Two Windows-only desktop boot bugs that caused spurious reinstall/repair loops:
1. findOnPath() searched the empty extension BEFORE PATHEXT, so an
extensionless Git-Bash `hermes` shim shadowed the real hermes.cmd/.exe.
The shim then failed the shell:false --version probe and the resolver
fell through to bootstrap/repair even though a working CLI was on PATH.
Fix: try PATHEXT extensions first, keep the empty entry LAST so callers
that already include the extension (py.exe, pwsh.exe) still resolve.
2. handOffWindowsBootstrapRecovery() chose the destructive --repair over the
gentle --update by checking only venv\Scripts\hermes.exe -- the setuptools
console-script shim, written at the END of venv setup and absent in
interrupted/quarantined states. Fix: take --update when ANY real-install
signal is present (venv python, the shim, or .hermes-bootstrap-complete).
Adds windows-hermes-resolution.test.cjs (source-assertion pattern, wired into
test:desktop:platforms) guarding both regressions.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A broken/empty Windows launcher venv can see the source tree via PYTHONPATH
but lack PyYAML, so 'import hermes_cli' succeeds while the first real CLI
import dies — the desktop then trusts the bootstrap marker, spawns a dead
backend, and loops on 'gateway offline' (#52378).
- backend-probes.cjs: canImportHermesCli now runs 'import yaml; import
hermes_cli.config' (extracted as hermesRuntimeImportProbe) and accepts an
env override, so a dependency regression is caught without a real broken
venv fixture.
- main.cjs: isBootstrapComplete() routes through new isActiveRuntimeUsable(),
which requires the venv python to pass the runtime import probe (with
ACTIVE_HERMES_ROOT on PYTHONPATH) — not just exist on disk.
Salvaged from PR #38179. The PR's install.ps1 reset/clean + autocrlf changes
and their tests are dropped: current main already preserves dirty checkouts
via stash (the data-loss-safe #38542 path) rather than the PR's older
reset-based Repair-ManagedCheckoutBeforeUpdate approach.
`npm run build` ended with `bundle-electron-main.mjs`, which esbuild-bundled
electron/main.cjs and renamed the bundle on top of the tracked source file.
Because every `hermes desktop` runs `npm run build`, each launch rewrote a
checked-in source file (~7.5k-line source -> ~14.8k-line bundle), dirtying the
working tree with a build artifact that `git restore` couldn't keep (the next
launch re-clobbered it) and forcing autostash/restore conflicts on update.
The bundle only existed to inline `simple-git` so the packaged app.asar (which
ships no node_modules) wouldn't crash at launch with "Cannot find module
'simple-git'". Replace it with the mechanism the repo already uses for the
other hoisted runtime dep (node-pty): stage the dependency closure and resolve
it from process.resourcesPath at runtime.
- stage-native-deps.cjs: resolve simple-git's runtime closure (walking
dependencies + optionalDependencies, so a version bump that adds a transitive
dep can't silently reintroduce the crash) and stage it under
build/native-deps/vendor/node_modules/. The `vendor/` nesting is load-bearing:
electron-builder drops a node_modules dir at the ROOT of an extraResources
copy but keeps a nested one.
- git-review-ops.cjs: fall back to the staged
native-deps/vendor/node_modules/simple-git when the hoisted require() fails;
dev runs resolve the hoisted copy and never hit the fallback.
- package.json: drop the bundler from the `build` script so main.cjs is never a
build target again.
- nix/desktop.nix: drop the direct bundler call (the closure rides the existing
`cp -rn native-deps` into $out) and patch process.resourcesPath in
git-review-ops.cjs alongside main.cjs.
- delete scripts/bundle-electron-main.mjs.
Verified: electron-builder's own file filter keeps the full staged closure
(0 dropped), and a packaged win-unpacked build launches with the git-review
pane resolving simple-git from the staged vendor path.
The plain-Linux overlay re-enable (#53185) left nativeOverlayWidth() at 0
for plain Linux, so the native min/max/close buttons painted on top of the
app's right-edge titlebar tools. Reserve the fallback width everywhere the
WCO overlay is painted (Windows, WSLg, plain Linux); macOS still reserves 0
since it uses traffic lights.
Commit da5484b61 disabled the Window Controls Overlay on all Linux
(non-Windows, non-WSL) with the note that WCO is a Windows/macOS-only
Electron feature. However, several Linux compositors (KDE/KWin,
GNOME/Mutter) do support it — plain Electron titleBarOverlay paints
native min/max/close buttons that were working before that change.
Narrow the exclusion to only WSLg, where the RDP host draws its own
window controls and an Electron overlay would leave a dead gap.
Fixes: da5484b61 ("fix(desktop): WSL2 clipboard image paste + Linux titlebar overlay")
The MoA reference-block display (each reference model's output shown as a
labelled thinking block before the aggregator responds) previously existed
only in the classic CLI. The facade already emits moa.reference / moa.aggregating
through tool_progress_callback; this wires the TUI and desktop consumers.
- tui_gateway/server.py: _on_tool_progress relays moa.reference (label / text /
index / count) and moa.aggregating to the Ink/desktop client as their own
events.
- ui-tui: gatewayTypes adds the two event shapes; createGatewayEventHandler
routes them; turnController.recordMoaReference pushes a committed
thinking-style segment tagged with the source model. Shown regardless of
showReasoning — references ARE the mixture-of-agents process the user opted
into, not ordinary reasoning. moa.aggregating is a status-only transition
(no transcript entry).
- apps/desktop: use-message-stream appends each reference as a labelled
reasoning chunk via the existing reasoning disclosure; GatewayEventPayload
gains label/index/aggregator.
Tests: tui_gateway emit (3), Ink handler render + showReasoning-independence +
aggregating-no-segment (3). TUI typecheck/lint clean; desktop typecheck/lint
clean.
* fix(windows): stop terminal-window popups from background spawns
Native-Windows desktop/gateway users saw cmd/conhost windows flash on
gateway restart, image paste, the dashboard Projects tree, voice notes,
and ~5 min after closing the app (detached cron). Two root causes:
- Console-subsystem exes (taskkill, schtasks, wmic, netstat, tasklist,
agent-browser, git, ffmpeg, powershell, git-bash) spawned via raw
subprocess allocate a fresh console when the launching process has
none (pythonw desktop backend / detached gateway) - even with output
captured.
- uv venv pythonw shims re-exec console python.exe, so Python children
get a console regardless of how they're launched.
Fixes:
- Single hidden-spawn primitive (_subprocess_compat.run/.popen) that ORs
CREATE_NO_WINDOW on Windows, no-op on POSIX. Route every Hermes-owned
console-exe spawn through it.
- FreeConsole() catch-all in hermes_bootstrap: any Python child that
exclusively owns an auto-allocated console detaches it at startup
(GetConsoleProcessList()==1 gate leaves shared interactive consoles
untouched).
- Replace PowerShell/wmic gateway PID scans with in-process psutil.
- Skip schtasks queries on non-interactive desktop restarts.
- Prefer native agent-browser .exe over .cmd shims.
- Guard test bans raw subprocess spawns of the Windows-only console
tools repo-wide so the popup class can't regress.
* fix(windows): scope FreeConsole to background entry points; fix merge fallout
Console detach review (per #53810 feedback): GetConsoleProcessList()==1 can't
tell a uv pythonw->python phantom console apart from a user opening the
interactive CLI/TUI in its own fresh console (double-click, shortcut, ConPTY) —
both report a single attached process with a tty. Running FreeConsole() in the
import-time bootstrap therefore risked detaching a legitimately-interactive
terminal.
- Extract FreeConsole into explicit hermes_bootstrap.detach_orphan_console();
remove it from apply_windows_utf8_bootstrap() (import side effect).
- Call it only from known background mains: gateway run, dashboard backend
(start_server, what the desktop spawns), cron standalone, tui_gateway entry,
slash worker. Interactive CLI/TUI never calls it.
- Behavior-contract tests: frees only when solo owner, leaves shared console,
no-op without console / on POSIX, and asserts it's not an import side effect.
Merge fallout from origin/main (#53791):
- local.py: 3-way merge left a dangling **_popen_kwargs (NameError crashing
every terminal init). _subprocess_compat.popen already hides the window, so
drop it.
- discord adapter: merge stacked an undefined windows_hide_flags() onto the
primitive call; drop the redundant arg.
- test_gateway: scan now goes psutil-first (zero spawn); rewrite the
case-variant test to drive that production path.
* test(claw): mock _subprocess_compat.run seam for Windows process scan
claw.py's Windows tasklist/powershell scan routes through the hidden-spawn
primitive; the tests still patched claw_mod.subprocess, so on win32 the mock
was never hit and real spawns returned nothing. Patch the actual seam.
A Radix <Select> renders a blank trigger when its `value` matches no
<SelectItem>. The Settings model pickers built their options solely from
each provider's curated `models` list, so a model added via config that
isn't in that list (e.g. anthropic/claude-opus-4.7 on nous) selected
nothing and showed an empty selector.
Union the active value into the options via a small `withActive` helper,
applied to the main, auxiliary, MoA reference, and MoA aggregator model
selects so the configured model always stays visible and selectable.
On the desktop Channels / Messaging page, the "Open setup guide" button was
rendered as a bare <a href={platform.docs_url} target="_blank"> with no guard.
Plugin-provided platforms (Microsoft Teams, Google Chat, Line, Raft, Yuanbao,
…) ship an empty docs_url, so the anchor's href was "".
In a packaged build, Electron resolves an empty href against the current
document — the app's own index.html inside the asar bundle — and
shell.openPath then fails with an OS "file not found" dialog. This is exactly
the Windows error reported for Messaging → Teams → Open guide.
Fix (3 changes):
1. fix(desktop) — Only render the "Open setup guide" button when docs_url is
non-empty, and route clicks through openExternalLink so a relative/empty
value can never be treated as a local bundle path. Fixes the whole class
(every plugin platform), not just Teams.
2. fix(messaging) — Give the Teams platform plugin a real docs_url (Microsoft
Teams setup guide) so its card shows a working button instead of nothing.
3. fix(messaging) — Give the Google Chat platform plugin a real docs_url
(Google Chat setup guide) so its card shows a working button instead of
nothing. Originally from #48940; folded in here because that PR's test
was broken (it queried the HTTP endpoint, but google_chat is a dynamic
enum member that only appears after the adapter module is imported).
Test plan:
- apps/desktop — new src/app/messaging/index.test.tsx: button is hidden when
docs_url is empty; a real URL opens via the validated external opener (does
not navigate).
- apps/desktop typecheck (tsc --noEmit) clean.
- backend — test_teams_messaging_metadata_links_setup_guide: the Teams catalog
entry exposes the setup-guide docs_url.
- backend — test_google_chat_messaging_metadata_links_setup_guide: the Google
Chat catalog entry exposes the setup-guide docs_url.
Co-authored-by: xxxigm <tuancanhnguyen706@gmail.com>
Co-authored-by: p-andhika <andhika.prakasiwi@gmail.com>
The desktop MoA settings 'Add preset', 'Set default', and 'Delete' buttons
mutated local React state only and never called the save endpoint, so a newly
constructed preset vanished on refresh. Each now builds the next config and
calls saveMoa() so the change is written to config.yaml via PUT /api/model/moa.
A MoA preset whose reference or aggregator slot points at the moa virtual
provider creates a recursive MoA tree. The runtime guards in moa_loop.py only
surface this mid-turn (references silently skipped, aggregator raises). Reject
it at the config chokepoint (_clean_slot) so it can never be saved, and hide it
from the desktop/dashboard slot pickers so it isn't offered as a dead choice.
Embeds reach out to third parties on render, so default to a placeholder that
mirrors the tool-approval UX: "Load <service>" (this embed) or "Always allow
<service>" (persisted). A desktop-local store ($embedMode ask|always|off +
per-service allowlist) gates the fetch with zero gateway round-trip; an
Appearance setting controls the global default. Local renderers (mermaid, svg,
alerts) are never gated. Addresses review feedback on outbound third-party
requests.
A clarify/approval/sudo/secret prompt blocks the turn on the user, but the UI
treated it as an in-flight turn: the "thinking" timer kept ticking and Esc
interrupted the run — discarding a question you might want to come back to. Add
$activeSessionAwaitingInput (the pet's awaitingInput concept, scoped to the
active session) and use it to suppress the stall indicator and disarm Esc while a
prompt waits. Clear the session's prompts (and needsInput) on Stop and on turn
end so a resolved/aborted turn can't leave a dead panel or a stuck "needs input"
dot.
The inline clarify panel used its own card tokens, an animated ring, and
oversized spacing — out of step with every other tool row. Rebuild it on the
shared --ui-*/--conversation-* tokens: a compact panel, letter-key badges
(A/B/C…) that double as a/b/c… shortcuts, an inline content-sizing "Other" field
(CSS field-sizing — no view swap, no layout shift on focus), and a Continue
button so picking an option selects rather than auto-sends. Selection lives on
the letter badge alone (solid primary; outlined while Other is focused-but-empty).
Also settle the panel into the standard tool block once the turn stops running,
so a stopped turn no longer strands a live, unanswerable prompt.
Add a content-agnostic Zoomable primitive (useZoomPan hook + overlay viewer):
click to open full-screen, wheel-zoom toward the cursor, drag to pan, toolbar
zoom/reset, and an optional copy action. Wire Mermaid diagrams into it with
copy-as-PNG; reusable for other inline content later.
Wire the embeds module into the markdown surface: bare provider autolinks unfurl
to inline embeds, ```mermaid/```svg fences route to the rich renderers, and
`> [!NOTE]`-style blockquotes become alert callouts. Labeled links stay plain.
Per-kind renderers, each a lazy split chunk: plain-iframe video/maps (wheel
chains to the transcript; maps gate scroll behind ⌘), the in-document
blockquote-script path for X/Instagram, the dark Spotify player, and the
YouTube iframe. Adds Mermaid and DOMPurify-sanitised SVG fences and GFM alert
callouts, all sized to 33dvh and theme-matched to avoid white color-scheme
artifacts. Main-process stamps a Referer on YouTube embed requests.
Pure, synchronous URL→descriptor matchers for YouTube, Vimeo, Instagram,
Pinterest, TikTok, X, Spotify, Google Maps and OpenStreetMap, plus the shared
embed primitives (error boundary, fail card, escape-html, dark-mode hook,
sizing token). Declares the mermaid + dompurify deps used by the fenced
renderers.
Bring apps/desktop and ui-tui to a clean state for typecheck, eslint,
and prettier:
- Run prettier across both trees (printWidth/wrap drift; prettier is not
CI-enforced for these JS projects, so main had accumulated drift).
- Apply eslint --fix for padding-line-between-statements and perfectionist
import/export sorting.
- Manual fixes for non-auto-fixable rules:
- remove unused node:net import in electron/main.cjs (uses Electron net)
- replace inline `typeof import(...)` annotations with top-level
`import type * as EnvModule` in two ui-tui test files
- scoped eslint-disable no-control-regex on intentional sentinel/ANSI
regexes (mathUnicode.ts, text.ts)
- resolve react-hooks/exhaustive-deps per-case: correct swapped/missing
deps, collapse redundant session.* members, and justified disables on
settings mount-only data-load effects to preserve run-once behavior
No behavior changes; test pass/fail counts are unchanged from the main
baseline.
Alt+wheel now scales about the pixel under the pointer instead of growing from a
corner, so the pet stays put under the cursor instead of running away. In-window
shifts its top-left; the overlay repositions its OS window (cursor-anchored on
wheel, bottom-center for slider-driven changes).
Hold Alt/Option and scroll over the mascot to resize it (same on Mac and
Windows); the modifier keeps a plain scroll passing through to the page. The
gesture drives the same `display.pet.scale` path as the settings slider.
The popped-out overlay grows its OS window to fit the pet at any scale (anchored
bottom-center) so the sprite is never clipped by the window edge, and the
in-window pet re-clamps against its actual size so growing near an edge can't
crop it. Also makes the overlay click-through per-pixel: only solid sprite
pixels (plus bubble / mail button) are interactive, transparent margins pass
clicks through.
Live-measure WCO width in the renderer, drop the right rail below the titlebar
band, and re-enable GPU compositing under WSLg when /dev/dxg is present.
WSLg bridges clipboard text but not images — pull host screenshots via
PowerShell. Disable titleBarOverlay on plain Linux; gate overlay width per
platform in titlebar-overlay-width.cjs.
When idle with a top-level delegate_task still in flight, render a static,
shimmering system-note at the transcript tail instead of a spinner (which
reads as "stuck"). Reuses the shared steer / slash-status chrome (centered,
0.6875rem, muted, Codicon) so it sits in the thread like every other meta
line, and mirrors the primary child's latest stream line, falling back to
generic copy. i18n across en/ja/zh/zh-hant; markdown prose/heading rhythm
tuned so a re-entered turn breathes.
Track top-level delegate_task work that dispatches in the background and
re-enters as a fresh turn. $backgroundResume returns {count, activity} for
the active session while idle — count of parked tasks plus the primary
child's latest stream line (tool/progress/thinking) when readable.
Use the app's amber warn color for the unsaved-edits tab dot (was inheriting
the label text color) and add a tab-bg ring + soft drop shadow so it stays
legible where it overlaps the filename.
Extends the pane store with heightOverride (alongside widthOverride) and a
get/set/clear API, and wires the pane shell + desktop controller so the
bottom-row terminal pane can be resized on the Y axis with its size persisted.
Adds a CodeMirror 6 spot editor to the right-rail file preview so users can
make quick edits in-app without leaving for an IDE. Entering edit mode is a
pure in-place swap of the read view — same fixed-height header, same gutter
geometry/typography (mirrors SourceView 1:1) so nothing shifts — toggled via
the Edit button, a bare `e` when the pane is hovered/focused, or the tab.
- Save path is transport-agnostic (writeDesktopFileText): local Electron IPC
or a new hardened POST /api/fs/write-text on the dashboard server (path
validation, parent-must-exist, regular-files-only, size cap, atomic
temp-file + os.replace), behind the existing auth middleware.
- Stale-on-disk guard re-reads before writing and offers overwrite vs
discard-and-reload instead of clobbering external/agent edits.
- VS Code-style modified dot on the tab; ⌘/Ctrl+S and ⌘/Ctrl+Enter save,
Esc cancels; GitHub highlight style matched to the read view's Shiki theme.
- Typing stays render-free (draft in a ref; dirty flips once at the boundary).
