Global setup and the worker-scoped fallback now share one seedTestRepository()
that runs its commit and worktree add with per-command
`-c core.hooksPath=/dev/null -c commit.gpgsign=false`, so a contributor's
global hooks or always-on signing no longer abort e2e before any spec runs,
and the test repo's own hook config stays untouched for specs that install hooks.
Fixes#31587
Co-authored-by: Neil <neil@stably.ai>
getClaudeAccountLabel returned 'System default' and 'Claude account' as
English literals, so the account-switch toast, switch-sessions dialog and
restart toasts showed them untranslated inside otherwise localized copy.
Route both through translate() with new accounts.claude keys, reusing
each locale's existing Accounts pane wording for System default.
Fixes#30101
Co-authored-by: Claude <noreply@anthropic.com>
sqlite3 3.54+ (macOS 27) prints a styled, versioned prompt instead of
`sqlite> `, so the #13767 repro tests timed out waiting for it. Start
sqlite3 with `-cmd ".prompt 'sqlite> '"` so the prompt is deterministic
on every sqlite3 version.
Fixes#27052
Co-authored-by: outof-place <contact@outofplace.space>
A restart that replaces a hidden pane's live PTY declared automatic-restore,
so after main stopped the old PTY the daemon deferred the 9th and later
restarts on live_cap indefinitely. Declare these restarts (and their deferred
retries) as user-resume, and drop back to automatic-restore once the spawn is
admitted so later wakes stay capped.
Fixes#34285
Co-authored-by: Neil <4138956+nwparker@users.noreply.github.com>
Read git ls-files and git diff --name-only with -z so non-ASCII paths are
not C-quoted and still match ast-grep's real filenames; otherwise new
flaky-test findings in such files were silently dropped.
Fixes#34395
Co-authored-by: Neil <neil@stably.ai>
In a drive-letter workspace, strip the leading slash from URL-style
/X:/ file hrefs before resolving, so links like /D:/Temp/report.txt
open D:/Temp/report.txt. POSIX, WSL UNC, and UNC hrefs are unchanged.
Fixes#34310
Mermaid draws every flowchart label in the theme's text colour, so a
`style` or `classDef` fill can hide it: light text on a pale node in dark
mode, dark text on a deep one in light mode. After rendering, labels that
fall below 4.5:1 on their node's fill switch to whichever theme colour
reads better, on desktop previews, chat and the phone.
Co-authored-by: Nico Hu <haoyu.hu@lexmount.com>
* feat(mobile): copy button on shell command rows in chat
* test(mobile): stub the copy button where chat tests count tool row presses
* fix(chat): decode the login-shell script the way the shell does
Codex joins argv by switching quote styles, so stripping only the outer
quotes left \" escapes and quote splices in about a fifth of real
commands. Argv arrays are now shell-quoted when joined.
* fix(chat): keep the shell-path match to the outer command; reuse the label's full command
Argv arrays go back to a plain join: no argv-shaped shell calls exist in
real transcripts, and quoting them broke PowerShell and cmd.exe argv.
* test(mobile): leave quote decoding to the shared unwrap test
* refactor(chat): slice double-quoted runs and name the row's command fields plainly
* style(mobile): format copy button mocks in chat row tests
* refactor(mobile): share the queued-card projection with desktop
Image-only queued drafts showed a blank card on the phone.
* fix(mobile): honor the host's Queue follow-ups setting
With it off, a mid-turn send from the phone still queued while desktop sent it into the running turn.
* fix(mobile): keep a queued-card edit when the card leaves the queue
The phone closed the editor and dropped the typing; it now goes to the chat box as on desktop.
* fix(mobile): hold /clear and /compact for background work and sends
The phone fired the command where desktop holds it, showed start failures twice, and left the agent unnamed in failures.
* fix(mobile): hand a queued edit back to the chat it came from
* fix(mobile): scope a command's send hold to its own chat
* fix(mobile): read a published null queue as empty
* refactor(mobile): share the returned queued-card caption rule
* test(mobile): drop queued-card tests the shared rules already own
* refactor(native-chat): share the needs-attention rule and drop leftover wrappers
* fix(mobile): type the structured sends mock, drop the moved queued-cards path and ship queued edit copy at boot
* fix(native-chat): drop the queued card's unused image-label import after the rebase
* fix(mobile): order the phone's queue reorder preview by position, as the shared cards sort
* test(mobile): keep the queued-messages test within its line budget
* fix(mobile): settle terminal chat working state and turn timing with desktop's rules
Move desktop's live-status merge, Stop suppression, hook-state resolver and
terminal turn timing to src/shared/native-chat. Mobile reads the transcript
lifecycle the host already sends, so a provider turn end settles Working the
same way desktop does. Terminal chats now get the turn bar, timer and folding,
and the streaming preview uses the shared deriveNativeChatStreamingText.
* fix(mobile): send desktop's terminal Stop keys, a second Escape only for OpenCode
Desktop sends one Escape on Stop. It adds a paced second Escape only for
OpenCode, which confirms an interrupt that way. Claude Code and Codex read
Esc-Esc at an idle prompt as rewind. The rule moves to
native-chat-agent-support, and the phone now paces the second Escape at
desktop's answer step instead of 80ms.
* fix(mobile): word Orca-stop and usage-limit notices like desktop
Move desktop's Orca-stop sentences to src/shared with an English table that
desktop translates. The phone names the Orca host it pairs with, and when
that host is unnamed it says "the remote server". Usage-limit rows rebuild
desktop's sentence from the failure fact. The usageLimitLineText fallback,
which only the phone used, is deleted.
* test(mobile): pass the overlay's host name in the prompt controller harness
* test(mobile): re-record Stop goldens at desktop's step, split status-notice tests, pin the turn clock and ship notice copy at boot
* fix(mobile): offer structured Stop while a send is out and read queue-sends-next as working
Move desktop's structured Stop rule, its Stop-joining helpers and the
queue-sends-next rule to src/shared. Both sides now use them. The phone
counts its in-flight sends, so a conversation-stop host offers Stop before
the turn opens. Between queued turns the phone reads as working, and
commands that can't wait in line hold like they do on desktop. Untargeted
Stops are no longer joined on legacy hosts.
* fix(mobile): offer terminal Stop again when its Escape went nowhere
Stop hides Working on the phone like it does on desktop. The phone's Escape
can fail outright, though, while desktop writes to a local PTY. When no Escape
could have landed, the phone now drops that suppression. The streaming preview
also treats a pending echo as the transcript's tail, as desktop does, so an
earlier reply can't hide a new preview.
* fix(mobile): hold structured Stop while a send it covers is in flight
On desktop, Stop's held and Stopping state depends on whether the chat can
stop, and that includes a send still in flight. The phone now does the same.
Once Stop is pressed before the turn opens, the button holds and repeated
presses send no more untargeted cancels.
* fix(mobile): end the terminal Stop hold when a new message is sent
Desktop clears Stop's working suppression on an optimistic send, so a new turn
that keeps the same hook epoch (Claude with working subagents) still offers
Stop. The phone now does the same when a new pending echo appears.
* refactor(mobile): trim chat turn wiring after the desktop rule move
Drop the controller's unread nativeChatAgentWorking field. Re-render only when
a chat starts or stops having a send out. Keep the structured turn indicator
stable across stream frames. Type the Stop test table instead of casting.
* refactor(mobile): track structured sends in flight once for Stop and command holds
* test(mobile): re-record Stop goldens now that Stop settles with its Escape delivery
* fix(mobile): escape terminal image paths like desktop drops
* fix(mobile): keep launch-draft prefill over older user turns
* fix(mobile): rank @ file matches with desktop's fuzzy Quick Open search
* fix(mobile): use desktop's composer triggers for mid-prompt / and $ picks
Picked @ files are now quoted when the path has spaces.
* fix(mobile): send picked skill paths on structured chats
* fix(mobile): list the host's disk-scanned skills in the / and $ menus
The phone names the workspace and the host resolves its folder, refusing SSH
workspaces as desktop does; hosts without skills.discover-workspace.v1 list none.
* fix(mobile): forget a picked skill once its token is deleted
* fix(mobile): drop a picked skill as soon as its token is deleted, and retry failed skill scans
* fix(mobile): treat a picked leading skill as chat and stop failed skill scans from refiring
* test(mobile): drop the menu ranking case the shared picker tests own
* refactor(mobile): keep skill picks on the draft record and reuse shared send rules
A rejected send now keeps its picks and leading-skill origin on both send paths.
* fix(mobile): replace a re-picked skill's file and restore picks with a rejected send
* fix(mobile): read the chat agent for the skill catalog from the active resolution
* style(native-chat): format the composer autocomplete files
* fix(native-chat): import the picker's apply and send-classify rules from the shared composer module
* refactor(mobile): import the composer's draft-store hooks once
* style(mobile): format the notice row's style import
* fix(mobile): resolve terminal chat agents in desktop's order
Mobile now uses the shared detected-then-launch precedence and transcript-readability check; the phone's remembered agent only fills in after them.
* fix(mobile): hide Codex foreground commands from the background-task strip
* refactor(native-chat): share diff-card gutter, basename and task count
* fix(mobile): show clipped tool output sizes in KB/MB like desktop
* refactor(native-chat): share the tool-run header rules with mobile
* fix(native-chat): share tool-line detail rules; skip unsafe-only search hits
* fix(mobile): copy message text with desktop's rule
Copy now leaves out roster and background-task fallback twins and wordless provider frames.
* refactor(native-chat): memoize shared tool-line view and output sizes; share result preview
* fix(native-chat): read a lone result's preview from its narrowed body
* chore(i18n): ship chat tool and stop copy in the runtime English catalog
* refactor(native-chat): move the pending-send echo rule to shared
Mobile reuses it instead of its own matcher.
* fix(mobile): match chat echoes after their send boundary, sink unmatched ones
Uses desktop's rule: rows before the send no longer retire an echo, a send that
never lands no longer lets later glued sends retire, and an unmatched echo sits
below later turns.
* fix(mobile): keep glued photo captions and folded boundaries in echo matching
* fix(mobile): give image preview echoes the send time the occurrence rule reads
* fix(mobile): match echoes on image-normalized rows and cap open sends at eight
Same input and bound as desktop; each open send rescans the transcript.
* test(mobile): drop a pending-order case the tail test already covers
* refactor(mobile): read the send boundary once and skip echo normalization when idle
* test(native-chat): type the occurrence case's pending send
* test(mobile): drop the pending-echo wall-clock entry from the flaky-test baseline
* refactor(native-chat): share session option labels with mobile
Mobile's options pill now shortens effort levels (Med, XHigh, Min) and names
choices and options the way desktop does, from one English table that desktop
translates.
* refactor(native-chat): run mobile terminal chat options on the desktop surface
The terminal-chat option surface, its appliers, the option cache and the
dispatch types move to src/shared, and the phone builds its pickers from them.
Grok and every other agent with a catalog now get model and effort pickers on
the phone, picks are saved as the host's launch defaults, the OMP live switch
gate and discovery mapping come from one copy, and authoritative discovery
retires persisted models it no longer lists. The option stamp moves to shared.
* refactor(mobile): merge the session option type imports
* fix(mobile): keep terminal chat option picks on their own chat
The phone keeps one option surface per chat, so a pick that lands after leaving
and returning shows there, and a picker command only switches the chat it came
from. Discovered models merge with the seed catalog as on desktop, so Grok keeps
its effort picker. A Claude model switch the phone cannot confirm is never
saved as a launch default.
* fix(mobile): start terminal chat options over when a tab changes agent
A tab that runs another agent drops its kept option surfaces, report markers
and cached record, so a fresh session shows its own model. A pick that lands
after the phone moved to another chat is not saved as a launch default.
* fix(mobile): fence retired terminal chat option surfaces
A write that lands after its tab started another agent no longer publishes the
old agent's record, and a changed OMP switch capability rebuilds the chat's
surface from the cache instead of reviving an older one.
* refactor(native-chat): simplify the shared session option wiring
Desktop translation pieces use literal English defaults so the localization gate can read them, the selected-choice lookup is shared once, and the phone keeps one option surface per tab with the agent change detected from the cached record.
* refactor(mobile): return no option surface for an agent without a catalog
* test(native-chat): expect the shared Effort label in the startup picker test
* test(native-chat): keep only the label cases main still checks after the rebase
* refactor(native-chat): import the session option snapshot rules once in the PTY option surface
* refactor(native-chat): share structured option pick bookkeeping with mobile
Desktop and the phone now track picks per option through one shared book: a
repeat of an option on its way waits, other options go together, a lost reply
is not a pick done, and a stale reply only commits its own option. The phone
also re-reads options each turn and after the provider starts, and shows picks
at once while the host is still starting the provider.
* refactor(native-chat): share the option pick blocking rule
* fix(native-chat): read in-flight option picks from the shared book for launch send options
* fix(native-chat): keep the structured options hook within its line budget and pass the phone turn in the e2e harness
* fix(native-chat): list a resting Claude chat's / commands from the account it will start on
A resting chat's next start follows the selected Claude account, but its / menu was read from the account saved when the chat was created. Read it from the same account a new chat would be created on, and rescan when the selection changes.
* refactor(native-chat): resolve the at-rest account in the shared catalog so its key matches the scan
* test(native-chat): build the second at-rest chat record from a lease fixture
The title and icon-link regexes rescanned the rest of the body from every
unclosed opener, so a 192 KB page of repeated '<link ' or '<title>' blocked
the main or runtime event loop for seconds. Scan tags forward only.
* fix(mobile): build the Source Control action sheet from desktop's menu rows
Adds Force Push and desktop's order, enable rules and hints.
* fix(mobile): rebase from desktop's effective base and keep the Force Push warning in the action sheet
* fix(mobile): keep the rebase row's base when the branch compare fails
* fix(mobile): keep upstream compares working when the default base lookup fails
* fix(mobile): keep the rebase target apart from the branch compare so failed or unsupported compares keep it
* fix(mobile): leave force pushes to desktop when the linked review has its own push target
* refactor(mobile): rebase onto the shown base and share the force-push block with the primary action
* fix(windows): probe fnm's real default alias dir for Node fallback
On Windows, fnm stores the default alias under FNM_DIR, else
%APPDATA%\fnm (legacy ~\.fnm), and node.exe sits directly in
aliases\default with no bin subdir. Probe those dirs and append them to
PATH so fnm's default only backfills a missing Node instead of shadowing
an inherited one.
Fixes#11356
Co-authored-by: niceSprite <16523864+iFwu@users.noreply.github.com>
Co-authored-by: Claude <noreply@anthropic.com>
* test: keep onTestFinished callbacks returning void
---------
Co-authored-by: Neil <neil@stably.ai>
Co-authored-by: Claude <noreply@anthropic.com>
Scroll-spy pages rewrite the URL fragment with history.replaceState as
you scroll. The webview navigation handlers and setBrowserPageUrl treated
those same-document updates as a new document, retiring saved-note
badges and cancelling an open comment card. Fragment-only in-place
updates now keep the geometry; document navigations and in-place route
changes (path or query) still retire it, and the store no longer guesses
document boundaries from URL changes.
Without --enter, --text reaches the PTY unmodified, so integrators can send
Escape or Ctrl+U directly. With --enter and without --interrupt, prompts to a
recognized running agent are bracketed-pasted and embedded Escape is made inert.
Fixes#14032
Co-authored-by: Neil <neil@stably.ai>
An unexpected ACP agent exit reported only the whole-run stderr tail, so a
startup warning read as the cause, and the observed exit code/signal was
dropped. Add the exit code or signal to the ended reason and Details, file
the stderr tail as a log diagnostic like the Claude and Codex lanes, and
read the exit status from ManagedProviderProcess.exitStatus through a
shared formatter that Claude reuses.
Fixes#33035
Co-authored-by: Brennan Benson <79079362+brennanb2025@users.noreply.github.com>
Co-authored-by: Neil <neil@stably.ai>
QuickCommandsPane, useAddRepoCloneFlow, UntitledFileRenameDialog and the
workspace board column resize hook guarded a render-phase setState with a
ref written in the same render. When a sibling suspends, React discards
that render and its queued updates but keeps the ref write, so the retry
skipped the adjustment (no Add dialog, empty clone destination, stale
save-as name, stale column width). Track each guard in state instead, and
sync the column-resize handler refs in a layout effect.
The other two sites from the issue (WorktreeMetaDialog prevIsOpenRef and
use-native-chat-draft lastScopeKey) no longer exist on main: the dialog
remounts a keyed draft and the chat draft reads via useSyncExternalStore.
Fixes#25527
Co-authored-by: Matheus Breguêz <matbrgz@gmail.com>
Jcode accepts a string or an array of strings per hook event, but Orca's
[hooks] parser only understood single-line strings, so any array made
status and install report error and no hook was written.
Parse string arrays, append Orca's command as the last element (after the
user's pre_tool gates), repair a stale Orca element in place, collapse
duplicate managed elements on reinstall, and remove only Orca's element on
uninstall. Edits are offset splices on the original text, so user entries,
comments and layout survive; malformed or non-string arrays still fail
closed.
Fixes#33998
Co-authored-by: Neil <neil@stably.ai>
Codex paginated rollouts persist an item_completed ContextCompaction item
with no content, and legacy rollouts persist an event_msg context_compacted.
The Codex decoder dropped both, so the native Chat timeline never showed
that context was compacted. Emit a system row with the shared compaction
copy and presentation so it renders as the compaction divider.
Fixes#14188
Co-authored-by: Neil <neil@stably.ai>
The details tokenizer's start hook matched any `<details` substring, so
marked cut a paragraph at a `<details>` mention inside inline code and its
HTML-block rule swallowed the rest, rewriting the file on save. HTML blocks
open only at a line start, so anchor the hook there.
Fixes#19799
Co-authored-by: Neil <neil@stably.ai>
The observed (`worker-start --setup run`) and sequenced (`wait-for-setup`) setup
commands are typed into the user's interactive shell. Pair-inserting line
editors such as zsh-autopair rewrite the `( ` of the `( bash runner )` wrapper
into `( )`, so bash -lc receives a stray `)` and rejects the whole line: setup
never runs, no completion marker is printed, and the worker stays at
setup_state: running (or the agent gate waits out its timeout).
The runner command is always a single external command (bash/cmd.exe/
powershell.exe), so the subshell added nothing; run it directly.
Fixes#18059
Co-authored-by: Neil <neil@stably.ai>
Codex can render a separate `? for shortcuts` row below its status row. The
composer classifier returned on the first nonblank bottom row, so it never
examined the real status row above the hint and the IME placeholder mask
never activated, leaving the placeholder beside composing text.
Fixes#24337
Co-authored-by: Neil <neil@stably.ai>
Co-authored-by: Andy Zhang <16486422+v01dstar@users.noreply.github.com>
Local Setup named only Node, pnpm, and Bun. A first pnpm install also
compiles node-pty and the native addons with node-gyp and loads them in
Electron, and *.electron.test.ts fall back to xvfb-run on headless Linux.
Document the per-platform build toolchain, the Linux Electron runtime
libraries plus xvfb, and the Corepack/pnpm version caveats.
Fixes#31589
Co-authored-by: Neil <neil@stably.ai>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Automatic hibernation checks and stops each agent pane on its own and only
protects terminal tabs that are visible on screen, so other tabs in the
active worktree can still sleep. Update the hibernation page to match:
drop the "hibernate together as a unit" claim, scope the foreground rule
to visible tabs (active tab, other split groups, Activity page), separate
manual whole-worktree sleep from automatic per-pane reclaim, and describe
resume as happening when you return to the agent's tab.
Fixes#33940
Co-authored-by: Neil <neil@stably.ai>
The orchestration skill description only advertised coordinator work, so a
dispatched worker never loaded it and built its heartbeat loop from the
preamble alone. Neither the preamble nor the guide said to stop a background
heartbeat loop before worker_done, so the loop outlived the task.
Add worker-role triggers (heartbeats, ask, worker_done) to the description and
stub, and tell workers in the preamble and guide to restart an expired
heartbeat loop mid-task and stop it before worker_done.
Fixes#26605
Main now tags each Orca-authored PR/MR create failure with a reason code
(the English error is unchanged for mobile, older clients and logs), and
the renderer maps the reason to translated copy in the Source Control
notice, the Checks panel and the pre-create blocked notice. Sign-in steps
are single-sourced across providers.
Fixes#33321
Co-authored-by: Neil <4138956+nwparker@users.noreply.github.com>
Hardened-runtime TCC refused EventKit and Contacts requests from terminal-launched
tools without a prompt: Orca and its terminal host lacked the addressbook and
calendars entitlements and the matching purpose strings. Declare them the same
way camera, microphone and location already are; the terminal host inherits both.
Fixes#12933
Co-authored-by: Claude <noreply@anthropic.com>
Japanese IMEs convert one clause at a time, but the terminal's preedit overlay
painted the whole composition uniformly, so the clause the candidate window
targets was invisible. Read the IME's selection from xterm's textarea on the
insertCompositionText input event (Chromium only updates it there) and paint
that range of the preedit in the theme's selection colors. Cell-grid advance
corrections are unchanged and nothing reaches the PTY; a collapsed caret or a
range outside the preedit highlights nothing.
Fixes#27873
Co-authored-by: Neil <neil@stably.ai>
Younghwan ChaeandNeil
NeilandClaude
MananandNeil
Blue DaydreamingandNeil
BingZ
OrcaWinandOrcaWin
Kelvin AmoabaandNico Hu
Théophile DiotandNeil
erishandNeil
Frederic BarthelemyandNeil
mmarabelandNeil
berceandNeil
Seiji KoharaandNeil
Takuya AkiyaandNeil