) // uncompressed size
-    local.writeUInt16LE(nameBuf.length, 26)
-    nameBuf.copy(local, 30)
-
-    locals.push(local, body)
-
-    const central = Buffer.alloc(46 + nameBuf.length)
-    central.writeUInt32LE(0x02014b50, 0)
-    central.writeUInt16LE(0, 10) // method: stored
-    central.writeUInt32LE(body.length, 20)
-    central.writeUInt32LE(body.length, 24)
-    central.writeUInt16LE(nameBuf.length, 28)
-    central.writeUInt32LE(offset, 42) // local header offset
-    nameBuf.copy(central, 46)
-
-    centrals.push(central)
-    offset += local.length + body.length
-  }
-
-  const centralStart = offset
-  const centralBuf = Buffer.concat(centrals)
-
-  const eocd = Buffer.alloc(22)
-  eocd.writeUInt32LE(0x06054b50, 0)
-  eocd.writeUInt16LE(entries.length, 8)
-  eocd.writeUInt16LE(entries.length, 10)
-  eocd.writeUInt32LE(centralBuf.length, 12)
-  eocd.writeUInt32LE(centralStart, 16)
-
-  return Buffer.concat([...locals, centralBuf, eocd])
-}
-
-test('readCentralDirectory finds every entry', () => {
-  const zip = makeZip([
-    { name: 'extension/package.json', data: '{}' },
-    { name: 'extension/themes/x.json', data: '{}' }
-  ])
-
-  const records = readCentralDirectory(zip)
-  assert.ok(records.has('extension/package.json'))
-  assert.ok(records.has('extension/themes/x.json'))
-})
-
-test('extractThemes reads contributed color themes (resolving ./ paths)', () => {
-  const pkg = JSON.stringify({
-    name: 'theme-dracula',
-    displayName: 'Dracula',
-    contributes: {
-      themes: [{ label: 'Dracula', uiTheme: 'vs-dark', path: './themes/dracula.json' }]
-    }
-  })
-  const themeJson = JSON.stringify({ name: 'Dracula', type: 'dark', colors: { 'editor.background': '#282a36' } })
-
-  const zip = makeZip([
-    { name: 'extension/package.json', data: pkg },
-    { name: 'extension/themes/dracula.json', data: themeJson }
-  ])
-
-  const themes = extractThemes(zip)
-  assert.strictEqual(themes.length, 1)
-  assert.strictEqual(themes[0].label, 'Dracula')
-  assert.strictEqual(themes[0].uiTheme, 'vs-dark')
-  assert.match(themes[0].contents, /editor\.background/)
-})
-
-test('extractThemes returns empty when the extension contributes no themes', () => {
-  const zip = makeZip([{ name: 'extension/package.json', data: JSON.stringify({ name: 'x', contributes: {} }) }])
-  assert.deepStrictEqual(extractThemes(zip), [])
-})
-
-test('extractThemes throws when the manifest is missing', () => {
-  const zip = makeZip([{ name: 'extension/other.txt', data: 'hi' }])
-  assert.throws(() => extractThemes(zip), /manifest missing/i)
-})
-
-test('looksLikeIconTheme filters icon/product-icon packs out of theme search', () => {
-  const { looksLikeIconTheme } = __testing
-
-  // Tagged contribution points are the strongest signal.
-  assert.strictEqual(looksLikeIconTheme({ tags: ['theme', 'icon-theme'] }), true)
-  assert.strictEqual(looksLikeIconTheme({ tags: ['product-icon-theme'] }), true)
-
-  // Name/description fallback for packs that don't tag themselves.
-  assert.strictEqual(looksLikeIconTheme({ displayName: 'Material Icon Theme' }), true)
-  assert.strictEqual(looksLikeIconTheme({ shortDescription: 'A pack of file icons.' }), true)
-
-  // Real color themes survive.
-  assert.strictEqual(looksLikeIconTheme({ displayName: 'Dracula Official', tags: ['theme', 'color-theme'] }), false)
-  assert.strictEqual(looksLikeIconTheme({ displayName: 'One Dark Pro' }), false)
-})

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

@@ -1,54 +0,0 @@
-'use strict'
-
-const test = require('node:test')
-const assert = require('node:assert/strict')
-const fs = require('node:fs')
-const path = require('node:path')
-
-const ELECTRON_DIR = __dirname
-
-function readElectronFile(name) {
-  return fs.readFileSync(path.join(ELECTRON_DIR, name), 'utf8')
-}
-
-function requireHiddenChildOptions(source, needle) {
-  const index = source.indexOf(needle)
-  assert.notEqual(index, -1, `missing call site: ${needle}`)
-  const snippet = source.slice(index, index + 700)
-  assert.match(
-    snippet,
-    /hiddenWindowsChildOptions\(/,
-    `expected ${needle} to wrap child-process options with hiddenWindowsChildOptions`
-  )
-}
-
-test('desktop background child processes opt into hidden Windows consoles', () => {
-  const source = readElectronFile('main.cjs')
-
-  assert.match(source, /function hiddenWindowsChildOptions\(options = \{\}\)/)
-
-  requireHiddenChildOptions(source, "execFileSync(\n          'reg'")
-  requireHiddenChildOptions(source, 'execFileSync(pyExe')
-  requireHiddenChildOptions(source, 'spawn(resolveGitBinary()')
-  requireHiddenChildOptions(source, "execFileSync('taskkill'")
-  requireHiddenChildOptions(source, 'spawn(command, args')
-  requireHiddenChildOptions(source, "spawn('curl'")
-  requireHiddenChildOptions(source, 'spawn(backend.command, backend.args')
-  requireHiddenChildOptions(source, 'hermesProcess = spawn(backend.command, backend.args')
-  requireHiddenChildOptions(source, "spawn(py, ['-m', 'hermes_cli.main', 'uninstall', '--gui-summary']")
-})
-
-test('intentional or interactive desktop child processes stay documented', () => {
-  const source = readElectronFile('main.cjs')
-
-  assert.match(source, /windowsHide: false/)
-  assert.match(source, /nodePty\.spawn\(command, args/)
-  assert.match(source, /spawn\('cmd\.exe', \['\/c', 'start'/)
-})
-
-test('bootstrap PowerShell runner hides Windows console children', () => {
-  const source = readElectronFile('bootstrap-runner.cjs')
-
-  assert.match(source, /function hiddenWindowsChildOptions\(options = \{\}\)/)
-  requireHiddenChildOptions(source, 'spawn(ps, fullArgs')
-})

apps/desktop/electron/workspace-cwd.cjs

@@ -1,38 +0,0 @@
-const path = require('node:path')
-
-/** True when `dir` lives inside a packaged app bundle / install tree. */
-function isPackagedInstallPath(dir, { installRoots, isPackaged }) {
-  if (!isPackaged || !dir) {
-    return false
-  }
-
-  let resolved
-
-  try {
-    resolved = path.resolve(String(dir))
-  } catch {
-    return false
-  }
-
-  const roots = new Set(
-    (installRoots ?? [])
-      .filter(Boolean)
-      .map(candidate => path.resolve(String(candidate)))
-  )
-
-  for (const root of roots) {
-    if (resolved === root) {
-      return true
-    }
-
-    const rel = path.relative(root, resolved)
-
-    if (rel && !rel.startsWith('..') && !path.isAbsolute(rel)) {
-      return true
-    }
-  }
-
-  return false
-}
-
-module.exports = { isPackagedInstallPath }

apps/desktop/electron/workspace-cwd.test.cjs

@@ -1,45 +0,0 @@
-/**
- * Tests for electron/workspace-cwd.cjs.
- *
- * Run with: node --test electron/workspace-cwd.test.cjs
- */
-
-const test = require('node:test')
-const assert = require('node:assert/strict')
-const path = require('node:path')
-
-const { isPackagedInstallPath } = require('./workspace-cwd.cjs')
-
-const installRoot = path.resolve('/opt/Hermes')
-
-test('isPackagedInstallPath returns false when not packaged', () => {
-  assert.equal(
-    isPackagedInstallPath(installRoot, { isPackaged: false, installRoots: [installRoot] }),
-    false
-  )
-})
-
-test('isPackagedInstallPath flags the install root itself', () => {
-  assert.equal(
-    isPackagedInstallPath(installRoot, { isPackaged: true, installRoots: [installRoot] }),
-    true
-  )
-})
-
-test('isPackagedInstallPath flags paths nested under the install root', () => {
-  const nested = path.join(installRoot, 'resources', 'app.asar')
-
-  assert.equal(
-    isPackagedInstallPath(nested, { isPackaged: true, installRoots: [installRoot] }),
-    true
-  )
-})
-
-test('isPackagedInstallPath ignores paths outside the install root', () => {
-  const homeProject = path.resolve('/home/user/projects/demo')
-
-  assert.equal(
-    isPackagedInstallPath(homeProject, { isPackaged: true, installRoots: [installRoot] }),
-    false
-  )
-})

apps/desktop/eslint.config.mjs

@@ -3,6 +3,7 @@ import typescriptEslint from '@typescript-eslint/eslint-plugin'
 import typescriptParser from '@typescript-eslint/parser'
 import perfectionist from 'eslint-plugin-perfectionist'
 import reactPlugin from 'eslint-plugin-react'
+import reactCompiler from 'eslint-plugin-react-compiler'
 import hooksPlugin from 'eslint-plugin-react-hooks'
 import unusedImports from 'eslint-plugin-unused-imports'
 import globals from 'globals'
@@ -46,6 +47,7 @@ export default [
       'custom-rules': customRules,
       perfectionist,
       react: reactPlugin,
+      'react-compiler': reactCompiler,
       'react-hooks': hooksPlugin,
       'unused-imports': unusedImports
     },
@@ -96,6 +98,7 @@ export default [
       'perfectionist/sort-jsx-props': ['error', { order: 'asc', type: 'natural' }],
       'perfectionist/sort-named-exports': ['error', { order: 'asc', type: 'natural' }],
       'perfectionist/sort-named-imports': ['error', { order: 'asc', type: 'natural' }],
+      'react-compiler/react-compiler': 'warn',
       'react-hooks/exhaustive-deps': 'warn',
       'react-hooks/rules-of-hooks': 'error',
       'unused-imports/no-unused-imports': 'error'

apps/desktop/package.json

@@ -18,7 +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 && 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",
     "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",
@@ -35,8 +35,8 @@
     "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-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/windows-child-process.test.cjs electron/update-remote.test.cjs",
-    "typecheck": "tsc -p . --noEmit",
+    "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",
+    "type-check": "tsc -b",
     "lint": "eslint src/ electron/",
     "lint:fix": "eslint src/ electron/ --fix",
     "fmt": "prettier --write 'src/**/*.{ts,tsx}' 'electron/**/*.{js,cjs}' 'vite.config.ts'",
@@ -103,19 +103,20 @@
     "@testing-library/dom": "^10.4.0",
     "@testing-library/react": "^16.3.2",
     "@types/hast": "^3.0.4",
-    "@types/node": "^24.12.0",
+    "@types/node": "^24.12.2",
     "@types/react": "^19.2.14",
     "@types/react-dom": "^19.2.3",
     "@typescript-eslint/eslint-plugin": "^8.59.1",
     "@typescript-eslint/parser": "^8.59.1",
     "@vitejs/plugin-react": "^6.0.1",
-    "concurrently": "^10.0.3",
+    "concurrently": "^9.2.1",
     "cross-env": "^10.1.0",
     "electron": "^40.9.3",
     "electron-builder": "^26.8.1",
     "eslint": "^9.39.4",
     "eslint-plugin-perfectionist": "^5.9.0",
     "eslint-plugin-react": "^7.37.5",
+    "eslint-plugin-react-compiler": "^19.1.0-rc.2",
     "eslint-plugin-react-hooks": "^7.1.1",
     "eslint-plugin-unused-imports": "^4.4.1",
     "globals": "^16.5.0",
@@ -165,8 +166,7 @@
     "afterSign": "scripts/notarize.cjs",
     "asarUnpack": [
       "**/*.node",
-      "**/prebuilds/**",
-      "dist/**"
+      "**/prebuilds/**"
     ],
     "mac": {
       "category": "public.app-category.developer-tools",

apps/desktop/scripts/assert-dist-built.cjs

@@ -1,70 +0,0 @@
-"use strict"
-
-// Build-time guard: refuse to hand a half-built renderer to electron-builder.
-//
-// `npm run pack` / `npm run dist*` are `npm run build && npm run builder`.
-// If the `build` step (tsc -b && vite build) fails but packaging proceeds
-// anyway — a stale checkout that fails typecheck, an interrupted vite build,
-// or npm not short-circuiting `&&` in some shells — electron-builder happily
-// packages an app with an empty or missing `dist/`. The result launches but
-// blank-pages with `ERR_FILE_NOT_FOUND` for dist/index.html, with no clue why.
-//
-// This runs at the tail of `build`, after vite build, so any packaging path
-// inherits it. It fails loud and early instead of shipping a broken bundle.
-// See issues #39484 (renderer blank page) and #41327 / #39472 (dashboard 404).
-
-const fs = require("fs")
-const path = require("path")
-
-// Pure check — returns { ok: true } or { ok: false, error: "..." }.
-// Kept side-effect-free so it can be unit tested without spawning a process.
-function checkDistBuilt(distDir) {
-  if (!fs.existsSync(distDir) || !fs.statSync(distDir).isDirectory()) {
-    return { ok: false, error: `no dist directory at ${distDir}` }
-  }
-
-  const indexHtml = path.join(distDir, "index.html")
-  if (!fs.existsSync(indexHtml) || !fs.statSync(indexHtml).isFile()) {
-    return { ok: false, error: `dist/index.html is missing at ${indexHtml}` }
-  }
-  if (fs.statSync(indexHtml).size === 0) {
-    return { ok: false, error: `dist/index.html is empty at ${indexHtml}` }
-  }
-
-  // index.html alone isn't enough — vite emits hashed JS into dist/assets.
-  // An index.html with no script bundle still blank-pages.
-  const assetsDir = path.join(distDir, "assets")
-  const hasAssets =
-    fs.existsSync(assetsDir) &&
-    fs.statSync(assetsDir).isDirectory() &&
-    fs.readdirSync(assetsDir).some(name => name.endsWith(".js"))
-  if (!hasAssets) {
-    return { ok: false, error: `dist/assets has no built JS bundle (expected vite output under ${assetsDir})` }
-  }
-
-  return { ok: true }
-}
-
-function main() {
-  const desktopRoot = path.resolve(__dirname, "..")
-  const distDir = path.join(desktopRoot, "dist")
-  const result = checkDistBuilt(distDir)
-
-  if (!result.ok) {
-    console.error(`\n✗ assert-dist-built: ${result.error}`)
-    console.error("  The renderer bundle is missing or incomplete, so packaging")
-    console.error("  would produce an app that launches to a blank page.")
-    console.error("  Re-run the build and check the tsc/vite output above for the")
-    console.error("  real failure, then package again:")
-    console.error(`    cd ${desktopRoot} && npm run build\n`)
-    process.exit(1)
-  }
-
-  console.log("✓ assert-dist-built: dist/index.html + assets present")
-}
-
-if (require.main === module) {
-  main()
-}
-
-module.exports = { checkDistBuilt }

apps/desktop/scripts/assert-dist-built.test.cjs

@@ -1,84 +0,0 @@
-const assert = require('node:assert/strict')
-const fs = require('node:fs')
-const os = require('node:os')
-const path = require('node:path')
-const test = require('node:test')
-
-const { checkDistBuilt } = require('../scripts/assert-dist-built.cjs')
-
-function makeDist(extra) {
-  const tempRoot = fs.mkdtempSync(path.join(os.tmpdir(), 'hermes-assert-dist-'))
-  const distDir = path.join(tempRoot, 'dist')
-  fs.mkdirSync(distDir, { recursive: true })
-  if (extra) extra(distDir)
-  return { tempRoot, distDir }
-}
-
-test('checkDistBuilt passes when index.html + an assets JS bundle exist', () => {
-  const { tempRoot, distDir } = makeDist(d => {
-    fs.writeFileSync(path.join(d, 'index.html'), '<!doctype html><div id=root></div>', 'utf8')
-    fs.mkdirSync(path.join(d, 'assets'))
-    fs.writeFileSync(path.join(d, 'assets', 'index-abc123.js'), 'console.log(1)', 'utf8')
-  })
-  try {
-    assert.deepEqual(checkDistBuilt(distDir), { ok: true })
-  } finally {
-    fs.rmSync(tempRoot, { recursive: true, force: true })
-  }
-})
-
-test('checkDistBuilt fails when the dist directory is absent', () => {
-  const tempRoot = fs.mkdtempSync(path.join(os.tmpdir(), 'hermes-assert-dist-'))
-  try {
-    const result = checkDistBuilt(path.join(tempRoot, 'dist'))
-    assert.equal(result.ok, false)
-    assert.match(result.error, /no dist directory/)
-  } finally {
-    fs.rmSync(tempRoot, { recursive: true, force: true })
-  }
-})
-
-test('checkDistBuilt fails when index.html is missing', () => {
-  const { tempRoot, distDir } = makeDist(d => {
-    fs.mkdirSync(path.join(d, 'assets'))
-    fs.writeFileSync(path.join(d, 'assets', 'index-abc123.js'), 'console.log(1)', 'utf8')
-  })
-  try {
-    const result = checkDistBuilt(distDir)
-    assert.equal(result.ok, false)
-    assert.match(result.error, /index\.html is missing/)
-  } finally {
-    fs.rmSync(tempRoot, { recursive: true, force: true })
-  }
-})
-
-test('checkDistBuilt fails when index.html is empty', () => {
-  const { tempRoot, distDir } = makeDist(d => {
-    fs.writeFileSync(path.join(d, 'index.html'), '', 'utf8')
-    fs.mkdirSync(path.join(d, 'assets'))
-    fs.writeFileSync(path.join(d, 'assets', 'index-abc123.js'), 'console.log(1)', 'utf8')
-  })
-  try {
-    const result = checkDistBuilt(distDir)
-    assert.equal(result.ok, false)
-    assert.match(result.error, /index\.html is empty/)
-  } finally {
-    fs.rmSync(tempRoot, { recursive: true, force: true })
-  }
-})
-
-test('checkDistBuilt fails when assets/ has no JS bundle', () => {
-  const { tempRoot, distDir } = makeDist(d => {
-    fs.writeFileSync(path.join(d, 'index.html'), '<!doctype html>', 'utf8')
-    fs.mkdirSync(path.join(d, 'assets'))
-    // CSS only, no JS — still a blank page at runtime.
-    fs.writeFileSync(path.join(d, 'assets', 'index-abc123.css'), 'body{}', 'utf8')
-  })
-  try {
-    const result = checkDistBuilt(distDir)
-    assert.equal(result.ok, false)
-    assert.match(result.error, /no built JS bundle/)
-  } finally {
-    fs.rmSync(tempRoot, { recursive: true, force: true })
-  }
-})

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

@@ -3,9 +3,8 @@ import { useStore } from '@nanostores/react'
 import { Codicon } from '@/components/ui/codicon'
 import { Tip } from '@/components/ui/tooltip'
 import { useI18n } from '@/i18n'
-import { AlertCircle, FileText, FolderOpen, ImageIcon, Link, Loader2, Terminal } from '@/lib/icons'
+import { FileText, FolderOpen, ImageIcon, Link, Terminal } from '@/lib/icons'
 import { normalizeOrLocalPreviewTarget } from '@/lib/local-preview'
-import { cn } from '@/lib/utils'
 import type { ComposerAttachment } from '@/store/composer'
 import { notifyError } from '@/store/notifications'
 import { setCurrentSessionPreviewTarget } from '@/store/preview'
@@ -32,9 +31,7 @@ function AttachmentPill({ attachment, onRemove }: { attachment: ComposerAttachme
   const c = t.composer
   const Icon = { folder: FolderOpen, url: Link, image: ImageIcon, file: FileText, terminal: Terminal }[attachment.kind]
   const cwd = useStore($currentCwd)
-  const isUploading = attachment.uploadState === 'uploading'
-  const hasUploadError = attachment.uploadState === 'error'
-  const canPreview = attachment.kind !== 'folder' && attachment.kind !== 'terminal' && !isUploading
+  const canPreview = attachment.kind !== 'folder' && attachment.kind !== 'terminal'
   const detail = attachment.detail && attachment.detail !== attachment.label ? attachment.detail : undefined
 
   async function openPreview() {
@@ -62,15 +59,7 @@ function AttachmentPill({ attachment, onRemove }: { attachment: ComposerAttachme
         throw new Error(c.couldNotPreview(attachment.label))
       }
 
-      // We already hold the image bytes (the card thumbnail) — render those
-      // directly so a screenshot/clipboard image previews even when its only
-      // on-disk copy is a transient path the renderer can't re-read.
-      const withBytes =
-        attachment.kind === 'image' && attachment.previewUrl
-          ? { ...preview, dataUrl: attachment.previewUrl, previewKind: 'image' as const }
-          : preview
-
-      setCurrentSessionPreviewTarget(withBytes, 'manual', target)
+      setCurrentSessionPreviewTarget(preview, 'manual', target)
     } catch (error) {
       notifyError(error, c.previewUnavailable)
     }
@@ -80,51 +69,30 @@ function AttachmentPill({ attachment, onRemove }: { attachment: ComposerAttachme
     <Tip label={attachment.path || attachment.detail || attachment.label}>
       <div className="group/attachment relative min-w-0 shrink-0">
         <button
-          aria-busy={isUploading || undefined}
           aria-label={canPreview ? c.previewLabel(attachment.label) : attachment.label}
-          className={cn(
-            'flex max-w-56 items-center gap-2 rounded-2xl border bg-background/50 px-2 py-1.5 text-left shadow-[inset_0_1px_0_rgba(255,255,255,0.18)] transition-colors disabled:cursor-default',
-            hasUploadError
-              ? 'border-destructive/45 hover:border-destructive/60'
-              : 'border-border/60 hover:border-primary/35 hover:bg-accent/45'
-          )}
+          className="flex max-w-56 items-center gap-2 border border-border/60 bg-background/50 px-2 py-1.5 text-left shadow-[inset_0_1px_0_rgba(255,255,255,0.25)] transition-colors hover:border-primary/35 hover:bg-accent/45 disabled:cursor-default"
           disabled={!canPreview}
           onClick={() => void openPreview()}
           type="button"
         >
-          <span className="relative grid size-8 shrink-0 place-items-center overflow-hidden rounded-lg border border-border/55 bg-muted/35 text-muted-foreground">
-            {attachment.previewUrl && attachment.kind === 'image' ? (
-              <img
-                alt={attachment.label}
-                className="size-full object-cover"
-                draggable={false}
-                src={attachment.previewUrl}
-              />
-            ) : (
+          {attachment.previewUrl && attachment.kind === 'image' ? (
+            <img
+              alt={attachment.label}
+              className="size-8 shrink-0 border border-border/70 object-cover"
+              draggable={false}
+              src={attachment.previewUrl}
+            />
+          ) : (
+            <span className="grid size-8 shrink-0 place-items-center border border-border/55 bg-muted/35 text-muted-foreground">
               <Icon className="size-3.5" />
-            )}
-            {isUploading && (
-              <span className="absolute inset-0 grid place-items-center bg-background/60 backdrop-blur-[1px]">
-                <Loader2 className="size-3.5 animate-spin text-foreground/75" />
-              </span>
-            )}
-            {hasUploadError && (
-              <span className="absolute inset-0 grid place-items-center bg-destructive/15">
-                <AlertCircle className="size-3.5 text-destructive" />
-              </span>
-            )}
-          </span>
+            </span>
+          )}
           <span className="min-w-0">
             <span className="block truncate text-[0.72rem] font-medium leading-4 text-foreground/90">
               {attachment.label}
             </span>
             {detail && (
-              <span
-                className={cn(
-                  'block truncate text-[0.62rem] leading-3.5',
-                  hasUploadError ? 'text-destructive/80' : 'text-muted-foreground/65'
-                )}
-              >
+              <span className="block truncate font-mono text-[0.6rem] leading-3 text-muted-foreground/65">
                 {detail}
               </span>
             )}

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

@@ -3,25 +3,32 @@ import { ComposerPrimitive } from '@assistant-ui/react'
 import type { ReactNode } from 'react'
 
 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',
+  'absolute bottom-[calc(100%+0.25rem)] left-0 z-50',
+  'w-60 max-w-[calc(100vw-2rem)]',
+  'max-h-[min(23rem,calc(100vh-8rem))] overflow-y-auto overscroll-contain',
+  'rounded-lg border border-(--ui-stroke-secondary)',
+  'bg-[color-mix(in_srgb,var(--ui-bg-elevated)_96%,transparent)]',
+  'p-1 text-xs text-popover-foreground shadow-md',
   'backdrop-blur-md'
 ].join(' ')
 
 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',
+  'absolute left-0 top-[calc(100%+0.25rem)] z-50',
+  'w-60 max-w-[calc(100vw-2rem)]',
+  'max-h-[min(23rem,calc(100vh-8rem))] overflow-y-auto overscroll-contain',
+  'rounded-lg border border-(--ui-stroke-secondary)',
+  'bg-[color-mix(in_srgb,var(--ui-bg-elevated)_96%,transparent)]',
+  'p-1 text-xs text-popover-foreground shadow-md',
   'backdrop-blur-md'
 ].join(' ')
 
+export const COMPLETION_DRAWER_ROW_CLASS = [
+  'relative flex cursor-default select-none items-center gap-2 rounded-md px-2 py-1',
+  'w-full min-w-0 text-left text-xs outline-hidden transition-colors',
+  'hover:bg-(--ui-bg-tertiary)',
+  'data-[highlighted]:bg-(--ui-bg-tertiary) data-[highlighted]:text-foreground'
+].join(' ')
+
 export function ComposerCompletionDrawer({
   adapter,
   ariaLabel,

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

@@ -4,7 +4,6 @@ import { Tip } from '@/components/ui/tooltip'
 import { useI18n } from '@/i18n'
 import { triggerHaptic } from '@/lib/haptics'
 import { AudioLines, Layers3, Loader2, Square, SteeringWheel } from '@/lib/icons'
-import { formatCombo } from '@/lib/keybinds/combo'
 import { cn } from '@/lib/utils'
 
 import type { ConversationStatus } from './hooks/use-voice-conversation'
@@ -63,7 +62,6 @@ export function ComposerControls({
 }) {
   const { t } = useI18n()
   const c = t.composer
-  const steerLabel = `${c.steer} (${formatCombo('mod+enter')})`
 
   if (conversation.active) {
     return <ConversationPill {...conversation} disabled={disabled} />
@@ -75,9 +73,9 @@ 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={steerLabel}>
+        <Tip label={c.steer}>
           <Button
-            aria-label={steerLabel}
+            aria-label={c.steer}
             className={GHOST_ICON_BTN}
             disabled={disabled}
             onClick={onSteer}

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

@@ -1,189 +0,0 @@
-import { act, cleanup, fireEvent, render } from '@testing-library/react'
-import { useRef, useState } from 'react'
-import { afterEach, describe, expect, it, vi } from 'vitest'
-
-// No global setupFiles registers auto-cleanup, so unmount between tests —
-// otherwise a second render() leaks the first editor and getByTestId('editor')
-// matches multiple nodes.
-afterEach(cleanup)
-
-// Faithful mirror of index.tsx's Enter wiring (handleEditorKeyDown's Enter
-// branch + submitDraft), driven through REAL DOM keydown events on a
-// contentEditable.
-//
-// Regression repro for #39630: pressing Enter right after typing (fast typing /
-// IME) did nothing. The composer state (`draft` from useAuiState) and its
-// derived `hasComposerPayload` lag the DOM by a render, so the keydown handler
-// read empty state and either dropped the message, drained a queued prompt
-// instead of sending, or (while busy) refused to queue. The fix reads the live
-// editor text — `hasLivePayload` in the handler and a DOM re-sync at the top of
-// submitDraft — so the just-typed text always wins.
-//
-// We model the race deterministically the way the IME repro does: mutate the
-// editor's textContent WITHOUT firing an input event, so the React `draft`
-// state stays stale while the DOM already holds the text.
-function Harness({
-  busy = false,
-  queued = [],
-  onSubmit,
-  onQueue,
-  onCancel,
-  onDrain
-}: {
-  busy?: boolean
-  queued?: readonly string[]
-  onSubmit: (text: string) => void
-  onQueue: (text: string) => void
-  onCancel: () => void
-  onDrain: () => void
-}) {
-  const editorRef = useRef<HTMLDivElement>(null)
-  const draftRef = useRef('')
-  // Mirrors `useAuiState(s => s.composer.text)` — updated only via setText, so
-  // it lags the DOM until React re-renders (the source of the bug).
-  const [draft, setDraft] = useState('')
-  const attachments: unknown[] = []
-
-  const composerPlainText = (el: HTMLElement) => el.textContent ?? ''
-
-  const setText = (next: string) => {
-    draftRef.current = next
-    setDraft(next)
-  }
-
-  const submitDraft = () => {
-    const editor = editorRef.current
-    if (editor) {
-      const domText = composerPlainText(editor)
-      if (domText !== draftRef.current) {
-        draftRef.current = domText
-        setDraft(domText)
-      }
-    }
-
-    const text = draftRef.current
-    const payloadPresent = text.trim().length > 0 || attachments.length > 0
-
-    if (busy) {
-      if (payloadPresent) {
-        onQueue(text)
-      } else {
-        onCancel()
-      }
-    } else if (!payloadPresent && queued.length > 0) {
-      onDrain()
-    } else if (payloadPresent) {
-      onSubmit(text)
-    }
-  }
-
-  const handleKeyDown = (event: React.KeyboardEvent<HTMLDivElement>) => {
-    if (event.key === 'Enter' && !event.shiftKey) {
-      event.preventDefault()
-
-      const editorText = editorRef.current ? composerPlainText(editorRef.current) : draftRef.current
-      const hasLivePayload = editorText.trim().length > 0 || attachments.length > 0
-
-      if (!busy && !hasLivePayload && queued.length > 0) {
-        onDrain()
-
-        return
-      }
-
-      if (busy && !hasLivePayload) {
-        return
-      }
-
-      submitDraft()
-    }
-  }
-
-  // `draft` is read so the lint/compiler treats the stale-state mirror as live;
-  // the assertions prove the handler never relies on it.
-  void draft
-
-  return (
-    <div
-      contentEditable
-      data-testid="editor"
-      onInput={event => setText(composerPlainText(event.currentTarget))}
-      onKeyDown={handleKeyDown}
-      ref={editorRef}
-      suppressContentEditableWarning
-    />
-  )
-}
-
-describe('composer Enter submit — live DOM vs stale composer state (#39630)', () => {
-  it('sends the just-typed text on Enter even when composer state has not synced', async () => {
-    const onSubmit = vi.fn()
-    const { getByTestId } = render(
-      <Harness onCancel={vi.fn()} onDrain={vi.fn()} onQueue={vi.fn()} onSubmit={onSubmit} />
-    )
-    const editor = getByTestId('editor')
-
-    // Fast typing: the DOM has the text but NO input event fired, so `draft`
-    // state is still empty (the exact stale-state race).
-    await act(async () => {
-      editor.textContent = 'hello world'
-      fireEvent.keyDown(editor, { key: 'Enter' })
-    })
-
-    expect(onSubmit).toHaveBeenCalledWith('hello world')
-  })
-
-  it('queues a fast-typed message while busy instead of draining the queue or cancelling', async () => {
-    const onQueue = vi.fn()
-    const onDrain = vi.fn()
-    const onCancel = vi.fn()
-    const { getByTestId } = render(
-      <Harness busy onCancel={onCancel} onDrain={onDrain} onQueue={onQueue} onSubmit={vi.fn()} queued={['queued-1']} />
-    )
-    const editor = getByTestId('editor')
-
-    await act(async () => {
-      editor.textContent = 'urgent follow-up'
-      fireEvent.keyDown(editor, { key: 'Enter' })
-    })
-
-    expect(onQueue).toHaveBeenCalledWith('urgent follow-up')
-    expect(onDrain).not.toHaveBeenCalled()
-    expect(onCancel).not.toHaveBeenCalled()
-  })
-
-  it('treats an empty Enter while busy as a no-op (never an accidental Stop)', async () => {
-    const onCancel = vi.fn()
-    const onSubmit = vi.fn()
-    const onQueue = vi.fn()
-    const { getByTestId } = render(
-      <Harness busy onCancel={onCancel} onDrain={vi.fn()} onQueue={onQueue} onSubmit={onSubmit} />
-    )
-    const editor = getByTestId('editor')
-
-    await act(async () => {
-      editor.textContent = ''
-      fireEvent.keyDown(editor, { key: 'Enter' })
-    })
-
-    expect(onCancel).not.toHaveBeenCalled()
-    expect(onSubmit).not.toHaveBeenCalled()
-    expect(onQueue).not.toHaveBeenCalled()
-  })
-
-  it('drains the next queued prompt on Enter when idle with a truly empty editor', async () => {
-    const onDrain = vi.fn()
-    const onSubmit = vi.fn()
-    const { getByTestId } = render(
-      <Harness onCancel={vi.fn()} onDrain={onDrain} onQueue={vi.fn()} onSubmit={onSubmit} queued={['queued-1']} />
-    )
-    const editor = getByTestId('editor')
-
-    await act(async () => {
-      editor.textContent = ''
-      fireEvent.keyDown(editor, { key: 'Enter' })
-    })
-
-    expect(onDrain).toHaveBeenCalledTimes(1)
-    expect(onSubmit).not.toHaveBeenCalled()
-  })
-})

apps/desktop/src/app/chat/composer/hooks/use-live-completion-adapter.ts

@@ -5,13 +5,6 @@ export interface CompletionEntry {
   text: string
   display?: unknown
   meta?: unknown
-  /** Optional section label (e.g. "Commands", "Skills"). The popover renders a
-   *  header whenever this changes between consecutive items, so the fetcher must
-   *  emit entries already grouped contiguously. */
-  group?: string
-  /** Optional completion-action id. When set, picking the item runs that action
-   *  (e.g. opening an overlay) instead of inserting a chip + waiting for submit. */
-  action?: string
 }
 
 export interface CompletionPayload {

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

@@ -2,17 +2,12 @@ import type { Unstable_TriggerAdapter, Unstable_TriggerItem } from '@assistant-u
 import { useCallback } from 'react'
 
 import type { HermesGateway } from '@/hermes'
-import { sessionTitle } from '@/lib/chat-runtime'
 import {
   type CommandsCatalogLike,
-  desktopSkinSlashCompletions,
   desktopSlashDescription,
-  type DesktopThemeCommandOption,
   filterDesktopCommandsCatalog,
-  isDesktopSlashExtensionCommand,
   isDesktopSlashSuggestion
 } from '@/lib/desktop-slash-commands'
-import { $sessions } from '@/store/session'
 
 import type { CompletionEntry, CompletionPayload } from './use-live-completion-adapter'
 import { useLiveCompletionAdapter } from './use-live-completion-adapter'
@@ -21,10 +16,7 @@ interface SlashItemMetadata extends Record<string, string> {
   command: string
   display: string
   meta: string
-  group: string
   rawText: string
-  /** Completion-action id; empty for ordinary insert-a-chip completions. */
-  action: string
 }
 
 function textValue(value: unknown, fallback = ''): string {
@@ -46,21 +38,12 @@ function commandText(value: string): string {
   return value.startsWith('/') ? value : `/${value}`
 }
 
-/** How many recent sessions to surface inline before the "Browse all…" entry. */
-const SESSION_INLINE_LIMIT = 7
-
 /** Live `/` completions backed by the gateway's `complete.slash` RPC. */
-export function useSlashCompletions(options: {
-  gateway: HermesGateway | null
-  /** Desktop theme list — `/skin` is owned client-side, so its arg completions
-   *  come from here, not the backend (whose skin list is CLI/TUI-only). */
-  skinThemes?: DesktopThemeCommandOption[]
-  activeSkin?: string
-}): {
+export function useSlashCompletions(options: { gateway: HermesGateway | null }): {
   adapter: Unstable_TriggerAdapter
   loading: boolean
 } {
-  const { gateway, skinThemes, activeSkin } = options
+  const { gateway } = options
   const enabled = Boolean(gateway)
 
   const fetcher = useCallback(
@@ -71,136 +54,34 @@ export function useSlashCompletions(options: {
 
       const text = `/${query}`
 
-      // The desktop owns /skin entirely (client-side theme context). Surface its
-      // theme list inside this single popover instead of a bespoke one, and skip
-      // the backend skin completions (which describe CLI/TUI skins that don't
-      // apply here). Matches once we're past `/skin ` into the arg stage.
-      const skinArg = /^\/skin\s+(.*)$/is.exec(text)
-
-      if (skinArg && skinThemes) {
-        const items = desktopSkinSlashCompletions(skinThemes, activeSkin ?? '', skinArg[1] ?? '').map(entry => ({
-          text: entry.text,
-          display: entry.display,
-          meta: entry.meta,
-          group: 'Themes'
-        }))
-
-        return { items, query }
-      }
-
-      // /resume (and its aliases) completes recent sessions inline — the same
-      // client-side list the picker overlay shows — instead of the backend
-      // (whose /resume opens an interactive TUI picker we can't render here).
-      const sessionArg = /^\/(?:resume|sessions|switch)\s+(.*)$/is.exec(text)
-
-      if (sessionArg) {
-        const needle = (sessionArg[1] ?? '').trim().toLowerCase()
-
-        const matches = (
-          needle
-            ? $sessions.get().filter(
-                session =>
-                  sessionTitle(session).toLowerCase().includes(needle) ||
-                  (session.preview ?? '').toLowerCase().includes(needle) ||
-                  session.id.toLowerCase().includes(needle)
-              )
-            : $sessions.get()
-        ).slice(0, SESSION_INLINE_LIMIT)
-
-        const items: CompletionEntry[] = matches.map(session => ({
-          text: `/resume ${session.id}`,
-          display: sessionTitle(session),
-          meta: (session.preview ?? '').trim(),
-          group: 'Sessions'
-        }))
-
-        // Trailing "more" affordance (Cursor-style): picking it opens the full
-        // session picker overlay directly. `text` stays a bare `/resume` so that
-        // submitting it (Enter) still opens the overlay if the action is skipped.
-        items.push({
-          text: '/resume',
-          display: 'Browse all sessions…',
-          meta: '',
-          group: 'Sessions',
-          action: 'session-picker'
-        })
-
-        return { items, query }
-      }
-
       try {
         if (!query) {
           const catalog = filterDesktopCommandsCatalog(await gateway.request<CommandsCatalogLike>('commands.catalog'))
 
-          // Prefer the categorized layout so the popover renders section headers
-          // (Session, Tools & Skills, ...). Fall back to the flat list when the
-          // backend didn't categorize.
-          const sections = catalog.categories?.length
-            ? catalog.categories
-            : [{ name: '', pairs: catalog.pairs ?? [] }]
-
-          const items = sections.flatMap(section =>
-            section.pairs.map(([command, meta]) => ({
-              text: command,
-              display: command,
-              group: section.name || undefined,
-              meta
-            }))
-          )
+          const items = (catalog.pairs ?? []).map(([command, meta]) => ({
+            text: command,
+            display: command,
+            meta
+          }))
 
           return { items, query }
         }
 
-        const result = await gateway.request<{ items?: CompletionEntry[]; replace_from?: number }>(
-          'complete.slash',
-          { text }
-        )
+        const result = await gateway.request<{ items?: CompletionEntry[] }>('complete.slash', { text })
 
-        // Arg-completion items (replace_from > 1) carry just the arg stub —
-        // e.g. complete.slash returns `{text: "alice"}` for `/personality alic`
-        // with replace_from = 14. Rewrite those entries so the popover inserts
-        // the full `/personality alice` token instead of stranding `/alice`.
-        const replaceFrom = typeof result.replace_from === 'number' ? result.replace_from : 1
-        const isArgCompletion = replaceFrom > 1
-        const prefix = isArgCompletion ? text.slice(0, replaceFrom) : ''
-
-        const decorated = (result.items ?? [])
-          .map(item => {
-            if (!isArgCompletion) {
-              return item
-            }
-
-            const argText = typeof item.text === 'string' ? item.text : ''
-
-            return { ...item, text: `${prefix}${argText}` }
-          })
-          .filter(item => isArgCompletion || isDesktopSlashSuggestion(item.text))
+        const items = (result.items ?? [])
+          .filter(item => isDesktopSlashSuggestion(item.text))
           .map(item => ({
             ...item,
-            // Arg suggestions (e.g. `/handoff <platform>`) live under one
-            // header; otherwise split skills out from built-in commands.
-            group: isArgCompletion ? 'Options' : isDesktopSlashExtensionCommand(item.text) ? 'Skills' : 'Commands',
-            // Arg items carry their own meta (the personality/toolset/platform
-            // blurb). Only command rows get the registry description — looking
-            // one up for `/personality none` would clobber it with the parent
-            // command's text.
-            meta: isArgCompletion ? textValue(item.meta) : desktopSlashDescription(item.text, textValue(item.meta))
+            meta: desktopSlashDescription(item.text, textValue(item.meta))
           }))
 
-        // Keep each group contiguous so headers render once: Commands before
-        // Skills (stable within a group, preserving backend relevance order).
-        const groupOrder = ['Commands', 'Skills', 'Options']
-
-        const items = isArgCompletion
-          ? decorated
-          : [...decorated].sort((a, b) => groupOrder.indexOf(a.group) - groupOrder.indexOf(b.group))
-
         return { items, query }
       } catch {
         return { items: [], query }
       }
     },
-    [gateway, skinThemes, activeSkin]
+    [gateway]
   )
 
   const toItem = useCallback((entry: CompletionEntry, index: number): Unstable_TriggerItem => {
@@ -212,8 +93,6 @@ export function useSlashCompletions(options: {
       command,
       display,
       meta,
-      group: textValue(entry.group),
-      action: textValue(entry.action),
       // Provide rawText so hermesDirectiveFormatter.serialize uses the
       // direct-insertion path instead of the legacy @type:id fallback.
       // Without this, the item.id (which includes a "|index" suffix for

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

@@ -13,25 +13,17 @@ import {
   useState
 } from 'react'
 
-import { hermesDirectiveFormatter, type SlashChipKind } from '@/components/assistant-ui/directive-text'
+import { hermesDirectiveFormatter } from '@/components/assistant-ui/directive-text'
 import { Button } from '@/components/ui/button'
 import { useMediaQuery } from '@/hooks/use-media-query'
 import { useResizeObserver } from '@/hooks/use-resize-observer'
 import { useI18n } from '@/i18n'
 import { chatMessageText } from '@/lib/chat-messages'
 import { SLASH_COMMAND_RE } from '@/lib/chat-runtime'
-import { desktopSlashCommandTakesArgs } from '@/lib/desktop-slash-commands'
 import { DATA_IMAGE_URL_RE } from '@/lib/embedded-images'
 import { triggerHaptic } from '@/lib/haptics'
 import { cn } from '@/lib/utils'
-import {
-  $composerAttachments,
-  clearComposerAttachments,
-  clearSessionDraft,
-  type ComposerAttachment,
-  stashSessionDraft,
-  takeSessionDraft
-} from '@/store/composer'
+import { $composerAttachments, clearComposerAttachments, type ComposerAttachment } from '@/store/composer'
 import {
   browseBackward,
   browseForward,
@@ -48,11 +40,10 @@ import {
   shouldAutoDrainOnSettle,
   updateQueuedPrompt
 } from '@/store/composer-queue'
-import { $gatewayState, $messages, setSessionPickerOpen } from '@/store/session'
+import { $gatewayState, $messages } from '@/store/session'
 import { $threadScrolledUp } from '@/store/thread-scroll'
-import { useTheme } from '@/themes'
 
-import { extractDroppedFiles, HERMES_PATHS_MIME, partitionDroppedFiles } from '../hooks/use-composer-actions'
+import { extractDroppedFiles, HERMES_PATHS_MIME } from '../hooks/use-composer-actions'
 
 import { AttachmentList } from './attachments'
 import { ContextMenu } from './context-menu'
@@ -73,7 +64,7 @@ import { useVoiceConversation } from './hooks/use-voice-conversation'
 import { useVoiceRecorder } from './hooks/use-voice-recorder'
 import {
   dragHasAttachments,
-  droppedFileInlineRefs,
+  droppedFileInlineRef,
   type InlineRefInput,
   insertInlineRefsIntoEditor
 } from './inline-refs'
@@ -83,9 +74,9 @@ import {
   placeCaretEnd,
   refChipElement,
   renderComposerContents,
-  RICH_INPUT_SLOT,
-  slashChipElement
+  RICH_INPUT_SLOT
 } from './rich-editor'
+import { SkinSlashPopover } from './skin-slash-popover'
 import { detectTrigger, extractClipboardImageBlobs, textBeforeCaret, type TriggerState } from './text-utils'
 import { ComposerTriggerPopover } from './trigger-popover'
 import type { ChatBarProps } from './types'
@@ -104,30 +95,6 @@ const COMPOSER_FADE_BACKGROUND =
 
 const pickPlaceholder = (pool: readonly string[]) => pool[Math.floor(Math.random() * pool.length)]
 
-/** Completion items can carry an `action` (set in use-slash-completions) that
- *  runs a side effect on pick instead of inserting a chip — e.g. the session
- *  picker's "Browse all…" entry opens the overlay. Table-driven so new action
- *  items are a registry row, not a composer branch. */
-const COMPLETION_ACTIONS: Record<string, () => void> = {
-  'session-picker': () => setSessionPickerOpen(true)
-}
-
-/** Map a picked `/` completion to its pill accent. Driven by the completion
- *  group set in use-slash-completions (Skills / Themes / Commands|Options). */
-function slashChipKindForItem(item: Unstable_TriggerItem): SlashChipKind {
-  const group = (item.metadata as { group?: unknown } | undefined)?.group
-
-  if (group === 'Skills') {
-    return 'skill'
-  }
-
-  if (group === 'Themes') {
-    return 'theme'
-  }
-
-  return 'command'
-}
-
 interface QueueEditState {
   attachments: ComposerAttachment[]
   draft: string
@@ -137,10 +104,6 @@ interface QueueEditState {
 
 const cloneAttachments = (attachments: ComposerAttachment[]) => attachments.map(a => ({ ...a }))
 
-// Quiet period after the last keystroke before persisting the draft;
-// unmount/pagehide flushes bypass it.
-const DRAFT_PERSIST_DEBOUNCE_MS = 400
-
 export function ChatBar({
   busy,
   cwd,
@@ -182,9 +145,6 @@ export function ChatBar({
   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 drainingQueueRef = useRef(false)
   const urlInputRef = useRef<HTMLInputElement | null>(null)
 
@@ -196,17 +156,14 @@ export function ChatBar({
   const [dragActive, setDragActive] = useState(false)
   const [queueEdit, setQueueEdit] = useState<QueueEditState | null>(null)
   const [focusRequestId, setFocusRequestId] = useState(0)
-  const queueEditRef = useRef(queueEdit)
-  queueEditRef.current = queueEdit
   const dragDepthRef = useRef(0)
   const composingRef = useRef(false) // true during IME composition (CJK input)
   const lastSpokenIdRef = useRef<string | null>(null)
 
   const narrow = useMediaQuery('(max-width: 30rem)')
 
-  const { availableThemes, themeName } = useTheme()
   const at = useAtCompletions({ gateway: gateway ?? null, sessionId: sessionId ?? null, cwd: cwd ?? null })
-  const slash = useSlashCompletions({ activeSkin: themeName, gateway: gateway ?? null, skinThemes: availableThemes })
+  const slash = useSlashCompletions({ gateway: gateway ?? null })
 
   const stacked = expanded || narrow || tight
   const trimmedDraft = draft.trim()
@@ -214,12 +171,10 @@ export function ChatBar({
   const canSubmit = busy || hasComposerPayload
   const editingQueuedPrompt = queueEdit ? (queuedPrompts.find(entry => entry.id === queueEdit.entryId) ?? null) : null
   const busyAction = busy && hasComposerPayload ? 'queue' : 'stop'
-
   // Steer only makes sense mid-turn, text-only (the gateway can't carry images
   // into a tool result) and never for a slash command (those execute inline).
   const canSteer =
     busy && !!onSteer && attachments.length === 0 && trimmedDraft.length > 0 && !SLASH_COMMAND_RE.test(trimmedDraft)
-
   const showHelpHint = draft === '?'
 
   const { t } = useI18n()
@@ -507,6 +462,12 @@ export function ChatBar({
     })
   }, [])
 
+  const selectSkinSlashCommand = (command: string) => {
+    draftRef.current = command
+    aui.composer().setText(command)
+    requestMainFocus()
+  }
+
   const handlePaste = (event: ClipboardEvent<HTMLDivElement>) => {
     const imageBlobs = extractClipboardImageBlobs(event.clipboardData)
 
@@ -659,50 +620,16 @@ export function ChatBar({
       return
     }
 
-    // Action items (e.g. "Browse all sessions…") run a side effect instead of
-    // inserting a chip: strip the typed trigger token, then fire the action.
-    const completionAction = (item.metadata as { action?: unknown } | undefined)?.action
-    const runAction = typeof completionAction === 'string' ? COMPLETION_ACTIONS[completionAction] : undefined
-
-    if (runAction) {
-      const current = composerPlainText(editor)
-      const prefix = current.slice(0, Math.max(0, current.length - trigger.tokenLength))
-
-      renderComposerContents(editor, prefix)
-      placeCaretEnd(editor)
-      draftRef.current = composerPlainText(editor)
-      aui.composer().setText(draftRef.current)
-      closeTrigger()
-      runAction()
-      requestMainFocus()
-
-      return
-    }
-
     const serialized = hermesDirectiveFormatter.serialize(item)
     const starter = serialized.endsWith(':')
-
-    // Picking a bare arg-taking command (e.g. `/personality`) shouldn't commit
-    // it — expand to its options step so the popover shows the inline list, just
-    // as typing `/personality ` by hand would. A serialized value with a space is
-    // 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 text = starter || serialized.endsWith(' ') ? serialized : `${serialized} `
     const directive = !starter && serialized.match(/^@([^:]+):(.+)$/)
-    // No pill while expanding — the bare command stays plain text until an arg
-    // is picked, at which point a single pill is emitted for the full command.
-    const slashKind = !expandsToArgs && trigger.kind === '/' ? slashChipKindForItem(item) : null
-    const keepTriggerOpen = starter || expandsToArgs
 
     const finish = () => {
       draftRef.current = composerPlainText(editor)
       aui.composer().setText(draftRef.current)
       requestMainFocus()
-      keepTriggerOpen ? window.setTimeout(refreshTrigger, 0) : closeTrigger()
+      starter ? window.setTimeout(refreshTrigger, 0) : closeTrigger()
     }
 
     const sel = window.getSelection()
@@ -712,20 +639,7 @@ export function ChatBar({
 
     if (!sel || !range || node?.nodeType !== Node.TEXT_NODE || offset < trigger.tokenLength) {
       const current = composerPlainText(editor)
-      const prefix = current.slice(0, Math.max(0, current.length - trigger.tokenLength))
-
-      if (slashKind) {
-        // Two-step arg picks (e.g. `/handoff` pill already inserted, now picking
-        // the platform) land here because the caret sits past a contenteditable
-        // chip. Rebuild the prefix and re-emit a single pill for the full command.
-        renderComposerContents(editor, prefix)
-        editor.append(slashChipElement(serialized, slashKind), document.createTextNode(' '))
-        placeCaretEnd(editor)
-
-        return finish()
-      }
-
-      renderComposerContents(editor, `${prefix}${text}`)
+      renderComposerContents(editor, `${current.slice(0, Math.max(0, current.length - trigger.tokenLength))}${text}`)
       placeCaretEnd(editor)
 
       return finish()
@@ -736,13 +650,8 @@ export function ChatBar({
     replaceRange.setEnd(node, offset)
     replaceRange.deleteContents()
 
-    const chip = slashKind
-      ? slashChipElement(serialized, slashKind)
-      : directive
-        ? refChipElement(directive[1], directive[2])
-        : null
-
-    if (chip) {
+    if (directive) {
+      const chip = refChipElement(directive[1], directive[2])
       const space = document.createTextNode(' ')
       const fragment = document.createDocumentFragment()
       fragment.append(chip, space)
@@ -905,16 +814,7 @@ export function ChatBar({
     if (event.key === 'Enter' && !event.shiftKey) {
       event.preventDefault()
 
-      // Decide from the DOM, not React state. `hasComposerPayload` is derived
-      // from the AUI composer state, which lags the latest keystroke by a
-      // render, so on fast typing / IME the just-typed text isn't in state yet.
-      // Without the live read, a real message typed while prompts are queued
-      // would drain the queue instead of sending. submitDraft() re-syncs and
-      // sends the live editor text.
-      const editorText = editorRef.current ? composerPlainText(editorRef.current) : draftRef.current
-      const hasLivePayload = editorText.trim().length > 0 || attachments.length > 0
-
-      if (!busy && !hasLivePayload && queuedPrompts.length > 0) {
+      if (!busy && !hasComposerPayload && queuedPrompts.length > 0) {
         void drainNextQueued()
 
         return
@@ -922,10 +822,7 @@ export function ChatBar({
 
       // Empty Enter while busy is a no-op — interrupting is explicit (Stop/Esc),
       // never a stray Enter after sending. With a payload, submitDraft queues it.
-      // Gate on the live DOM payload (not the render-lagged composer state) so a
-      // message typed fast / via IME while busy still reaches submitDraft() and
-      // gets queued instead of being mistaken for an empty Enter.
-      if (busy && !hasLivePayload) {
+      if (busy && !hasComposerPayload) {
         return
       }
 
@@ -1022,25 +919,24 @@ export function ChatBar({
       return
     }
 
-    // In-app drags (project tree / gutter) are workspace-relative paths the
-    // gateway resolves directly, so they stay inline @file:/@line: refs. OS
-    // drops are absolute local paths a remote gateway can't read (and images
-    // need byte upload for vision), so route them through the upload pipeline.
-    const { inAppRefs, osDrops } = partitionDroppedFiles(candidates)
-    const refs = droppedFileInlineRefs(inAppRefs, cwd)
+    if (Array.from(event.dataTransfer.types || []).includes(HERMES_PATHS_MIME)) {
+      const refs = candidates
+        .map(candidate => droppedFileInlineRef(candidate, cwd))
+        .filter((ref): ref is string => Boolean(ref))
 
-    if (refs.length && insertInlineRefs(refs)) {
-      triggerHaptic('selection')
+      if (insertInlineRefs(refs)) {
+        triggerHaptic('selection')
+      }
+
+      return
     }
 
-    if (osDrops.length) {
-      void Promise.resolve(onAttachDroppedItems(osDrops)).then(attached => {
-        if (attached) {
-          triggerHaptic('selection')
-          requestMainFocus()
-        }
-      })
-    }
+    void Promise.resolve(onAttachDroppedItems(candidates)).then(attached => {
+      if (attached) {
+        triggerHaptic('selection')
+        requestMainFocus()
+      }
+    })
   }
 
   const handleInputDragOver = (event: ReactDragEvent<HTMLDivElement>) => {
@@ -1060,7 +956,11 @@ export function ChatBar({
 
     const candidates = extractDroppedFiles(event.dataTransfer)
 
-    if (!candidates.length) {
+    const refs = candidates
+      .map(candidate => droppedFileInlineRef(candidate, cwd))
+      .filter((ref): ref is string => Boolean(ref))
+
+    if (!refs.length) {
       return
     }
 
@@ -1068,27 +968,9 @@ export function ChatBar({
     event.stopPropagation()
     resetDragState()
 
-    // Dropping straight onto the text box used to inline-ref *every* file —
-    // including OS/Finder drops, whose absolute local path a remote gateway
-    // can't read and whose image bytes never reached vision. Split by origin:
-    // in-app drags stay inline refs; OS drops go through the upload pipeline.
-    // (When no upload handler is wired, fall back to inline refs for all.)
-    const attach = onAttachDroppedItems
-    const { inAppRefs, osDrops } = partitionDroppedFiles(candidates)
-    const refs = droppedFileInlineRefs(attach ? inAppRefs : candidates, cwd)
-
-    if (refs.length && insertInlineRefs(refs)) {
+    if (insertInlineRefs(refs)) {
       triggerHaptic('selection')
     }
-
-    if (attach && osDrops.length) {
-      void Promise.resolve(attach(osDrops)).then(attached => {
-        if (attached) {
-          triggerHaptic('selection')
-          requestMainFocus()
-        }
-      })
-    }
   }
 
   const clearDraft = useCallback(() => {
@@ -1113,69 +995,6 @@ export function ChatBar({
     }
   }
 
-  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
-  // on enter. Keyed writes are idempotent, so no skip-sentinel.
-  useEffect(() => {
-    const { attachments, text } = takeSessionDraft(activeQueueSessionKey)
-    loadIntoComposer(text, attachments)
-
-    return () => {
-      const editing = queueEditRef.current
-
-      if (editing?.sessionKey === activeQueueSessionKey) {
-        stashAt(activeQueueSessionKey, editing.draft, editing.attachments)
-      } else if (!isBrowsingHistory(sessionId)) {
-        stashAt(activeQueueSessionKey)
-      }
-    }
-  }, [activeQueueSessionKey]) // eslint-disable-line react-hooks/exhaustive-deps
-
-  // Debounced stash into the active scope. Skipped while browsing history or
-  // editing a queued prompt — recalled text must not clobber the real draft.
-  useEffect(() => {
-    if (isBrowsingHistory(sessionId) || queueEdit) {
-      return
-    }
-
-    pendingDraftPersistRef.current = { scope: activeQueueSessionKey, text: draft }
-
-    const handle = window.setTimeout(() => {
-      pendingDraftPersistRef.current = null
-      stashAt(activeQueueSessionKey, draft)
-    }, DRAFT_PERSIST_DEBOUNCE_MS)
-
-    return () => window.clearTimeout(handle)
-  }, [activeQueueSessionKey, draft, queueEdit, sessionId])
-
-  // pagehide is load-bearing: React skips effect cleanups on reload, so Cmd+R
-  // inside the debounce window would drop trailing keystrokes without this.
-  useEffect(() => {
-    const flushPendingDraftPersist = () => {
-      const pending = pendingDraftPersistRef.current
-
-      if (!pending) {
-        return
-      }
-
-      pendingDraftPersistRef.current = null
-      stashAt(pending.scope, pending.text)
-    }
-
-    window.addEventListener('pagehide', flushPendingDraftPersist)
-
-    return () => {
-      window.removeEventListener('pagehide', flushPendingDraftPersist)
-      flushPendingDraftPersist()
-    }
-  }, [])
-
   const beginQueuedEdit = (entry: QueuedPromptEntry) => {
     if (!activeQueueSessionKey || queueEdit) {
       return
@@ -1378,61 +1197,21 @@ export function ChatBar({
     }
   }, [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.
+  // Clean up queue edit when its target disappears (session swap or external delete).
   useEffect(() => {
     if (!queueEdit) {
       return
     }
 
-    if (queueEdit.sessionKey === activeQueueSessionKey) {
-      if (editingQueuedPrompt) {
-        return
-      }
-
-      loadIntoComposer(queueEdit.draft, queueEdit.attachments)
+    if (queueEdit.sessionKey === activeQueueSessionKey && editingQueuedPrompt) {
+      return
     }
 
+    loadIntoComposer(queueEdit.draft, queueEdit.attachments)
     setQueueEdit(null)
   }, [activeQueueSessionKey, editingQueuedPrompt, queueEdit]) // eslint-disable-line react-hooks/exhaustive-deps
 
-  const dispatchSubmit = (text: string, attachments?: ComposerAttachment[]) => {
-    const submittedScope = activeQueueSessionKeyRef.current
-    const submittedAttachments = attachments ?? []
-
-    const restore = () => {
-      loadIntoComposer(text, submittedAttachments)
-      stashAt(activeQueueSessionKeyRef.current, text, submittedAttachments)
-    }
-
-    void Promise.resolve(attachments ? onSubmit(text, { attachments }) : onSubmit(text))
-      .then(accepted => void (accepted === false ? restore() : clearSessionDraft(submittedScope)))
-      .catch(restore)
-  }
-
   const submitDraft = () => {
-    // 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
-    // not have synced yet — reading state here drops the message (Enter looks
-    // like it does nothing; typing a trailing space only "fixes" it because the
-    // extra input event forces a state sync). draftRef is updated on every
-    // input event; refresh it from the editor once more to also cover an
-    // in-flight keystroke that hasn't fired its input event yet.
-    const editor = editorRef.current
-
-    if (editor) {
-      const domText = composerPlainText(editor)
-
-      if (domText !== draftRef.current) {
-        draftRef.current = domText
-        aui.composer().setText(domText)
-      }
-    }
-
-    const text = draftRef.current
-    const payloadPresent = text.trim().length > 0 || attachments.length > 0
-
     if (queueEdit) {
       exitQueuedEdit('save')
     } else if (busy) {
@@ -1443,11 +1222,12 @@ export function ChatBar({
       // busy guard for commands that genuinely need an idle session (skill
       // /send directives).  Queuing them would make every slash command wait
       // for the current turn to finish, which is how the TUI never behaves.
-      if (!attachments.length && SLASH_COMMAND_RE.test(text.trim())) {
+      if (!attachments.length && SLASH_COMMAND_RE.test(draft.trim())) {
+        const submitted = draft
         triggerHaptic('submit')
         clearDraft()
-        dispatchSubmit(text)
-      } else if (payloadPresent) {
+        void onSubmit(submitted)
+      } else if (hasComposerPayload) {
         queueCurrentDraft()
       } else {
         // Stop button (the only way to reach here while busy with an empty
@@ -1455,15 +1235,15 @@ export function ChatBar({
         triggerHaptic('cancel')
         void Promise.resolve(onCancel())
       }
-    } else if (!payloadPresent && queuedPrompts.length > 0) {
+    } else if (!hasComposerPayload && queuedPrompts.length > 0) {
       void drainNextQueued()
-    } else if (payloadPresent) {
-      const submittedAttachments = cloneAttachments(attachments)
+    } else if (draft.trim() || attachments.length > 0) {
+      const submitted = draft
       triggerHaptic('submit')
       resetBrowseState(sessionId)
       clearDraft()
       clearComposerAttachments()
-      dispatchSubmit(text, submittedAttachments)
+      void onSubmit(submitted, { attachments })
     }
 
     focusInput()
@@ -1688,6 +1468,7 @@ export function ChatBar({
               onPick={replaceTriggerWithChip}
             />
           )}
+          <SkinSlashPopover draft={draft} onSelect={selectSkinSlashCommand} />
           {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

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

@@ -83,12 +83,6 @@ export function droppedFileInlineRef(candidate: DroppedFile, cwd: string | null
   return `@${kind}:${formatRefValue(rel)}`
 }
 
-/** Resolve a batch of drops to their inline `@file:`/`@line:`/`@folder:` refs,
- * dropping any that carry no path. */
-export function droppedFileInlineRefs(candidates: DroppedFile[], cwd: string | null | undefined): string[] {
-  return candidates.map(candidate => droppedFileInlineRef(candidate, cwd)).filter((ref): ref is string => Boolean(ref))
-}
-
 export function insertInlineRefsIntoEditor(editor: HTMLDivElement, refs: readonly InlineRefInput[]) {
   if (!refs.length) {
     return null

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

@@ -10,10 +10,7 @@ import {
   DIRECTIVE_CHIP_CLASS,
   directiveIconElement,
   directiveIconSvg,
-  formatRefValue,
-  slashChipClass,
-  type SlashChipKind,
-  slashIconElement
+  formatRefValue
 } from '@/components/assistant-ui/directive-text'
 
 export const RICH_INPUT_SLOT = 'composer-rich-input'
@@ -80,24 +77,6 @@ export function refChipElement(kind: string, rawValue: string, displayLabel?: st
   return chip
 }
 
-/** A non-editable pill for a picked slash command (`/skin nous`, `/tropes`).
- *  `data-ref-text` carries the literal command so `composerPlainText` round-trips
- *  it back to the exact text that gets submitted. */
-export function slashChipElement(command: string, kind: SlashChipKind, label?: string) {
-  const chip = document.createElement('span')
-  const text = document.createElement('span')
-
-  chip.contentEditable = 'false'
-  chip.dataset.refText = command
-  chip.dataset.slashKind = kind
-  chip.className = slashChipClass(kind)
-  text.className = 'truncate'
-  text.textContent = label || command
-  chip.append(slashIconElement(kind), text)
-
-  return chip
-}
-
 function appendTextWithBreaks(target: DocumentFragment | HTMLElement, text: string) {
   const lines = text.split('\n')
 

apps/desktop/src/app/chat/composer/skin-slash-popover.tsx

@@ -0,0 +1,61 @@
+import { useI18n } from '@/i18n'
+import { desktopSkinSlashCompletions } from '@/lib/desktop-slash-commands'
+import { triggerHaptic } from '@/lib/haptics'
+import { useTheme } from '@/themes/context'
+
+import { COMPLETION_DRAWER_CLASS, COMPLETION_DRAWER_ROW_CLASS, CompletionDrawerEmpty } from './completion-drawer'
+
+interface SkinSlashPopoverProps {
+  draft: string
+  onSelect: (command: string) => void
+}
+
+export function SkinSlashPopover({ draft, onSelect }: SkinSlashPopoverProps) {
+  const { t } = useI18n()
+  const c = t.composer
+  const { availableThemes, themeName } = useTheme()
+  const match = draft.match(/^\/skin\s+(\S*)$/i)
+
+  if (!match) {
+    return null
+  }
+
+  const items = desktopSkinSlashCompletions(availableThemes, themeName, match[1] ?? '')
+
+  return (
+    <div
+      aria-label={c.themeSuggestions}
+      className={COMPLETION_DRAWER_CLASS}
+      data-slot="composer-skin-completion-drawer"
+      data-state="open"
+      role="listbox"
+    >
+      <div className="grid gap-0.5 pt-0.5">
+        {items.length === 0 ? (
+          <CompletionDrawerEmpty title={c.noMatchingThemes}>
+            {c.themeTryPre}
+            <span className="font-mono text-foreground/80">/skin list</span>
+            {c.themeTryPost}
+          </CompletionDrawerEmpty>
+        ) : (
+          items.map(item => (
+            <button
+              className={COMPLETION_DRAWER_ROW_CLASS}
+              key={item.text}
+              onClick={() => {
+                triggerHaptic('selection')
+                onSelect(item.text)
+              }}
+              onMouseDown={event => event.preventDefault()}
+              role="option"
+              type="button"
+            >
+              <span className="shrink-0 font-mono font-medium leading-5 text-foreground">{item.display}</span>
+              <span className="min-w-0 truncate leading-5 text-muted-foreground/80">{item.meta}</span>
+            </button>
+          ))
+        )}
+      </div>
+    </div>
+  )
+}

apps/desktop/src/app/chat/composer/text-utils.test.ts

@@ -22,33 +22,6 @@ describe('detectTrigger', () => {
   it('returns null for plain text', () => {
     expect(detectTrigger('hello there')).toBeNull()
   })
-
-  it('keeps the slash trigger live while typing args', () => {
-    expect(detectTrigger('/personality ')).toEqual({
-      kind: '/',
-      query: 'personality ',
-      tokenLength: 13
-    })
-    expect(detectTrigger('/personality alic')).toEqual({
-      kind: '/',
-      query: 'personality alic',
-      tokenLength: 17
-    })
-    expect(detectTrigger('/tools enable foo')).toEqual({
-      kind: '/',
-      query: 'tools enable foo',
-      tokenLength: 17
-    })
-  })
-
-  it('does not treat file-style paths as slash triggers', () => {
-    expect(detectTrigger('src/foo/bar')).toBeNull()
-    expect(detectTrigger('/path/to/file')).toBeNull()
-  })
-
-  it('still anchors at-mention triggers strictly at the token edge', () => {
-    expect(detectTrigger('@file:path with space')).toBeNull()
-  })
 })
 
 describe('extractClipboardImageBlobs', () => {

apps/desktop/src/app/chat/composer/text-utils.ts

@@ -6,13 +6,7 @@ export interface TriggerState {
   tokenLength: number
 }
 
-// `@` triggers stop at the first whitespace — `@file:path` and `@diff` are
-// single tokens. `/` triggers keep going so the popover stays live while the
-// user types args (`/personality alic` → arg completer suggests `alice`).
-// Restricting the slash command name to `[a-zA-Z][\w-]*` avoids matching file
-// paths like `src/foo/bar`.
-const AT_TRIGGER_RE = /(?:^|[\s])(@)([^\s@/]*)$/
-const SLASH_TRIGGER_RE = /(?:^|[\s])(\/)((?:[a-zA-Z][\w-]*(?:\s+\S*)*)?)$/
+const TRIGGER_RE = /(?:^|[\s])([@/])([^\s@/]*)$/
 
 /** Stable key for paste dedupe — `items` and `files` often mirror the same image as different objects. */
 export function blobDedupeKey(blob: Blob): string {
@@ -103,17 +97,11 @@ export function textBeforeCaret(editor: HTMLDivElement): string | null {
 }
 
 export function detectTrigger(textBefore: string): TriggerState | null {
-  const slash = SLASH_TRIGGER_RE.exec(textBefore)
+  const match = TRIGGER_RE.exec(textBefore)
 
-  if (slash) {
-    return { kind: '/', query: slash[2], tokenLength: 1 + slash[2].length }
+  if (!match) {
+    return null
   }
 
-  const at = AT_TRIGGER_RE.exec(textBefore)
-
-  if (at) {
-    return { kind: '@', query: at[2], tokenLength: 1 + at[2].length }
-  }
-
-  return null
+  return { kind: match[1] as '@' | '/', query: match[2], tokenLength: 1 + match[2].length }
 }

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

@@ -34,17 +34,9 @@ describe('ComposerTriggerPopover i18n', () => {
   })
 
   it('renders localized loading copy for slash commands', () => {
-    renderPopover('/', true)
+    const { container } = renderPopover('/', true)
 
-    // While loading the popover shows only the spinner + loading copy — the
-    // `/help` empty-state hint is reserved for the resolved (not-loading) state.
     expect(screen.getByText('查找中…')).toBeTruthy()
-  })
-
-  it('renders the slash empty-state hint when not loading', () => {
-    const { container } = renderPopover('/')
-
-    expect(screen.getByText('没有匹配项。')).toBeTruthy()
     expect(container.textContent).toContain('/help')
   })
 })

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

@@ -1,7 +1,5 @@
 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 { useI18n } from '@/i18n'
 import { cn } from '@/lib/utils'
@@ -9,6 +7,7 @@ import { cn } from '@/lib/utils'
 import {
   COMPLETION_DRAWER_BELOW_CLASS,
   COMPLETION_DRAWER_CLASS,
+  COMPLETION_DRAWER_ROW_CLASS,
   CompletionDrawerEmpty
 } from './completion-drawer'
 
@@ -24,7 +23,11 @@ const AT_ICON_BY_TYPE: Record<string, string> = {
   url: 'globe'
 }
 
-function atIcon(item: Unstable_TriggerItem) {
+function completionIcon(kind: '@' | '/', item: Unstable_TriggerItem) {
+  if (kind === '/') {
+    return 'terminal'
+  }
+
   const meta = item.metadata as { rawText?: string } | undefined
   const raw = meta?.rawText || item.label
 
@@ -39,18 +42,6 @@ function atIcon(item: Unstable_TriggerItem) {
   return AT_ICON_BY_TYPE[item.type] || AT_ICON_BY_TYPE.simple
 }
 
-interface RowMeta {
-  display?: string
-  group?: string
-  meta?: string
-}
-
-const ROW_BASE_CLASS = [
-  'relative flex w-full cursor-default select-none rounded-md px-2 py-1 text-left',
-  'outline-hidden transition-colors hover:bg-(--ui-bg-tertiary)',
-  'data-[highlighted]:bg-(--ui-bg-tertiary) data-[highlighted]:text-foreground'
-].join(' ')
-
 interface ComposerTriggerPopoverProps {
   activeIndex: number
   items: readonly Unstable_TriggerItem[]
@@ -72,9 +63,6 @@ export function ComposerTriggerPopover({
 }: ComposerTriggerPopoverProps) {
   const { t } = useI18n()
   const copy = t.composer
-  const isSlash = kind === '/'
-
-  let lastGroup: string | undefined
 
   return (
     <div
@@ -85,94 +73,41 @@ export function ComposerTriggerPopover({
       role="listbox"
     >
       {items.length === 0 ? (
-        loading ? (
-          <div className="flex items-center gap-2 px-2 py-1.5 text-(--ui-text-tertiary)">
-            <BrailleSpinner ariaLabel={copy.lookupLoading} className="text-foreground/70" spinner="braille" />
-            <span>{copy.lookupLoading}</span>
-          </div>
-        ) : (
-          <CompletionDrawerEmpty title={copy.lookupNoMatches}>
-            {kind === '@' ? (
-              <>
-                {copy.lookupTry} <span className="font-mono text-foreground/80">@file:</span> {copy.lookupOr}{' '}
-                <span className="font-mono text-foreground/80">@folder:</span>.
-              </>
-            ) : (
-              <>
-                {copy.lookupTry} <span className="font-mono text-foreground/80">/help</span>.
-              </>
-            )}
-          </CompletionDrawerEmpty>
-        )
+        <CompletionDrawerEmpty title={loading ? copy.lookupLoading : copy.lookupNoMatches}>
+          {kind === '@' ? (
+            <>
+              {copy.lookupTry} <span className="font-mono text-foreground/80">@file:</span> {copy.lookupOr}{' '}
+              <span className="font-mono text-foreground/80">@folder:</span>.
+            </>
+          ) : (
+            <>
+              {copy.lookupTry} <span className="font-mono text-foreground/80">/help</span>.
+            </>
+          )}
+        </CompletionDrawerEmpty>
       ) : (
         items.map((item, index) => {
-          const meta = item.metadata as RowMeta | undefined
-          const display = meta?.display ?? (isSlash ? `/${item.label}` : item.label)
+          const meta = item.metadata as { display?: string; meta?: string } | undefined
+          const display = meta?.display ?? (kind === '/' ? `/${item.label}` : item.label)
           const description = meta?.meta || item.description
-          const group = meta?.group?.trim()
-          const showHeader = isSlash && Boolean(group) && group !== lastGroup
-          const isFirstHeader = lastGroup === undefined
-          lastGroup = group || lastGroup
-          const active = index === activeIndex
 
           return (
-            <Fragment key={item.id}>
-              {showHeader && (
-                <div
-                  className={cn(
-                    'select-none px-2 pb-0.5 text-[0.625rem] font-semibold uppercase tracking-wider text-(--ui-text-tertiary)',
-                    isFirstHeader ? 'pt-0.5' : 'pt-2'
-                  )}
-                >
-                  {group}
-                </div>
+            <button
+              className={cn(COMPLETION_DRAWER_ROW_CLASS, index === activeIndex && 'bg-(--ui-bg-tertiary)')}
+              data-highlighted={index === activeIndex ? '' : undefined}
+              key={item.id}
+              onClick={() => onPick(item)}
+              onMouseEnter={() => onHover(index)}
+              type="button"
+            >
+              <span className="grid size-3.5 shrink-0 place-items-center text-(--ui-text-tertiary)">
+                <Codicon name={completionIcon(kind, item)} size="0.875rem" />
+              </span>
+              <span className="min-w-0 shrink truncate font-mono font-medium leading-5 text-foreground">{display}</span>
+              {description && (
+                <span className="min-w-0 flex-1 truncate leading-5 text-(--ui-text-tertiary)">{description}</span>
               )}
-              <button
-                className={cn(ROW_BASE_CLASS, isSlash ? 'flex-col gap-0' : 'items-center gap-2')}
-                data-highlighted={active ? '' : undefined}
-                onClick={() => onPick(item)}
-                onMouseEnter={() => onHover(index)}
-                type="button"
-              >
-                {isSlash ? (
-                  <>
-                    {/* Active row (keyboard nav or hover) un-truncates inline so
-                        long command names / descriptions stay readable without a
-                        floating tooltip. */}
-                    <span
-                      className={cn(
-                        'text-[0.8125rem] font-medium leading-snug text-foreground',
-                        active ? 'whitespace-normal break-words' : 'truncate'
-                      )}
-                    >
-                      {display}
-                    </span>
-                    {description && (
-                      <span
-                        className={cn(
-                          'text-[0.6875rem] leading-snug text-(--ui-text-tertiary)',
-                          active ? 'whitespace-normal break-words' : 'truncate'
-                        )}
-                      >
-                        {description}
-                      </span>
-                    )}
-                  </>
-                ) : (
-                  <>
-                    <span className="grid size-4 shrink-0 place-items-center text-(--ui-text-tertiary)">
-                      <Codicon name={atIcon(item)} size="0.875rem" />
-                    </span>
-                    <span className="min-w-0 shrink truncate font-mono font-medium leading-5 text-foreground">
-                      {display}
-                    </span>
-                    {description && (
-                      <span className="min-w-0 flex-1 truncate leading-5 text-(--ui-text-tertiary)">{description}</span>
-                    )}
-                  </>
-                )}
-              </button>
-            </Fragment>
+            </button>
           )
         })
       )}

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

@@ -1,57 +0,0 @@
-import { describe, expect, it } from 'vitest'
-
-import { type DroppedFile, partitionDroppedFiles } from './use-composer-actions'
-
-// A Finder/Explorer drop carries a native File handle; an in-app drag (project
-// tree, gutter line ref) is path-only. The split decides whether a drop becomes
-// an inline @file: ref (in-app, workspace-relative, gateway-resolvable) or goes
-// through the upload pipeline (OS drop — absolute local path a remote gateway
-// can't read, plus image bytes for vision).
-const osDrop = (path: string): DroppedFile => ({ file: new File(['x'], path.split('/').pop() || 'f'), path })
-const inAppRef = (path: string, extra: Partial<DroppedFile> = {}): DroppedFile => ({ path, ...extra })
-
-describe('partitionDroppedFiles', () => {
-  it('routes File-bearing OS drops to osDrops and path-only in-app drags to inAppRefs', () => {
-    const finderPdf = osDrop('/Users/mahmoud/Downloads/DEVIS_signed.pdf')
-    const projectFile = inAppRef('src/index.ts')
-
-    const { inAppRefs, osDrops } = partitionDroppedFiles([finderPdf, projectFile])
-
-    expect(osDrops).toEqual([finderPdf])
-    expect(inAppRefs).toEqual([projectFile])
-  })
-
-  it('treats an OS screenshot drop as an upload target (so it gets byte upload + vision)', () => {
-    const screenshot = osDrop('/var/folders/tmp/Screenshot 2026-06-09.png')
-
-    const { inAppRefs, osDrops } = partitionDroppedFiles([screenshot])
-
-    expect(osDrops).toEqual([screenshot])
-    expect(inAppRefs).toEqual([])
-  })
-
-  it('keeps gutter line-range drags inline (no File handle)', () => {
-    const lineRef = inAppRef('src/app.ts', { line: 10, lineEnd: 20 })
-
-    const { inAppRefs, osDrops } = partitionDroppedFiles([lineRef])
-
-    expect(osDrops).toEqual([])
-    expect(inAppRefs).toEqual([lineRef])
-  })
-
-  it('splits a mixed drop and preserves order within each group', () => {
-    const a = inAppRef('a.ts')
-    const b = osDrop('/abs/b.pdf')
-    const c = inAppRef('c.ts')
-    const d = osDrop('/abs/d.png')
-
-    const { inAppRefs, osDrops } = partitionDroppedFiles([a, b, c, d])
-
-    expect(inAppRefs).toEqual([a, c])
-    expect(osDrops).toEqual([b, d])
-  })
-
-  it('returns empty groups for an empty drop', () => {
-    expect(partitionDroppedFiles([])).toEqual({ inAppRefs: [], osDrops: [] })
-  })
-})

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

@@ -33,7 +33,7 @@ function blobExtension(blob: Blob): string {
   return (mime && BLOB_MIME_EXTENSION[mime]) || '.png'
 }
 
-export function isImagePath(filePath: string): boolean {
+function isImagePath(filePath: string): boolean {
   return IMAGE_EXTENSION_PATTERN.test(filePath)
 }
 
@@ -181,35 +181,6 @@ export function extractDroppedFiles(transfer: DataTransfer): DroppedFile[] {
   return result
 }
 
-/**
- * Split dropped entries by origin. OS/Finder drops carry a native `File`
- * handle; in-app drags (project tree, gutter line refs) are path-only.
- *
- * The distinction is load-bearing: an in-app path is workspace-relative and
- * resolves on the gateway as-is, so it stays an inline `@file:`/`@line:` ref.
- * An OS drop is an absolute path on *this* machine — the gateway can't read it
- * in remote mode, and an image needs its bytes uploaded to get vision either
- * way. So OS drops must go through the attachment/upload pipeline rather than
- * leaking a local path into the prompt text.
- */
-export function partitionDroppedFiles(candidates: DroppedFile[]): {
-  osDrops: DroppedFile[]
-  inAppRefs: DroppedFile[]
-} {
-  const osDrops: DroppedFile[] = []
-  const inAppRefs: DroppedFile[] = []
-
-  for (const candidate of candidates) {
-    if (candidate.file) {
-      osDrops.push(candidate)
-    } else {
-      inAppRefs.push(candidate)
-    }
-  }
-
-  return { osDrops, inAppRefs }
-}
-
 interface ComposerActionsOptions {
   activeSessionId: string | null
   currentCwd: string

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

@@ -49,9 +49,9 @@ import { ChatDropOverlay } from './chat-drop-overlay'
 import { ChatSwapOverlay } from './chat-swap-overlay'
 import { ChatBar, ChatBarFallback } from './composer'
 import { requestComposerInsert, requestComposerInsertRefs } from './composer/focus'
-import { droppedFileInlineRefs, type SessionDragPayload, sessionInlineRef } from './composer/inline-refs'
+import { droppedFileInlineRef, type SessionDragPayload, sessionInlineRef } from './composer/inline-refs'
 import type { ChatBarState } from './composer/types'
-import { type DroppedFile, partitionDroppedFiles } from './hooks/use-composer-actions'
+import type { DroppedFile } from './hooks/use-composer-actions'
 import { useFileDropZone } from './hooks/use-file-drop-zone'
 import { SessionActionsMenu } from './sidebar/session-actions-menu'
 import { lastVisibleMessageIsUser, threadLoadingState } from './thread-loading'
@@ -124,13 +124,7 @@ function ChatHeader({
 
   return (
     <header className={cn(titlebarHeaderBaseClass, isRoutedSessionView && titlebarHeaderShadowClass)}>
-      <div
-        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)'
-        }}
-      >
+      <div className="min-w-0 flex-1">
         <SessionActionsMenu
           align="start"
           onDelete={selectedSessionId ? onDeleteSelectedSession : undefined}
@@ -141,11 +135,11 @@ function ChatHeader({
           title={title}
         >
           <Button
-            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]"
+            className="pointer-events-auto h-6 min-w-0 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"
           >
-            <h2 className="min-w-0 flex-1 truncate text-[0.75rem] font-medium leading-none">{title}</h2>
+            <h2 className="max-w-[52vw] truncate text-[0.75rem] font-medium leading-none">{title}</h2>
             <Codicon className="shrink-0 text-(--ui-text-tertiary)" name="chevron-down" size="0.8125rem" />
           </Button>
         </SessionActionsMenu>
@@ -302,25 +296,19 @@ export function ChatView({
   })
 
   // 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
-  // drops carry absolute local paths that don't exist on a remote gateway (and
-  // images need byte upload for vision), so route them through the attachment
-  // pipeline — otherwise the local path leaks into the prompt verbatim.
+  // input — appending the same inline `@file:` ref chips the composer drop
+  // produces (vs. attachment cards) so both surfaces behave identically.
   const onDropFiles = useCallback(
     (candidates: DroppedFile[]) => {
-      const { inAppRefs, osDrops } = partitionDroppedFiles(candidates)
-      const refs = droppedFileInlineRefs(inAppRefs, currentCwd)
+      const refs = candidates
+        .map(candidate => droppedFileInlineRef(candidate, currentCwd))
+        .filter((ref): ref is string => Boolean(ref))
 
       if (refs.length) {
         requestComposerInsert(refs.join(' '), { mode: 'inline', target: 'main' })
       }
-
-      if (osDrops.length) {
-        void onAttachDroppedItems(osDrops)
-      }
     },
-    [currentCwd, onAttachDroppedItems]
+    [currentCwd]
   )
 
   // Dropping a sidebar session inserts an @session link the agent can resolve

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

@@ -446,9 +446,7 @@ export function LocalFilePreview({ reloadKey, target }: { reloadKey: number; tar
 
       try {
         if (isImage) {
-          // Prefer bytes the caller already handed us (a pasted/dropped
-          // screenshot) over re-reading a path that may be transient/unreadable.
-          const dataUrl = target.dataUrl || (await window.hermesDesktop.readFileDataUrl(filePath))
+          const dataUrl = await window.hermesDesktop.readFileDataUrl(filePath)
 
           if (active) {
             setState({ dataUrl, loading: false })
@@ -486,7 +484,7 @@ export function LocalFilePreview({ reloadKey, target }: { reloadKey: number; tar
     return () => {
       active = false
     }
-  }, [blockedByTarget, filePath, forcePreview, isImage, isText, reloadKey, target.dataUrl, target.language])
+  }, [blockedByTarget, filePath, forcePreview, isImage, isText, reloadKey, target.language])
 
   if (state.loading) {
     return <PageLoader label={t.preview.loading} />

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

@@ -14,8 +14,6 @@ import type { CronJob } from '@/types/hermes'
 import { jobState, jobTitle, STATE_DOT } from '../../cron/job-state'
 import { SidebarPanelLabel } from '../../shell/sidebar-label'
 
-import { SidebarLoadMoreRow } from './load-more-row'
-
 const INACTIVE_STATES = new Set(['completed', 'disabled', 'error', 'paused'])
 
 // Recent runs shown in the inline quick-peek — enough to glance at history
@@ -26,11 +24,6 @@ const PEEK_RUN_LIMIT = 5
 // open peek so a freshly-fired run shows up within a few seconds.
 const PEEK_POLL_INTERVAL_MS = 8000
 
-// Keep the section compact: show a few jobs up front, reveal more in larger
-// steps on demand (mirrors the messaging sections in the sidebar).
-const INITIAL_VISIBLE_JOBS = 3
-const LOAD_MORE_STEP = 10
-
 const relativeFmt = new Intl.RelativeTimeFormat(undefined, { numeric: 'auto', style: 'short' })
 
 // Localized "in 5 min" / "2 hr ago" without hand-rolled strings — picks the
@@ -40,25 +33,17 @@ function relativeTime(targetMs: number, nowMs: number): string {
   const abs = Math.abs(diff)
   const sign = diff < 0 ? -1 : 1
 
-  if (abs < 60_000) {
-    return relativeFmt.format(sign * Math.round(abs / 1000), 'second')
-  }
+  if (abs < 60_000) {return relativeFmt.format(sign * Math.round(abs / 1000), 'second')}
 
-  if (abs < 3_600_000) {
-    return relativeFmt.format(sign * Math.round(abs / 60_000), 'minute')
-  }
+  if (abs < 3_600_000) {return relativeFmt.format(sign * Math.round(abs / 60_000), 'minute')}
 
-  if (abs < 86_400_000) {
-    return relativeFmt.format(sign * Math.round(abs / 3_600_000), 'hour')
-  }
+  if (abs < 86_400_000) {return relativeFmt.format(sign * Math.round(abs / 3_600_000), 'hour')}
 
   return relativeFmt.format(sign * Math.round(abs / 86_400_000), 'day')
 }
 
 function nextRunMs(job: CronJob): null | number {
-  if (!job.next_run_at) {
-    return null
-  }
+  if (!job.next_run_at) {return null}
 
   const ms = Date.parse(job.next_run_at)
 
@@ -69,9 +54,7 @@ function nextRunMs(job: CronJob): null | number {
 // the timestamp is what tells them apart. Compact (no year, no seconds) for the
 // narrow sidebar.
 function formatRunTime(seconds?: null | number): string {
-  if (!seconds) {
-    return '—'
-  }
+  if (!seconds) {return '—'}
 
   const date = new Date(seconds * 1000)
 
@@ -107,15 +90,11 @@ export function SidebarCronJobsSection({
   const [nowMs, setNowMs] = useState(() => Date.now())
   // Single-open inline peek so the section stays scannable.
   const [peekJobId, setPeekJobId] = useState<null | string>(null)
-  // Rows revealed so far; starts compact, grows in steps via "load more".
-  const [visibleCount, setVisibleCount] = useState(INITIAL_VISIBLE_JOBS)
 
   // One clock for the whole section (rows are pure) so the countdowns tick
   // without re-rendering the rest of the sidebar. Only runs while expanded.
   useEffect(() => {
-    if (!open) {
-      return
-    }
+    if (!open) {return}
 
     const id = window.setInterval(() => setNowMs(Date.now()), 1000)
 
@@ -129,25 +108,17 @@ export function SidebarCronJobsSection({
       const an = nextRunMs(a)
       const bn = nextRunMs(b)
 
-      if (an !== null && bn !== null && an !== bn) {
-        return an - bn
-      }
+      if (an !== null && bn !== null && an !== bn) {return an - bn}
 
-      if (an === null && bn !== null) {
-        return 1
-      }
+      if (an === null && bn !== null) {return 1}
 
-      if (an !== null && bn === null) {
-        return -1
-      }
+      if (an !== null && bn === null) {return -1}
 
       return jobTitle(a).localeCompare(jobTitle(b))
     })
   }, [jobs])
 
-  const cap = Math.min(visibleCount, max)
-  const shown = sorted.slice(0, cap)
-  const hiddenCount = Math.min(sorted.length, max) - shown.length
+  const shown = sorted.slice(0, max)
   // When capped, signal "50+" rather than implying the list is complete.
   const countLabel = jobs.length > max ? `${max}+` : String(jobs.length)
 
@@ -168,7 +139,7 @@ export function SidebarCronJobsSection({
         </button>
       </div>
       {open && (
-        <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">
+        <SidebarGroupContent className="flex max-h-72 shrink-0 flex-col gap-px overflow-y-auto overscroll-contain pb-1.75">
           {shown.map(job => (
             <CronJobSidebarRow
               expanded={peekJobId === job.id}
@@ -181,12 +152,6 @@ export function SidebarCronJobsSection({
               onTrigger={() => onTriggerJob(job.id)}
             />
           ))}
-          {hiddenCount > 0 && (
-            <SidebarLoadMoreRow
-              onClick={() => setVisibleCount(count => count + LOAD_MORE_STEP)}
-              step={Math.min(LOAD_MORE_STEP, hiddenCount)}
-            />
-          )}
         </SidebarGroupContent>
       )}
     </SidebarGroup>
@@ -216,7 +181,11 @@ function CronJobSidebarRow({
   const next = nextRunMs(job)
   const label = jobTitle(job)
 
-  const meta = INACTIVE_STATES.has(state) ? (c.states[state] ?? state) : next !== null ? relativeTime(next, nowMs) : '—'
+  const meta = INACTIVE_STATES.has(state)
+    ? (c.states[state] ?? state)
+    : next !== null
+      ? relativeTime(next, nowMs)
+      : '—'
 
   return (
     <div>
@@ -288,7 +257,13 @@ function CronJobSidebarRow({
   )
 }
 
-function CronJobSidebarRuns({ jobId, onOpenRun }: { jobId: string; onOpenRun: (sessionId: string) => void }) {
+function CronJobSidebarRuns({
+  jobId,
+  onOpenRun
+}: {
+  jobId: string
+  onOpenRun: (sessionId: string) => void
+}) {
   const { t } = useI18n()
   const c = t.cron
   const selectedSessionId = useStore($selectedStoredSessionId)
@@ -300,22 +275,16 @@ function CronJobSidebarRuns({ jobId, onOpenRun }: { jobId: string; onOpenRun: (s
     const load = () =>
       getCronJobRuns(jobId, PEEK_RUN_LIMIT)
         .then(result => {
-          if (!cancelled) {
-            setRuns(result)
-          }
+          if (!cancelled) {setRuns(result)}
         })
         .catch(() => {
-          if (!cancelled) {
-            setRuns(prev => prev ?? [])
-          }
+          if (!cancelled) {setRuns(prev => prev ?? [])}
         })
 
     void load()
 
     const intervalId = window.setInterval(() => {
-      if (document.visibilityState === 'visible') {
-        void load()
-      }
+      if (document.visibilityState === 'visible') {void load()}
     }, PEEK_POLL_INTERVAL_MS)
 
     return () => {

apps/desktop/src/app/chat/sidebar/load-more-row.tsx

@@ -1,30 +0,0 @@
-import { Codicon } from '@/components/ui/codicon'
-import { useI18n } from '@/i18n'
-
-interface SidebarLoadMoreRowProps {
-  step: number
-  onClick: () => void
-  loading?: boolean
-}
-
-// "Load N more" affordance shared by the recents, messaging, and cron sections.
-// The chevron sits in the same w-3.5 column the rows use for their dot, so it
-// lines up with the list above.
-export function SidebarLoadMoreRow({ step, onClick, loading = false }: SidebarLoadMoreRowProps) {
-  const { t } = useI18n()
-  const label = loading ? t.sidebar.loading : step > 0 ? t.sidebar.loadCount(step) : t.sidebar.loadMore
-
-  return (
-    <button
-      className="flex min-h-5 items-center gap-1.5 self-start bg-transparent pl-2 text-left text-[0.6875rem] text-(--ui-text-tertiary) transition-colors duration-100 ease-out hover:text-foreground hover:transition-none disabled:cursor-default disabled:opacity-60 disabled:hover:text-(--ui-text-tertiary)"
-      disabled={loading}
-      onClick={onClick}
-      type="button"
-    >
-      <span className="grid w-3.5 shrink-0 place-items-center">
-        <Codicon className="opacity-70" name={loading ? 'loading' : 'chevron-down'} size="0.75rem" spinning={loading} />
-      </span>
-      <span>{label}</span>
-    </button>
-  )
-}

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

@@ -83,9 +83,8 @@ const stepThroughCells: Modifier = ({ containerNodeRect, draggingNodeRect, trans
 // Arc-Spaces-style profile rail at the sidebar foot: a default↔all toggle pinned
 // left, the colored named profiles scrolling between, and Manage pinned right.
 // The active profile pops in its own color — the "where am I" cue. Single-
-// profile users see the "+" (create their first profile) and the Manage
-// overflow (edit the default profile's SOUL.md); the colored named squares
-// and the default↔all toggle only appear once a second profile exists.
+// profile users see only the "+" (create their first profile); everything else
+// appears once a second profile exists.
 export function ProfileRail() {
   const { t } = useI18n()
   const p = t.profiles
@@ -269,11 +268,9 @@ export function ProfileRail() {
         </Tip>
       </div>
 
-      {/* Always reachable, even with only the default profile: the manage
-          overlay is the only place to edit a profile's SOUL.md, and a
-          single-profile user must be able to edit the default's persona
-          without first creating a throwaway second profile. */}
-      <ProfilePill active={false} glyph="ellipsis" label={p.manageProfiles} onSelect={() => navigate(PROFILES_ROUTE)} />
+      {multiProfile && (
+        <ProfilePill active={false} glyph="ellipsis" label={p.manageProfiles} onSelect={() => navigate(PROFILES_ROUTE)} />
+      )}
 
       {/* Land in the new profile on a fresh chat (selectProfile triggers the
           new-session reset), not stuck on the session you were just in. */}

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

@@ -21,7 +21,6 @@ import { triggerHaptic } from '@/lib/haptics'
 import { exportSession } from '@/lib/session-export'
 import { notify, notifyError } from '@/store/notifications'
 import { setSessions } from '@/store/session'
-import { canOpenSessionWindow, openSessionInNewWindow } from '@/store/windows'
 
 interface SessionActions {
   sessionId: string
@@ -69,19 +68,6 @@ function useSessionActions({ sessionId, title, pinned = false, profile, onPin, o
         void writeClipboardText(sessionId).catch(err => notifyError(err, r.copyIdFailed))
       }
     },
-    ...(canOpenSessionWindow()
-      ? [
-          {
-            disabled: !sessionId,
-            icon: 'link-external',
-            label: r.newWindow,
-            onSelect: () => {
-              triggerHaptic('selection')
-              void openSessionInNewWindow(sessionId)
-            }
-          }
-        ]
-      : []),
     {
       disabled: !sessionId,
       icon: 'cloud-download',

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

@@ -2,18 +2,14 @@ import { useStore } from '@nanostores/react'
 import type * as React from 'react'
 
 import { writeSessionDrag } from '@/app/chat/composer/inline-refs'
-import { PlatformAvatar } from '@/app/messaging/platform-icon'
 import { Button } from '@/components/ui/button'
 import { Codicon } from '@/components/ui/codicon'
-import { Tip } from '@/components/ui/tooltip'
 import type { SessionInfo } from '@/hermes'
 import { type Translations, useI18n } from '@/i18n'
 import { sessionTitle } from '@/lib/chat-runtime'
 import { triggerHaptic } from '@/lib/haptics'
-import { handoffOriginSource, sessionSourceLabel } from '@/lib/session-source'
 import { cn } from '@/lib/utils'
 import { $attentionSessionIds } from '@/store/session'
-import { canOpenSessionWindow, openSessionInNewWindow } from '@/store/windows'
 
 import { SessionActionsMenu, SessionContextMenu } from './session-actions-menu'
 
@@ -71,11 +67,6 @@ export function SidebarSessionRow({
   const title = sessionTitle(session)
   const age = formatAge(session.last_active || session.started_at, r)
   const handleLabel = `Reorder ${title}`
-  // A handed-off session's live source is local, but it originated on a
-  // messaging platform — surface that origin as a small badge so e.g. a
-  // Telegram thread continued here still reads as Telegram.
-  const handoffSource = handoffOriginSource(session.handoff_state, session.handoff_platform)
-  const handoffLabel = handoffSource ? sessionSourceLabel(handoffSource) ?? handoffSource : null
   // Subscribe per-row (the leaf) instead of drilling a set through the list —
   // the atom is tiny and rarely non-empty. True when a clarify prompt in this
   // session is waiting on the user.
@@ -133,15 +124,11 @@ export function SidebarSessionRow({
               return
             }
 
-            // ⌘-click (mac) / ⌃-click (win/linux) pops the chat into its own
-            // window — the universal "open in a new window" gesture. Archive
-            // lives in the row's ⋯ and right-click menus. Falls through to a
-            // normal resume when standalone windows aren't available (web embed).
-            if ((event.metaKey || event.ctrlKey) && canOpenSessionWindow()) {
+            if (event.metaKey || event.ctrlKey) {
               event.preventDefault()
               event.stopPropagation()
               triggerHaptic('selection')
-              void openSessionInNewWindow(session.id)
+              onArchive()
 
               return
             }
@@ -189,18 +176,9 @@ export function SidebarSessionRow({
                 needsInput ? 'overflow-visible' : 'overflow-hidden'
               )}
             >
-              <SidebarRowDot isWorking={isWorking} needsInput={needsInput} />
-            </span>
+            <SidebarRowDot isWorking={isWorking} needsInput={needsInput} />
+          </span>
           )}
-          {handoffSource && handoffLabel ? (
-            <Tip label={r.handoffOrigin(handoffLabel)}>
-              <PlatformAvatar
-                className="size-4 rounded-[4px] text-[0.5rem] [&_svg]:size-2.5"
-                platformId={handoffSource}
-                platformName={handoffLabel}
-              />
-            </Tip>
-          ) : null}
           <span className="min-w-0 flex-1 truncate text-[0.8125rem] font-normal text-(--ui-text-secondary) group-hover:text-foreground group-data-[working=true]:text-foreground/90">
             {title}
           </span>

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

@@ -4,10 +4,7 @@ import { Dialog as DialogPrimitive } from 'radix-ui'
 import { useCallback, useEffect, useMemo, useState } from 'react'
 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 { KbdGroup } from '@/components/ui/kbd'
 import { getHermesConfigRecord, listSessions } from '@/hermes'
 import { useI18n } from '@/i18n'
 import { sessionTitle } from '@/lib/chat-runtime'
@@ -15,11 +12,11 @@ import {
   Activity,
   Archive,
   BarChart3,
+  Check,
   ChevronLeft,
   ChevronRight,
   Clock,
   Cpu,
-  Download,
   Globe,
   type IconComponent,
   Info,
@@ -33,18 +30,13 @@ import {
   Settings,
   Settings2,
   Sun,
-  Terminal,
   Users,
   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'
-import { luminance } from '@/themes/color'
 import { type ThemeMode, useTheme } from '@/themes/context'
-import { isUserTheme, resolveTheme } from '@/themes/user-themes'
 
 import {
   AGENTS_ROUTE,
@@ -62,11 +54,8 @@ import { FIELD_LABELS, SECTIONS } from '../settings/constants'
 import { fieldCopyForSchemaKey } from '../settings/field-copy'
 import { prettyName } from '../settings/helpers'
 
-import { MarketplaceThemePage } from './marketplace-theme-page'
-
 interface PaletteItem {
-  /** Keybind action id — its live combo renders as a hotkey hint. */
-  action?: string
+  active?: boolean
   icon: IconComponent
   id: string
   /** Keep the palette open after running (live-preview pickers like theme/mode). */
@@ -80,16 +69,10 @@ interface PaletteItem {
 }
 
 interface PaletteGroup {
-  /** Optional: a headingless group renders as a bare action row (e.g. the
-   *  "Install theme…" entry pinned atop the theme picker). */
-  heading?: string
+  heading: string
   items: PaletteItem[]
 }
 
-// Nested page → its parent, so Back / Esc step up one level instead of closing
-// the palette. Pages absent here go straight back to the root list.
-const PAGE_PARENTS: Record<string, string> = { 'install-theme': 'theme' }
-
 /** A nested page reachable from a root item via `to`. */
 interface PalettePage {
   groups: PaletteGroup[]
@@ -103,22 +86,6 @@ interface SessionEntry {
   title: string
 }
 
-// cmdk defaults to fuzzy subsequence scoring, so "color" matches anything with
-// c…o…l…o…r scattered across it. Use case-insensitive multi-term substring
-// matching instead: every typed word must literally appear in the item's
-// value/keywords, which keeps results tight and predictable.
-const paletteFilter = (value: string, search: string, keywords?: string[]): number => {
-  const needle = search.trim().toLowerCase()
-
-  if (!needle) {
-    return 1
-  }
-
-  const haystack = `${value} ${keywords?.join(' ') ?? ''}`.toLowerCase()
-
-  return needle.split(/\s+/).every(term => haystack.includes(term)) ? 1 : 0
-}
-
 type SessionRow = Awaited<ReturnType<typeof listSessions>>['sessions'][number]
 
 const toSessionEntry = (session: SessionRow): SessionEntry => ({
@@ -179,32 +146,11 @@ const THEME_MODES: ReadonlyArray<{ icon: IconComponent; mode: ThemeMode }> = [
   { icon: Monitor, mode: 'system' }
 ]
 
-// Which Light/Dark groups a theme belongs in. Built-ins render in both modes
-// (the engine synthesises the missing side). Imported VS Code themes only carry
-// the variant(s) the extension shipped — a single dark theme like Dracula lives
-// under Dark only, while a GitHub/Solarized family (light + dark) lives in both.
-function themeSupportsMode(name: string, target: 'light' | 'dark'): boolean {
-  if (!isUserTheme(name)) {
-    return true
-  }
-
-  const resolved = resolveTheme(name)
-
-  if (!resolved) {
-    return true
-  }
-
-  const background = target === 'dark' ? (resolved.darkColors ?? resolved.colors).background : resolved.colors.background
-
-  return target === 'dark' ? luminance(background) <= 0.5 : luminance(background) > 0.5
-}
-
 export function CommandPalette() {
   const { t } = useI18n()
   const open = useStore($commandPaletteOpen)
-  const bindings = useStore($bindings)
   const navigate = useNavigate()
-  const { availableThemes, resolvedMode, setMode, setTheme, themeName } = useTheme()
+  const { availableThemes, mode, resolvedMode, setMode, setTheme, themeName } = useTheme()
   const [search, setSearch] = useState('')
   const [page, setPage] = useState<string | null>(null)
 
@@ -248,19 +194,10 @@ export function CommandPalette() {
   }, [open])
 
   const go = useCallback((path: string) => () => navigate(path), [navigate])
-
-  // Step up one nested page (or back to the root list), clearing the filter so
-  // the parent page doesn't reopen mid-search.
-  const goBack = useCallback(() => {
-    setSearch('')
-    setPage(prev => (prev ? (PAGE_PARENTS[prev] ?? null) : null))
-  }, [])
-
   const settingsSectionLabel = useCallback(
     (section: (typeof SECTIONS)[number]) => t.settings.sections[section.id] ?? section.label,
     [t.settings.sections]
   )
-
   const configFieldLabel = useCallback(
     (key: string) =>
       fieldCopyForSchemaKey(t.settings.fieldLabels, key) ??
@@ -277,61 +214,20 @@ export function CommandPalette() {
       {
         heading: cc.goTo,
         items: [
+          { icon: Plus, id: 'nav-new', keywords: ['chat', 'create'], label: cc.nav.newChat.title, run: go(NEW_CHAT_ROUTE) },
+          { icon: Settings, id: 'nav-settings', label: cc.nav.settings.title, run: go(SETTINGS_ROUTE) },
           {
-            action: 'session.new',
-            icon: Plus,
-            id: 'nav-new',
-            keywords: ['chat', 'create'],
-            label: cc.nav.newChat.title,
-            run: go(NEW_CHAT_ROUTE)
-          },
-          {
-            action: 'view.showTerminal',
-            icon: Terminal,
-            id: 'nav-terminal',
-            keywords: ['terminal', 'shell', 'console'],
-            label: t.keybinds.actions['view.showTerminal'],
-            run: () => setTerminalTakeover(true)
-          },
-          {
-            action: 'nav.settings',
-            icon: Settings,
-            id: 'nav-settings',
-            label: cc.nav.settings.title,
-            run: go(SETTINGS_ROUTE)
-          },
-          {
-            action: 'nav.skills',
             icon: Wrench,
             id: 'nav-skills',
             keywords: ['tools', 'toolsets'],
             label: cc.nav.skills.title,
             run: go(SKILLS_ROUTE)
           },
-          {
-            action: 'nav.messaging',
-            icon: MessageCircle,
-            id: 'nav-messaging',
-            label: cc.nav.messaging.title,
-            run: go(MESSAGING_ROUTE)
-          },
-          {
-            action: 'nav.artifacts',
-            icon: Package,
-            id: 'nav-artifacts',
-            label: cc.nav.artifacts.title,
-            run: go(ARTIFACTS_ROUTE)
-          },
-          {
-            action: 'nav.cron',
-            icon: Clock,
-            id: 'nav-cron',
-            keywords: ['schedule', 'jobs'],
-            label: t.shell.statusbar.cron,
-            run: go(CRON_ROUTE)
-          },
-          { action: 'nav.profiles', icon: Users, id: 'nav-profiles', label: t.profiles.title, run: go(PROFILES_ROUTE) },
-          { action: 'nav.agents', icon: Cpu, id: 'nav-agents', label: t.agents.title, run: go(AGENTS_ROUTE) }
+          { icon: MessageCircle, id: 'nav-messaging', label: cc.nav.messaging.title, run: go(MESSAGING_ROUTE) },
+          { icon: Package, id: 'nav-artifacts', label: cc.nav.artifacts.title, run: go(ARTIFACTS_ROUTE) },
+          { icon: Clock, id: 'nav-cron', keywords: ['schedule', 'jobs'], label: t.shell.statusbar.cron, run: go(CRON_ROUTE) },
+          { icon: Users, id: 'nav-profiles', label: t.profiles.title, run: go(PROFILES_ROUTE) },
+          { icon: Cpu, id: 'nav-agents', label: t.agents.title, run: go(AGENTS_ROUTE) }
         ]
       },
       {
@@ -477,40 +373,24 @@ export function CommandPalette() {
       theme: {
         title: t.settings.appearance.themeTitle,
         placeholder: t.settings.appearance.themeDesc,
-        groups: [
-          // Pinned at the top: drills into the Marketplace browser.
-          {
-            items: [
-              {
-                icon: Download,
-                id: 'theme-install',
-                keywords: ['install', 'marketplace', 'vscode', 'vs code', 'download', 'new', 'color'],
-                label: t.commandCenter.installTheme.title,
-                to: 'install-theme'
-              }
-            ]
-          },
-          // Built-ins and imported families list under the mode(s) they support;
-          // picking sets skin + mode at once. A multi-variant import (GitHub,
-          // Solarized) appears in both groups and switches variants with the mode.
-          ...(['light', 'dark'] as const).map(groupMode => ({
-            heading: groupMode === 'light' ? t.settings.modeOptions.light.label : t.settings.modeOptions.dark.label,
-            items: availableThemes
-              .filter(theme => themeSupportsMode(theme.name, groupMode))
-              .map(theme => ({
-                active: themeName === theme.name && resolvedMode === groupMode,
-                icon: groupMode === 'light' ? Sun : Moon,
-                id: `theme-${theme.name}-${groupMode}`,
-                keepOpen: true,
-                keywords: ['theme', 'appearance', 'palette', groupMode, theme.label, theme.description ?? ''],
-                label: theme.label,
-                run: () => {
-                  setTheme(theme.name)
-                  setMode(groupMode)
-                }
-              }))
+        // Skins aren't inherently light/dark — the same skin renders in either
+        // mode. Group by appearance so picking an entry sets skin + mode at
+        // once, and keep the palette open so each pick previews live.
+        groups: (['light', 'dark'] as const).map(groupMode => ({
+          heading: groupMode === 'light' ? t.settings.modeOptions.light.label : t.settings.modeOptions.dark.label,
+          items: availableThemes.map(theme => ({
+            active: themeName === theme.name && resolvedMode === groupMode,
+            icon: groupMode === 'light' ? Sun : Moon,
+            id: `theme-${theme.name}-${groupMode}`,
+            keepOpen: true,
+            keywords: ['theme', 'appearance', 'palette', groupMode, theme.label, theme.description ?? ''],
+            label: theme.label,
+            run: () => {
+              setTheme(theme.name)
+              setMode(groupMode)
+            }
           }))
-        ]
+        }))
       },
       'color-mode': {
         title: t.settings.appearance.colorMode,
@@ -519,6 +399,7 @@ export function CommandPalette() {
           {
             heading: t.settings.appearance.colorMode,
             items: THEME_MODES.map(entry => ({
+              active: mode === entry.mode,
               icon: entry.icon,
               id: `mode-${entry.mode}`,
               keepOpen: true,
@@ -528,16 +409,9 @@ export function CommandPalette() {
             }))
           }
         ]
-      },
-      // Server-driven page: items come from the Marketplace, rendered by
-      // <MarketplaceThemePage> (loader + live search + per-row install).
-      'install-theme': {
-        title: t.commandCenter.installTheme.title,
-        placeholder: t.commandCenter.installTheme.placeholder,
-        groups: []
       }
     }),
-    [availableThemes, resolvedMode, setMode, setTheme, t, themeName]
+    [availableThemes, mode, resolvedMode, setMode, setTheme, t, themeName]
   )
 
   const activePage = page ? subPages[page] : null
@@ -562,22 +436,17 @@ export function CommandPalette() {
   return (
     <DialogPrimitive.Root onOpenChange={setCommandPaletteOpen} open={open}>
       <DialogPrimitive.Portal>
-        {/* Transparent overlay: keeps click-away + focus trap, but no dim/blur. */}
-        <DialogPrimitive.Overlay className="fixed inset-0 z-[200]" />
+        <DialogPrimitive.Overlay className="fixed inset-0 z-[200] bg-black/15 backdrop-blur-[1px] data-[state=closed]:animate-out data-[state=closed]:fade-out-0 data-[state=open]:animate-in data-[state=open]:fade-in-0" />
         <DialogPrimitive.Content
           aria-describedby={undefined}
-          className={cn(
-            HUD_POSITION,
-            HUD_SURFACE,
-            'z-[210] w-[min(34rem,calc(100vw-2rem))] overflow-hidden duration-150 data-[state=closed]:animate-out data-[state=closed]:fade-out-0 data-[state=closed]:zoom-out-95 data-[state=open]:animate-in data-[state=open]:fade-in-0 data-[state=open]:slide-in-from-top-2 data-[state=open]:zoom-in-95'
-          )}
+          className="fixed left-1/2 top-[14vh] z-[210] w-[min(40rem,calc(100vw-2rem))] -translate-x-1/2 overflow-hidden rounded-xl border border-(--ui-stroke-secondary) bg-(--ui-chat-bubble-background) shadow-lg duration-150 data-[state=closed]:animate-out data-[state=closed]:fade-out-0 data-[state=closed]:zoom-out-95 data-[state=open]:animate-in data-[state=open]:fade-in-0 data-[state=open]:slide-in-from-top-2 data-[state=open]:zoom-in-95"
         >
           <DialogPrimitive.Title className="sr-only">{t.commandCenter.paletteTitle}</DialogPrimitive.Title>
-          <Command className="bg-transparent" filter={paletteFilter} loop>
+          <Command className="bg-transparent" loop>
             {activePage && (
               <button
                 className="flex w-full items-center gap-1.5 border-b border-border px-3 py-1.5 text-left text-xs text-muted-foreground transition-colors hover:text-foreground"
-                onClick={goBack}
+                onClick={() => setPage(null)}
                 type="button"
               >
                 <ChevronLeft className="size-3.5" />
@@ -587,7 +456,6 @@ export function CommandPalette() {
               </button>
             )}
             <CommandInput
-              className={HUD_TEXT}
               onKeyDown={event => {
                 if (!activePage) {
                   return
@@ -598,45 +466,38 @@ export function CommandPalette() {
                 if (event.key === 'Escape' || (event.key === 'Backspace' && search === '')) {
                   event.preventDefault()
                   event.stopPropagation()
-                  goBack()
+                  setPage(null)
                 }
               }}
               onValueChange={setSearch}
               placeholder={placeholder}
               value={search}
             />
-            <CommandList className="dt-portal-scrollbar max-h-[min(20rem,56vh)]">
-              {page === 'install-theme' ? (
-                <MarketplaceThemePage onPickTheme={setTheme} search={search} />
-              ) : (
-                <CommandEmpty>{t.commandCenter.noResults}</CommandEmpty>
-              )}
-              {visibleGroups.map((group, index) => (
+            <CommandList className="max-h-[min(24rem,60vh)]">
+              <CommandEmpty>{t.commandCenter.noResults}</CommandEmpty>
+              {visibleGroups.map(group => (
                 <CommandGroup
-                  className={HUD_HEADING}
+                  className="**:[[cmdk-group-heading]]:uppercase **:[[cmdk-group-heading]]:tracking-wider **:[[cmdk-group-heading]]:text-[0.6875rem] **:[[cmdk-group-heading]]:text-muted-foreground/70"
                   heading={group.heading}
-                  key={group.heading ?? `palette-group-${index}`}
+                  key={group.heading}
                 >
                   {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
-                        className={cn(HUD_ITEM, HUD_TEXT)}
+                        className="gap-2.5"
                         key={item.id}
                         keywords={item.keywords}
                         onSelect={() => handleSelect(item)}
                         value={`${item.label} ${item.keywords?.join(' ') ?? ''} ${item.id}`}
                       >
-                        <Icon className="size-3.5 shrink-0 text-muted-foreground" />
+                        <Icon className="size-4 shrink-0 text-muted-foreground" />
                         <span className="truncate">{item.label}</span>
-                        {keys && <KbdGroup className="ml-auto" keys={keys} />}
-                        {item.to && (
-                          <ChevronRight
-                            className={cn('size-3.5 shrink-0 text-muted-foreground/70', !keys && 'ml-auto')}
-                          />
+                        {item.to ? (
+                          <ChevronRight className="ml-auto size-4 shrink-0 text-muted-foreground/70" />
+                        ) : (
+                          <Check className={cn('ml-auto size-4 text-foreground', !item.active && 'invisible')} />
                         )}
                       </CommandItem>
                     )

apps/desktop/src/app/command-palette/marketplace-theme-page.tsx

@@ -1,157 +0,0 @@
-/**
- * Cmd-K "Install theme…" page.
- *
- * Browses the VS Code Marketplace for color themes: an empty query shows the
- * most-installed themes, typing runs a live (debounced) search against the
- * Marketplace. Selecting a row downloads + converts + installs it via the same
- * pipeline as the settings importer, then activates it — and stays open so the
- * user can grab several.
- */
-
-import { useQuery } from '@tanstack/react-query'
-import { useEffect, useState } from 'react'
-
-import { HUD_ITEM, HUD_TEXT } from '@/app/floating-hud'
-import type { DesktopMarketplaceSearchItem } from '@/global'
-import { useI18n } from '@/i18n'
-import { triggerHaptic } from '@/lib/haptics'
-import { Check, Download, Loader2, Palette } from '@/lib/icons'
-import { cn } from '@/lib/utils'
-import { installVscodeThemeFromMarketplace } from '@/themes/install'
-
-const compactNumber = new Intl.NumberFormat(undefined, { notation: 'compact', maximumFractionDigits: 1 })
-
-function useDebounced<T>(value: T, delayMs: number): T {
-  const [debounced, setDebounced] = useState(value)
-
-  useEffect(() => {
-    const handle = setTimeout(() => setDebounced(value), delayMs)
-
-    return () => clearTimeout(handle)
-  }, [value, delayMs])
-
-  return debounced
-}
-
-interface MarketplaceThemePageProps {
-  search: string
-  /** Activate a freshly installed theme by slug. */
-  onPickTheme: (name: string) => void
-}
-
-export function MarketplaceThemePage({ search, onPickTheme }: MarketplaceThemePageProps) {
-  const { t } = useI18n()
-  const copy = t.commandCenter.installTheme
-  const debouncedSearch = useDebounced(search.trim(), 300)
-  const [installingId, setInstallingId] = useState<string | null>(null)
-  const [installed, setInstalled] = useState<Record<string, true>>({})
-  const [installError, setInstallError] = useState<string | null>(null)
-
-  const query = useQuery({
-    queryKey: ['marketplace-themes', debouncedSearch],
-    queryFn: () => window.hermesDesktop?.themes?.searchMarketplace(debouncedSearch) ?? Promise.resolve([]),
-    staleTime: 5 * 60 * 1000
-  })
-
-  const install = async (item: DesktopMarketplaceSearchItem) => {
-    if (installingId) {
-      return
-    }
-
-    setInstallingId(item.extensionId)
-    setInstallError(null)
-
-    try {
-      const theme = await installVscodeThemeFromMarketplace(item.extensionId)
-
-      triggerHaptic('crisp')
-      setInstalled(prev => ({ ...prev, [item.extensionId]: true }))
-      onPickTheme(theme.name)
-    } catch (error) {
-      setInstallError(error instanceof Error ? error.message : copy.error)
-    } finally {
-      setInstallingId(null)
-    }
-  }
-
-  if (query.isLoading) {
-    return <Status icon={<Loader2 className="size-3.5 animate-spin" />} text={copy.loading} />
-  }
-
-  if (query.isError) {
-    return <Status text={copy.error} tone="error" />
-  }
-
-  const results = query.data ?? []
-
-  if (results.length === 0) {
-    return <Status text={copy.empty} />
-  }
-
-  return (
-    <div role="listbox">
-      {installError && <p className="px-2 pb-1 pt-1.5 text-[0.6875rem] text-(--ui-red)">{installError}</p>}
-      {results.map(item => {
-        const busy = installingId === item.extensionId
-        const done = installed[item.extensionId]
-
-        return (
-          <button
-            className={cn(
-              'flex w-full items-start rounded-md text-left transition-colors hover:bg-(--chrome-action-hover) disabled:opacity-60 aria-disabled:opacity-60',
-              HUD_ITEM,
-              HUD_TEXT
-            )}
-            disabled={Boolean(installingId) && !busy}
-            key={item.extensionId}
-            onClick={() => void install(item)}
-            onMouseDown={event => event.preventDefault()}
-            role="option"
-            type="button"
-          >
-            <Palette className="mt-0.5 size-3.5 shrink-0 text-muted-foreground" />
-            <span className="flex min-w-0 flex-col">
-              <span className="truncate font-medium">{item.displayName}</span>
-              <span className="truncate text-[0.6875rem] text-muted-foreground/80">
-                {item.publisher}
-                {item.installs > 0 ? ` · ${copy.installs(compactNumber.format(item.installs))}` : ''}
-              </span>
-            </span>
-            <span className="ml-auto mt-0.5 flex shrink-0 items-center gap-1 text-[0.6875rem] text-muted-foreground">
-              {busy ? (
-                <>
-                  <Loader2 className="size-3 animate-spin" />
-                  {copy.installing}
-                </>
-              ) : done ? (
-                <>
-                  <Check className="size-3 text-(--ui-green)" />
-                  {copy.installed}
-                </>
-              ) : (
-                <>
-                  <Download className="size-3" />
-                  {copy.install}
-                </>
-              )}
-            </span>
-          </button>
-        )
-      })}
-    </div>
-  )
-}
-
-function Status({ icon, text, tone }: { icon?: React.ReactNode; text: string; tone?: 'error' }) {
-  return (
-    <div
-      className={cn(
-        'flex items-center justify-center gap-2 px-2 py-6 text-xs',
-        tone === 'error' ? 'text-(--ui-red)' : 'text-muted-foreground'
-      )}
-    >
-      {icon}
-      {text}
-    </div>
-  )
-}

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

@@ -8,18 +8,11 @@ import { DesktopInstallOverlay } from '@/components/desktop-install-overlay'
 import { DesktopOnboardingOverlay } from '@/components/desktop-onboarding-overlay'
 import { GatewayConnectingOverlay } from '@/components/gateway-connecting-overlay'
 import { Pane, PaneMain } from '@/components/pane-shell'
-import { useMediaQuery } from '@/hooks/use-media-query'
 import { useSkinCommand } from '@/themes/use-skin-command'
 
 import { formatRefValue } from '../components/assistant-ui/directive-text'
 import { getCronJobs, getSessionMessages, listAllProfileSessions, type SessionInfo, triggerCronJob } from '../hermes'
 import { preserveLocalAssistantErrors, toChatMessages } from '../lib/chat-messages'
-import {
-  isMessagingSource,
-  LOCAL_SESSION_SOURCE_IDS,
-  MESSAGING_SESSION_SOURCE_IDS,
-  normalizeSessionSource
-} from '../lib/session-source'
 import { setCronFocusJobId, setCronJobs } from '../store/cron'
 import {
   $panesFlipped,
@@ -30,7 +23,6 @@ import {
   FILE_BROWSER_MAX_WIDTH,
   FILE_BROWSER_MIN_WIDTH,
   pinSession,
-  setSidebarOverlayMounted,
   SIDEBAR_DEFAULT_WIDTH,
   SIDEBAR_MAX_WIDTH,
   SIDEBAR_SESSIONS_PAGE_SIZE,
@@ -50,14 +42,11 @@ import {
   $currentCwd,
   $freshDraftReady,
   $gatewayState,
-  $messagingSessions,
   $selectedStoredSessionId,
   $sessions,
   $workingSessionIds,
   CRON_SECTION_LIMIT,
-  getRecentlySettledSessionIds,
   mergeSessionPage,
-  MESSAGING_SECTION_LIMIT,
   sessionPinId,
   setAwaitingResponse,
   setBusy,
@@ -67,16 +56,12 @@ import {
   setCurrentModel,
   setCurrentProvider,
   setMessages,
-  setMessagingPlatformTotals,
-  setMessagingSessions,
-  setMessagingTruncated,
   setSessionProfileTotals,
   setSessions,
   setSessionsLoading,
   setSessionsTotal
 } from '../store/session'
 import { openUpdatesWindow, startUpdatePoller, stopUpdatePoller } from '../store/updates'
-import { isSecondaryWindow } from '../store/windows'
 
 import { ChatView } from './chat'
 import { useComposerActions } from './chat/hooks/use-composer-actions'
@@ -91,15 +76,12 @@ import { CommandPalette } from './command-palette'
 import { useGatewayBoot } from './gateway/hooks/use-gateway-boot'
 import { useGatewayRequest } from './gateway/hooks/use-gateway-request'
 import { useKeybinds } from './hooks/use-keybinds'
-import { SIDEBAR_COLLAPSE_MEDIA_QUERY } from './layout-constants'
 import { ModelPickerOverlay } from './model-picker-overlay'
 import { ModelVisibilityOverlay } from './model-visibility-overlay'
 import { RightSidebarPane } from './right-sidebar'
 import { $terminalTakeover } from './right-sidebar/store'
 import { PersistentTerminal, TerminalSlot } from './right-sidebar/terminal/persistent'
 import { CRON_ROUTE, NEW_CHAT_ROUTE, routeSessionId, sessionRoute, SETTINGS_ROUTE } from './routes'
-import { SessionPickerOverlay } from './session-picker-overlay'
-import { SessionSwitcher } from './session-switcher'
 import { useContextSuggestions } from './session/hooks/use-context-suggestions'
 import { useCwdActions } from './session/hooks/use-cwd-actions'
 import { useHermesConfig } from './session/hooks/use-hermes-config'
@@ -135,39 +117,22 @@ const SkillsView = lazy(async () => ({ default: (await import('./skills')).Skill
 // this cadence while the app is open + visible so new runs surface promptly
 // instead of waiting for the next user-triggered refreshSessions().
 const CRON_POLL_INTERVAL_MS = 30_000
-// The recents list is local-only: cron rows have their own section, and each
-// messaging platform (telegram, discord, …) is fetched separately into its own
-// 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', ...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]
 
 // Cheap signature compare so the poll only swaps the atom (and re-renders the
 // sidebar) when the visible cron rows actually changed.
 function sameCronSignature(a: SessionInfo[], b: SessionInfo[]): boolean {
-  if (a.length !== b.length) {
-    return false
-  }
+  if (a.length !== b.length) {return false}
 
   return a.every((session, i) => session.id === b[i]?.id && session.title === b[i]?.title)
 }
 
 // Rows a session refresh must preserve even if the aggregator omits them:
-// in-flight first turns (message_count 0), pinned rows aged off the page, the
-// actively-viewed chat (its "working" flag clears a beat before the aggregator
-// sees the persisted row), and sessions whose turn just settled (same race, but
-// for a chat the user has already navigated away from). Pass `scope` to only
-// keep the active row when it belongs to the profile being paged.
+// in-flight first turns (message_count 0), pinned rows aged off the page, and
+// the actively-viewed chat (its "working" flag clears a beat before the
+// aggregator sees the persisted row). Pass `scope` to only keep the active row
+// when it belongs to the profile being paged.
 function sessionsToKeep(scope?: string): Set<string> {
-  const keep = new Set<string>([
-    ...$workingSessionIds.get(),
-    ...$pinnedSessionIds.get(),
-    ...getRecentlySettledSessionIds()
-  ])
-
+  const keep = new Set<string>([...$workingSessionIds.get(), ...$pinnedSessionIds.get()])
   const active = $selectedStoredSessionId.get()
 
   if (active) {
@@ -200,10 +165,6 @@ export function DesktopController() {
   const terminalTakeover = useStore($terminalTakeover)
   const panesFlipped = useStore($panesFlipped)
   const profileScope = useStore($profileScope)
-  // Below SIDEBAR_COLLAPSE_BREAKPOINT_PX there's no room for a docked rail —
-  // collapse both sidebars (without touching their stored open state) so the
-  // hover-reveal overlay becomes the way in. Restores once it's wide again.
-  const narrowViewport = useMediaQuery(SIDEBAR_COLLAPSE_MEDIA_QUERY)
 
   const routedSessionId = routeSessionId(location.pathname)
   const routeToken = `${location.pathname}:${location.search}:${location.hash}`
@@ -226,7 +187,7 @@ export function DesktopController() {
     toggleCommandCenter
   } = useOverlayRouting()
 
-  const terminalSidebarOpen = chatOpen && terminalTakeover
+  const terminalTakeoverActive = chatOpen && terminalTakeover
 
   const titlebarToolGroups = useGroupRegistry<TitlebarTool>()
   const statusbarItemGroups = useGroupRegistry<StatusbarItem>()
@@ -305,51 +266,6 @@ export function DesktopController() {
     }
   }, [])
 
-  // Messaging-platform sessions as their own slice, fetched separately from
-  // local recents so each platform renders a self-managed section and never
-  // competes with local chats for the recents page budget. One combined fetch
-  // seeds every platform; the sidebar splits the rows per source.
-  const refreshMessagingSessions = useCallback(async () => {
-    try {
-      const result = await listAllProfileSessions(MESSAGING_SECTION_LIMIT, 1, 'exclude', 'recent', 'all', {
-        excludeSources: MESSAGING_EXCLUDED_SOURCES
-      })
-
-      // Drop any non-messaging source the broad exclude didn't catch (custom
-      // sources) — those stay in local recents, not a platform section.
-      const rows = result.sessions.filter(s => isMessagingSource(s.source))
-
-      setMessagingSessions(prev => (sameCronSignature(prev, rows) ? prev : rows))
-      // Hit the cap → at least one platform may have more on disk than loaded,
-      // so platform sections offer their own per-platform "load more".
-      setMessagingTruncated(result.sessions.length >= MESSAGING_SECTION_LIMIT)
-    } catch {
-      // Non-fatal: the messaging sections just stay empty/stale.
-    }
-  }, [])
-
-  // Page a single platform's section independently (mirrors the per-profile
-  // pager): fetch that source's next window and merge it back in place, leaving
-  // every other platform's rows untouched. Resolves the platform's exact total.
-  const loadMoreMessagingForPlatform = useCallback(async (platform: string) => {
-    const inPlatform = (s: SessionInfo) => normalizeSessionSource(s.source) === platform
-    const loaded = $messagingSessions.get().filter(inPlatform).length
-
-    const result = await listAllProfileSessions(loaded + SIDEBAR_SESSIONS_PAGE_SIZE, 1, 'exclude', 'recent', 'all', {
-      source: platform
-    })
-
-    const incoming = result.sessions.filter(s => normalizeSessionSource(s.source) === platform)
-
-    setMessagingSessions(prev => [
-      ...prev.filter(s => !inPlatform(s)),
-      ...mergeSessionPage(prev.filter(inPlatform), incoming, sessionsToKeep())
-    ])
-
-    const total = result.total ?? incoming.length
-    setMessagingPlatformTotals(prev => ({ ...prev, [platform]: Math.max(total, incoming.length) }))
-  }, [])
-
   // Cron *jobs* drive the sidebar "Cron jobs" section. Jobs are created
   // synchronously (agent tool call or the cron UI), so refreshing here right
   // after an agent turn surfaces a new job immediately; the interval poll keeps
@@ -384,9 +300,8 @@ export function DesktopController() {
       // with few recent sessions isn't windowed out of the cross-profile
       // recency page — the empty-history-on-profile-switch bug.
       const sessionProfile = profileScope === ALL_PROFILES ? 'all' : profileScope
-
       const result = await listAllProfileSessions(limit, 1, 'exclude', 'recent', sessionProfile, {
-        excludeSources: SIDEBAR_EXCLUDED_SOURCES
+        excludeSources: ['cron']
       })
 
       if (refreshSessionsRequestRef.current === requestId) {
@@ -402,8 +317,7 @@ export function DesktopController() {
 
     void refreshCronSessions()
     void refreshCronJobs()
-    void refreshMessagingSessions()
-  }, [profileScope, refreshCronSessions, refreshCronJobs, refreshMessagingSessions])
+  }, [profileScope, refreshCronSessions, refreshCronJobs])
 
   const loadMoreSessions = useCallback(() => {
     bumpSessionsLimit()
@@ -418,15 +332,12 @@ export function DesktopController() {
     const loaded = $sessions.get().filter(inKey).length
 
     const result = await listAllProfileSessions(loaded + SIDEBAR_SESSIONS_PAGE_SIZE, 1, 'exclude', 'recent', key, {
-      excludeSources: SIDEBAR_EXCLUDED_SOURCES
+      excludeSources: ['cron']
     })
 
     const keep = sessionsToKeep(key)
 
-    setSessions(prev => [
-      ...prev.filter(s => !inKey(s)),
-      ...mergeSessionPage(prev.filter(inKey), result.sessions, keep)
-    ])
+    setSessions(prev => [...prev.filter(s => !inKey(s)), ...mergeSessionPage(prev.filter(inKey), result.sessions, keep)])
 
     const total = result.profile_totals?.[key] ?? result.total ?? result.sessions.length
     setSessionProfileTotals(prev => ({ ...prev, [key]: Math.max(total, result.sessions.length) }))
@@ -687,20 +598,19 @@ export function DesktopController() {
     submitText,
     transcribeVoiceAudio
   } = usePromptActions({
-    activeSessionId,
-    activeSessionIdRef,
-    branchCurrentSession: branchInNewChat,
-    busyRef,
-    createBackendSessionForSend,
-    handleSkinCommand,
-    refreshSessions,
-    requestGateway,
-    resumeStoredSession: resumeSession,
-    selectedStoredSessionIdRef,
-    startFreshSessionDraft,
-    sttEnabled,
-    updateSessionState
-  })
+      activeSessionId,
+      activeSessionIdRef,
+      branchCurrentSession: branchInNewChat,
+      busyRef,
+      createBackendSessionForSend,
+      handleSkinCommand,
+      refreshSessions,
+      requestGateway,
+      selectedStoredSessionIdRef,
+      startFreshSessionDraft,
+      sttEnabled,
+      updateSessionState
+    })
 
   useGatewayBoot({
     handleGatewayEvent: handleDesktopGatewayEvent,
@@ -726,14 +636,10 @@ export function DesktopController() {
   // in the background (advancing next-run/state and creating runs), so poll the
   // job list on an interval (and on tab re-focus) while connected.
   useEffect(() => {
-    if (gatewayState !== 'open') {
-      return
-    }
+    if (gatewayState !== 'open') {return}
 
     const tick = () => {
-      if (document.visibilityState === 'visible') {
-        void refreshCronJobs()
-      }
+      if (document.visibilityState === 'visible') {void refreshCronJobs()}
     }
 
     const intervalId = window.setInterval(tick, CRON_POLL_INTERVAL_MS)
@@ -745,13 +651,6 @@ export function DesktopController() {
     }
   }, [gatewayState, refreshCronJobs])
 
-  useEffect(() => {
-    if (gatewayState === 'open' && !activeSessionId && freshDraftReady) {
-      void refreshCurrentModel()
-      void refreshHermesConfig()
-    }
-  }, [activeSessionId, freshDraftReady, gatewayState, refreshCurrentModel, refreshHermesConfig])
-
   useRouteResume({
     activeSessionId,
     activeSessionIdRef,
@@ -770,7 +669,6 @@ export function DesktopController() {
 
   const { leftStatusbarItems, statusbarItems } = useStatusbarItems({
     agentsOpen,
-    chatOpen,
     commandCenterOpen,
     extraLeftItems: statusbarItemGroups.flat.left,
     extraRightItems: statusbarItemGroups.flat.right,
@@ -791,7 +689,6 @@ export function DesktopController() {
       currentView={currentView}
       onArchiveSession={sessionId => void archiveSession(sessionId)}
       onDeleteSession={sessionId => void removeSession(sessionId)}
-      onLoadMoreMessaging={loadMoreMessagingForPlatform}
       onLoadMoreProfileSessions={loadMoreSessionsForProfile}
       onLoadMoreSessions={loadMoreSessions}
       onManageCronJob={jobId => {
@@ -809,35 +706,27 @@ export function DesktopController() {
     />
   )
 
-  // One PTY-backed terminal mounted forever; <TerminalSlot /> placeholders decide
-  // where it shows. Lives in main's stacking context (not the root overlay layer)
-  // so pane resize handles still paint above it. Toggling never rebuilds the shell.
-  const mainOverlays = (
-    <PersistentTerminal cwd={currentCwd} onAddSelectionToChat={composer.addTerminalSelectionAttachment} />
-  )
-
   const overlays = (
     <>
-      {!isSecondaryWindow() && <DesktopInstallOverlay />}
-      {!isSecondaryWindow() && (
-        <DesktopOnboardingOverlay
-          enabled={gatewayState === 'open'}
-          onCompleted={() => {
-            void refreshHermesConfig()
-            void refreshCurrentModel()
-            void queryClient.invalidateQueries({ queryKey: ['model-options'] })
-          }}
-          requestGateway={requestGateway}
-        />
-      )}
+      <DesktopInstallOverlay />
+      {/* One PTY-backed terminal mounted forever; <TerminalSlot /> placeholders
+          decide where it shows. Toggling fullscreen never rebuilds the shell. */}
+      <PersistentTerminal cwd={currentCwd} onAddSelectionToChat={composer.addTerminalSelectionAttachment} />
+      <DesktopOnboardingOverlay
+        enabled={gatewayState === 'open'}
+        onCompleted={() => {
+          void refreshHermesConfig()
+          void refreshCurrentModel()
+          void queryClient.invalidateQueries({ queryKey: ['model-options'] })
+        }}
+        requestGateway={requestGateway}
+      />
       <ModelPickerOverlay gateway={gatewayRef.current || undefined} onSelect={selectModel} />
-      <SessionPickerOverlay onResume={resumeSession} />
       <ModelVisibilityOverlay gateway={gatewayRef.current || undefined} onOpenProviders={openProviderSettings} />
       <UpdatesOverlay />
       <GatewayConnectingOverlay />
       <BootFailureOverlay />
       <CommandPalette />
-      <SessionSwitcher />
 
       {settingsOpen && (
         <Suspense fallback={null}>
@@ -925,6 +814,12 @@ export function DesktopController() {
     />
   )
 
+  const takeoverTerminalView = (
+    <div className="relative flex h-full min-h-0 min-w-0 flex-col overflow-hidden bg-(--ui-chat-surface-background) pt-(--titlebar-height)">
+      <TerminalSlot />
+    </div>
+  )
+
   // Flipped layout mirrors the default: sessions sidebar → right, file
   // browser + preview rail → left. Same panes, swapped sides.
   const sidebarSide = panesFlipped ? 'right' : 'left'
@@ -951,8 +846,6 @@ export function DesktopController() {
     <Pane
       defaultOpen={false}
       disabled={!chatOpen}
-      forceCollapsed={narrowViewport}
-      hoverReveal
       id="file-browser"
       key="file-browser"
       maxWidth={FILE_BROWSER_MAX_WIDTH}
@@ -969,56 +862,30 @@ export function DesktopController() {
     </Pane>
   )
 
-  const terminalPane = (
-    <Pane
-      defaultOpen
-      disabled={!terminalSidebarOpen}
-      divider
-      id="terminal-sidebar"
-      key="terminal-sidebar"
-      maxWidth="80vw"
-      minWidth="22vw"
-      resizable
-      side={railSide}
-      width="42vw"
-    >
-      <div className="relative flex h-full min-h-0 min-w-0 flex-col overflow-hidden bg-(--ui-editor-surface-background) pt-(--titlebar-height)">
-        <TerminalSlot />
-      </div>
-    </Pane>
-  )
-
   return (
     <AppShell
       leftStatusbarItems={leftStatusbarItems}
       leftTitlebarTools={titlebarToolGroups.flat.left}
-      mainOverlays={mainOverlays}
       onOpenSettings={openSettings}
       overlays={overlays}
-      previewPaneOpen={chatOpen && Boolean(previewTarget || filePreviewTarget)}
       statusbarItems={statusbarItems}
-      terminalPaneOpen={terminalSidebarOpen}
       titlebarTools={titlebarToolGroups.flat.right}
     >
-      {!isSecondaryWindow() && (
-        <Pane
-          forceCollapsed={narrowViewport}
-          hoverReveal
-          id="chat-sidebar"
-          maxWidth={SIDEBAR_MAX_WIDTH}
-          minWidth={SIDEBAR_DEFAULT_WIDTH}
-          onOverlayActiveChange={setSidebarOverlayMounted}
-          resizable
-          side={sidebarSide}
-          width={`${SIDEBAR_DEFAULT_WIDTH}px`}
-        >
-          {sidebar}
-        </Pane>
-      )}
+      <Pane
+        disabled={terminalTakeoverActive}
+        id="chat-sidebar"
+        maxWidth={SIDEBAR_MAX_WIDTH}
+        minWidth={SIDEBAR_DEFAULT_WIDTH}
+        resizable
+        side={sidebarSide}
+        width={`${SIDEBAR_DEFAULT_WIDTH}px`}
+      >
+        {sidebar}
+      </Pane>
       <PaneMain>
         <Routes>
-          <Route element={chatView} index />
-          <Route element={chatView} path=":sessionId" />
+          <Route element={terminalTakeoverActive ? takeoverTerminalView : chatView} index />
+          <Route element={terminalTakeoverActive ? takeoverTerminalView : chatView} path=":sessionId" />
           <Route
             element={
               <Suspense fallback={null}>
@@ -1055,13 +922,11 @@ export function DesktopController() {
       </PaneMain>
       {/*
         Order within a side maps to column order. Default (rail on the right):
-        main | terminal | preview | file-browser. Flipped (rail on the left):
-        mirror to file-browser | preview | terminal | main so terminal stays
-        adjacent to the chat.
+        main | preview | file-browser. Flipped (rail on the left): mirror it to
+        file-browser | preview | main so preview stays adjacent to the chat.
       */}
-      {panesFlipped ? fileBrowserPane : terminalPane}
-      {previewPane}
-      {panesFlipped ? terminalPane : fileBrowserPane}
+      {panesFlipped ? fileBrowserPane : previewPane}
+      {panesFlipped ? previewPane : fileBrowserPane}
     </AppShell>
   )
 }

apps/desktop/src/app/floating-hud.ts

@@ -1,22 +0,0 @@
-// Shared chrome for the top-center floating HUDs (command palette + session
-// switcher). They pin just under the title bar, centered, and lean on a crisp
-// border + shadow to separate from the app — no dimming/blurring backdrop.
-// Each caller layers on its own z-index, width, and overflow.
-export const HUD_POSITION = 'fixed left-1/2 top-3 -translate-x-1/2'
-
-// Matches the app's borderless-overlay surface (dialog, keybind panel, …):
-// hairline `--stroke-nous` paired with the soft `--shadow-nous` float.
-export const HUD_SURFACE = 'rounded-xl border border-(--stroke-nous) bg-(--ui-chat-bubble-background) shadow-nous'
-
-// One row/text size for both HUDs (compact — two notches under `text-sm`).
-export const HUD_TEXT = 'text-xs'
-
-// Shared item layout + padding for both HUDs. Tight vertical rhythm so rows
-// don't feel chunky; overrides the shadcn `CommandItem` default (`px-2 py-1.5`).
-export const HUD_ITEM = 'gap-2 px-2 py-1'
-
-// Section headings styled like the sidebar panel labels: brand-tinted, uppercase,
-// tightly tracked — plain text, no sticky chrome bar. Targets the cmdk group
-// heading via the universal-descendant variant.
-export const HUD_HEADING =
-  '**:[[cmdk-group-heading]]:static **:[[cmdk-group-heading]]:bg-transparent **:[[cmdk-group-heading]]:px-2.5 **:[[cmdk-group-heading]]:pb-1 **:[[cmdk-group-heading]]:pt-2.5 **:[[cmdk-group-heading]]:text-[0.64rem] **:[[cmdk-group-heading]]:font-semibold **:[[cmdk-group-heading]]:uppercase **:[[cmdk-group-heading]]:tracking-[0.16em] **:[[cmdk-group-heading]]:text-(--theme-primary)'