When a field accepted the AX value write but kept its old value, the
helper had already moved the caret, so the fallback keys or Cmd+V went
in at the wrong place.
* fix(computer): stop echoing written text and secure read-backs in action results
* fix(computer): treat a truncated set-value read-back as unverifiable
* fix(computer): accept code-point truncated read-backs as unverifiable
* fix(computer-use): capture Wayland windows with grim
* fix(computer-use): skip Wayland captures of windows that are not on screen
Sway and Hyprland report windows on other workspaces at coordinates that grim
would fill with whatever is visible there. Require the one matching window to be
visible, name why a capture was skipped, check for a supported compositor before
asking for grim, and load GdkPixbuf without needing Gdk.
* fix(computer-use): keep Wayland screenshots aligned with accessibility frames
Skip the capture when the compositor window size differs from the AT-SPI window
bounds, since element frames are measured against those bounds. Treat a running
Sway or Hyprland as Wayland even when XDG_SESSION_TYPE is tty.
---------
Co-authored-by: Kelvin Amoaba <97001695+AmoabaKelvin@users.noreply.github.com>
* fix(computer): stop macOS type-text and paste-text entering text twice
Chromium answers an AX value read from a cache that trails a successful
write by about 100 ms. type-text and paste-text read the field once right
after writing, saw the old value, and typed or pasted the same text again
as keys. set-value read back the same stale value and reported that a
landed write did not take effect.
Both now wait up to 500 ms for the readback to settle, and text input
falls back to keys only when the field still holds its old value. Text
input into Chromium web content skips the AX write and sends real keys,
because the write fires no beforeinput/input events (none at all for a
contenteditable), so editors such as Discord's search never saw the text.
* fix(computer): bound the macOS value readback wait by wall clock
* docs(computer): say why text readback polls for async AX trees
* feat(appearance): let app chrome follow the terminal theme
Add an "App Chrome Appearance" setting (Default / Match Terminal) next to
Left Sidebar Appearance. In match-terminal mode the terminal theme's
background/foreground are written as document-root CSS variables
(--bg-titlebar, --background/--card/--popover/..., both sidebar token
families), so the tab strip, status bar, side panels, full-page views and
popovers all follow the terminal instead of the app theme. Left-sidebar
modes still win because they scope their own variables below the root.
The terminal color pair and token derivation move into
terminal-surface-colors.ts and are shared with the sidebar resolver.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(status-bar): configurable usage format template
Add Settings → Appearance → Status Bar → "Usage format": a template that
replaces the built-in per-provider usage text when non-empty. Placeholders
cover {provider}, {plan}, the 5h/7d/fable/30d windows as percentages
(following Used/Remaining), {x.reset} countdowns, {x.resetAt} clock times
and Gemini {buckets}; "[...]" groups drop out when a placeholder inside is
empty so one template works across providers with different windows.
Whitespace is rendered as authored (whitespace-pre).
Persisted as PersistedUIState.statusBarUsageFormat with a byProvider
override map reserved for later; the RPC UI-update schema gains the
optional field.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* chore(appearance): document touched functions; surface usage format in search
Add one-line JSDoc to every function this branch adds or edits so the
review docstring gate passes. Include the usage-format entry in the
Status Bar section's search match so searching "template" or
"placeholder" no longer hides the parent row, and mention popovers in
the App Chrome Appearance descriptions since the setting styles them.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(appearance): let file editors follow the terminal theme in match-terminal mode
With App Chrome Appearance = Match Terminal, the tab strip, status bar and
side panels took the terminal palette but every file pane kept the stock app
theme: Monaco (code, plain text, diffs, notebooks, markdown source) was
hard-wired to `vs`/`vs-dark`, and the markdown preview / rich editor painted
the fixed `--editor-surface` with GitHub-colored hljs tokens.
- `terminal-editor-palette.ts` derives an editor palette from the active
terminal theme (background/foreground/cursor/selection + syntax roles mapped
onto the ANSI slots) and builds a Monaco theme (`orca-terminal`) from it;
light/dark is rated by the terminal background, so a light terminal theme
under a dark OS setting still gets light styling.
- match-terminal chrome variables now include `--editor-surface` and
`--syntax-*`; the hljs stylesheets read those ahead of their defaults.
- `useEditorSurfaceAppearance` / `useMonacoEditorTheme` replace the inline
`isDark ? 'vs-dark' : 'vs'` computations; the diff-section chain passes the
Monaco theme name instead of a dark flag.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VbWbChctXeMvaR2zdbFY99
(cherry picked from commit dc85fadee0a2664adbf55a49dd3c97c9b10fc111)
* fix(status-bar): honor the usage template in compact usage mode
ProviderSegment only rendered a user template on the verbose branch, so
switching usage mode to compact silently fell back to the built-in
tightest-window label. A non-empty template now replaces the built-in
text in both modes; the mini bar stays verbose-only.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VbWbChctXeMvaR2zdbFY99
* docs(appearance): document remaining functions touched by the diff
Adds one-line JSDoc to the editor surfaces, diff viewers, palette helpers
and UI-slice entry points this branch modifies so the diff's docstring
coverage clears the pre-merge threshold.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VbWbChctXeMvaR2zdbFY99
* fix(editor): keep Monaco theme registration and preview clock out of render
The upgraded oxlint React rules flag reassigning a module variable during
render and calling Date.now() inside useMemo. Move the terminal Monaco
theme's signature cache into ensureTerminalMonacoTheme() and pin the
usage-format preview clock with a lazy useState initializer.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VbWbChctXeMvaR2zdbFY99
* localization: complete Brazilian Portuguese catalogs
* fix(i18n): localize remaining hard-coded UI copy and complete Simplified Chinese coverage
Route the English literals and shared English labels that ignored the UI
language through the existing catalog: host labels, plugin cards and consent,
CLI and SSH errors, the checks pill, browser profile labels, shortcut groups,
app and context menus, and the macOS computer-use permission window. Shared
labels are translated where they render, so src/shared stays English for
mobile, CLI and older paired hosts.
- en.json gains 314 keys bootstrapped from inline fallbacks; four duplicate
"Unassigned" keys collapse into translateUnassignedLabel()
- zh gains 337 entries and refreshes existing ones; the other nine locales get
the newly extracted keys
- joinSentences ends a CJK sentence that has no final stop with 。 before the
next one
- the desktop caller passes the computer-use helper its locale; orcad and the
CLI stay Electron-free
- three files the copy pushed past max-lines each shed a focused module
* test: expect labeled roles in the editable context menu
The edit menu now passes explicit Cut, Copy and Select All labels so the roles render in the UI language; the upstream template assertion still expected bare roles.
* fix(i18n): key translated memos on the UI language without bare void reads
Upstream's React Compiler ratchet rejects `void language;`, which the compiler drops, and treats a react-hooks lint suppression as a bailout. The two host-label hooks now pass the language through readUiLanguageMemoInput so it stays a real memo input. ContextualTourOverlay drops its language key: getContextualTour is a static lookup, and step copy is localized at render.
* fix(i18n): translate editor theme settings
* Fix reactive localization and catalog extraction
* Fix combined localization validation
* Localize nested import failures and refine catalogs
* Handle older nested import responses
* Fix localized merge prompts and quota windows
* Address localization and runtime review feedback
* Keep host search and configuration examples usable
* Preserve configured names in add-project host selector
* Preserve host label origins across views
* Use localized fallback for missing local host
* Preserve custom names and refresh localized host labels
* Refresh task host labels without repeating searches
* Preserve Portuguese mobile tab translations
* Address localization review feedback
* fix(status-bar): import window visibility interval from its new path (#40951)
---------
Co-authored-by: lisi <lisi.see@bytedance.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Co-authored-by: marlico77 <marlonssoficial@gmail.com>
Co-authored-by: x-Ai <5061489+x-Ai@users.noreply.github.com>
Co-authored-by: Kelvin Amoaba <97001695+AmoabaKelvin@users.noreply.github.com>
* fix(computer): preserve synthetic Unicode typing on Linux
* fix(computer): bound typing across xdotool versions
* fix(computer): hold Linux keymaps through text delivery
* test(computer): verify missing Linux keys through requests
* test(computer): verify task event payloads
* fix(computer): verify Linux window restoration
* fix(computer): restore managed Linux windows
Exclude unmanaged helper windows from PID fallback without
losing minimized windows.
* fix(windows): start the WSL holder outside Orca's jobs so a crash doesn't take it down
* test(windows): prove the job closed with a non-breakaway control before checking the holder
* feat(computer): send tree diffs after actions and add get-app-state --query
Action results now carry only the accessibility-tree lines that changed since
the caller's last view of the same window (~ changed, + added, - removed as
index ranges, or an explicit no-change line), plus the focused element. The
runtime sidecar keeps the last tree per session/worktree and app, so all three
providers get it. The CLI opts in with treeDiff; older runtimes strip the
param and keep returning the full tree, and --full-tree opts out.
The macOS helper now keeps an element's index across snapshots of a window:
AX identity first, then the next unclaimed index with the same signature for
views that rebuild their AX objects, and never reuses a removed index.
get-app-state --query prints matching elements with their ancestors and
--max-elements caps the output, both in the CLI over the rendered tree.
Custom action names with raw newlines no longer split an element over lines.
* fix(computer): preserve parent changes in snapshot diffs
* fix(computer): refuse stale index reuse for lookalike controls
* fix(computer): print full trees in action replies
* fix(computer): keep controls located in snapshot replies
* fix(computer): fail set-value when its read-back shows a different value
A value_mismatch came back as ok with exit 0, so callers that check only
the exit status took an ignored or misplaced write as success. The runtime
now turns it into a verification_failed error at the sidecar boundary,
which covers every provider and every caller of the RPC.
Part of #33511
* fix(computer): pair macOS AX windows with their CGWindowID
AppKit and WebKit windows don't answer AXWindowNumber, so matching a
capture to its accessibility window fell back to frames. Two windows with
the same frame then paired the screenshot of one with the tree and title
of whichever AX listed first, under the first one's id. Read the id with
_AXUIElementGetWindow, as yabai and Hammerspoon do, and keep AXWindowNumber
as the fallback.
Fixes#27252
* fix(computer): fail fast when a macOS app stops answering accessibility
Every request runs behind one helper lock, and a hung app cost a timeout
per AX read, so one get-app-state on it held every agent for up to the 60 s
RPC timeout. Pin the AX messaging timeout at 1.5 s, refuse an app that
doesn't answer a root probe, and stop a tree walk at the first read that
waits out the timeout, each with an action_timeout error.
A press or write that the app received but didn't answer now fails the
same way, saying it may have taken effect. Before, the click fell through
to AXConfirm and then a synthetic click, so a button whose handler blocked
ran twice, and type-text typed the text again as keys.
* fix(computer): report macOS accessibility actions that changed nothing
AXPress and other accessibility actions came back as plain unasserted
success even when the app ignored them, as Chromium does for a role=button
whose handler sits on a child. After such an action the helper now compares
the target's tree, frame and on-screen windows with the snapshot it took
just before. If they still match after re-reading for about a second, the
result says unverified/suspected_noop; if they change late, the result
carries the settled state instead.
The new reason is additive; older CLIs print it like any other reason.
Refs #31758
* fix(computer): propagate late macOS observation failures
* fix(computer): select existing text before replacement
* fix(computer): share native macOS window identity
* fix(computer): leave Linux and Windows windows in the background for accessibility actions
Element clicks no longer restore and activate the target window when an
AT-SPI action or UIA Invoke/SelectionItem/Toggle pattern handles them; only
the pointer fallback activates. On Linux, type-text and paste-text insert
into the focused editable field through AT-SPI EditableText without needing
window focus, and fall back to key synthesis or the clipboard otherwise.
Refs #12412
* fix(computer): refuse pointer input after failed activation
* fix(computer): stop after unconfirmed text writes
* fix(computer): insert the full UTF-8 text on Linux
* test(computer): focus each Linux text-entry trial
* fix(computer): verify Linux text mutations before deletion
AT-SPI can acknowledge writes without changing text. Read back
mutation counts and caret state before reporting success.
* fix(computer): verify complete Linux text replacements
* fix(computer): decline unreadable Linux text selections
* feat(computer): drive macOS apps in the background by default
Clicks no longer activate and raise the target first. Pointer and key
input go to the target window's pid through a new BackgroundInputTransport:
mouse and scroll events carry the window id in field 51 and a window-local
location, keys run inside a focus-without-raise key-window scope, and
chords carry no Unicode string so menu key equivalents fire. Scroll and
drag now land (#27306); typed text no longer doubles in Electron (#17942).
The HID tap and focus fence stay for --restore-window only. Results
report action.delivery, the handshake reports supports.input.background,
and a missing private symbol or a popup click fails with
background_unavailable instead of falling back to the foreground.
* test(computer): read typed key strings back instead of diffing live-state events
* fix(computer): keep Unicode characters in one key event
* fix(computer): preserve active macOS key windows
Refuse key-window changes in the active app and surface failed
focus records, including cleanup after partially delivered input.
* fix(computer): guard macOS click routes before delivery
* fix(computer): let paste events read the clipboard
* fix(computer): classify clicks in the requested macOS window
* fix(computer): fence macOS scroll and drag recipients
* Support Windows native Antigravity account credentials
* Preserve Antigravity prompt arguments and sanitize Windows file failures
* Preserve Windows credential bytes and distinguish native read failures
* Import Buffer type in Windows security declarations
* Expose native credential bytes without a Node type dependency
* Let Windows agy launches keep shell ownership
* fix: declare Windows account messages and cover owner scope
* fix: resolve Windows account authority before snapshot projection
* fix: share Windows credential parsing with saved account reader
* test: type saved account identity lookup mock
* test: isolate native account fixture host detection
* test: reuse filesystem mock for container marker control
* test: preserve native credential contract failure diagnostics
* test: normalize private read expectation and cover Windows boundary
* fix(windows): start the WSL holder with only a NUL handle to inherit
libuv starts every Windows child with bInheritHandles=TRUE and no handle
list, so a detached holder that outlives Orca kept any inheritable handle
Orca had open, such as a launcher's pipe or a DevTools socket. Start it
through CreateProcessW with PROC_THREAD_ATTRIBUTE_HANDLE_LIST naming one NUL
handle for its stdio. The build probes now check the newest export without
calling it, and the lane receipt labels ConPTY's own startup output.
* fix(windows): report isolated holder exits and kills the way libuv does
* fix(windows): keep Orca's std handles out of every spawned child
libuv spawns with bInheritHandles=TRUE, so an inheritable std handle reached
long-lived children such as the WSL holder and the daemon and held a
launcher's pipe open after Orca exited. Clear HANDLE_FLAG_INHERIT on the std
handles at the top of main startup; stdio 'inherit' still works because
libuv hands the child its own inheritable duplicate.
* test(windows): model Orca's Electron main, which keeps std handles inheritable
Plain Node clears its std handles' inherit flag at startup, so a Node child
never reproduced the leak. Electron main opts out of that, which is why Orca
leaks and why the control arm now runs Electron; a plain-Node arm pins the
difference.
* test(windows): say why the std-handle tests need longer timeouts
* fix(computer-use): keep Linux element indexes unique after browser tab compaction
Dropping inactive tabs shrank the records list mid-walk, so later nodes took len(records) and reused a retained tab's index; elementParam then acted on the tab.
* fix(computer-use): keep Windows element indexes unique after browser tab compaction
top costs ~263 ms of CPU per sample, so the memory panel only refreshed
footprints every 10 s. The new orca-process-footprint helper reads
phys_footprint for every same-user process in ~3 ms, so footprints now
refresh on the panel's 2 s poll. top stays as the fallback when the helper
is missing or fails.
Co-authored-by: m4air <m4air@Mac.localdomain>
SendKeys presses Shift for an uppercase letter (VkKeyScan reports the
shift state), so `orca computer hotkey --key CmdOrCtrl+A` (the documented
example) sent Ctrl+Shift+A on Windows. Lower-case a single-character key
when the hotkey has modifiers; Shift is still sent when written explicitly.
Extract ConvertTo-OrcaSendKeysHotkey so runtime-render.test.ps1 can assert
the SendKeys strings case-sensitively.
Fixes#27589
Co-authored-by: jespo2021 <joeespo@gmail.com>
Co-authored-by: Neil <neil@stably.ai>
* fix(computer-use): send correct named keys and hotkey letters on Linux
press_key passed the keysym for named keys (Return, Tab, arrows, Escape)
to AT-SPI as PRESSRELEASE, which expects a hardware keycode, so a
different key was pressed. Synthesize with SYM instead.
hotkey forwarded upper-case letters such as CmdOrCtrl+A to xdotool,
which reads them as shifted keysyms and sends Ctrl+Shift+A. Lower-case
single letters in key combinations.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* test(computer-use): drop the unreachable lone-key hotkey case
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com>
set_value looked up node.is_editable_text before attempt() ran. GI bindings
from at-spi2-core < 2.53 (e.g. Ubuntu 24.04) do not expose that function, so
the AttributeError escaped and every set-value failed. Probe with
get_editable_text_iface, which has been bound since 2.9.2 and checks the same
interface.
Closes#10569
Co-authored-by: bbingz <zzb@gxsmjx.com>
Windows managed hook skip paths (Devin/Grok/Cursor replay guards, and orca-agent-hook.exe) exit immediately instead of draining stdin through more.com, so a held-open writer can no longer hang the hook.
Fixes#24907
Supersedes #24909
Co-authored-by: inforanger <29155427+inforanger@users.noreply.github.com>
pnpm dev launches the app as com.stablyai.orca.dev, but the macOS helper's
isTrustedOrcaApplication only trusted com.stablyai.orca, com.github.Electron
and ids starting with com.stablyai.orca.dev. (with the trailing dot). Every
computer-use request from a dev build failed with "computer-use agent peer is
not authorized".
Move the bundle id rule into TrustedOrcaApplication in
OrcaComputerUseMacOSCore so it is tested, and trust com.stablyai.orca.dev
exactly as well. Look-alikes such as com.stablyai.orca.devtools stay
untrusted.
Fixes#26498
Co-authored-by: petersindex <287557875+petersindex@users.noreply.github.com>
Co-authored-by: Neil <neil@stably.ai>
* test: remove redundant cases and repair behavioral assertions
* docs: reconcile audit records with current main
* fix(deps): apply security updates from #32101
Apply the six-file DOMPurify 3.4.17 and Next.js 16.4.1 patch from PR #32101 to resolve the dependency security gate for this test audit. Original patch by Brennan Benson.
* test: satisfy changed-line quality gates
Add required braces and replace filter indexing with find. Rerun both CI quality gates and the 27 affected relay tests; refresh audit hashes and record the corrections.
---------
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com>
Co-authored-by: Brennan Benson <79079362+brennanb2025@users.noreply.github.com>
* fix(orcad): ship computer use with the server and say when it is unavailable
* test(computer): drop the type assertion in the sidecar app-environment mock
* fix(cli): carry the server reprovision hint into --json computer failures
* fix: refresh host environments before structured agent command resolution
* fix: ship the owning server CLI for managed remote agents
* test: include the shipped CLI in remote install fixtures
* Bind packaged agent CLI calls to their execution host
* Keep owning-server CLI binding scoped to managed agents
* Serve the last shell environment to chat starts and refresh it in the background
Only the first chat start waits for the login shell; later starts get the last
snapshot and refresh it in the background once it is 10 s old. On Windows hosts
without the registry addon (managed server slots), the saved PATH is reread
with reg.exe. The Windows server CLI launcher stager now checks the target
architecture, and the CLI help says ambient selectors are ignored inside a
managed server.
* Keep the last good shell environment when a refresh fails, and load the Windows profile once
A login-shell capture now says whether the shell produced the env or Orca fell back
to its own, and only a real capture replaces the snapshot. Windows loads the
PowerShell profile once per process and refreshes only the saved PATH. The reg.exe
fallback queries the whole key so a missing user Path no longer drops the saved PATH.
The first capture starts when the resolvers are created.
* Keep the last good login-shell env for other readers while a forced refresh runs
* Mark the Claude launch resolver as a reviewed router-aware module
It only types the chat command resolver it injects; main's launch-routing guard (#26801) matched the name.
* Keep a settled login-shell env for readers during a refresh and share the first run
- Any settled cached entry, a fallback included, now serves other readers while a forced capture
runs; only a real capture replaces it.
- Native chat's first capture shares a run a startup reader already began, so the login shell
runs once per process at startup on macOS/Linux.
- Type the injected Claude resolver without naming resolveClaudeCommand.
- Cover the runtime's Claude command options, the owning-host --host check, and the PTY flag drop.
* Drop an inherited owning-host flag from daemon terminal panes
The persistent terminal daemon merges its own environment into every pane, so an inherited
ORCA_CLI_OWNING_HOST reached desktop panes even though main strips it from the spawn env.
Drop it with the other per-launch CLI variables.
* Route agents to their own server through ordinary launch defaults; start the first chat after an install
The native-chat agent environment now sets the owning server's ORCA_USER_DATA_PATH and
empty ORCA_ENVIRONMENT / ORCA_PAIRING_CODE / ORCA_REMOTE_PAIRING, which the CLI already
reads as unset; empty rather than deleted, because the spawn merges the launch overlay
over the server's own env. The ORCA_CLI_OWNING_HOST flag, the CLI's skip of ambient
selectors, the launchers' export of it, the terminal clean-up, the help-text exception
and the WSL launcher guard are gone, so a person's terminal follows their own settings
as on main.
A launch that finds no program in the shell snapshot waits for one shared refresh (at
most one shell timeout) and resolves again; create support rereads the shell once
before refusing. On Windows a failed PowerShell profile load is retried on the next
refresh instead of being kept until restart.
* One shared shell reread when an agent's program is missing
The pre-create check for OpenCode, OMP and Pi forced its own login-shell
capture on every refusal: not shared with launches, one per concurrent
check, triggered by an installed-but-too-old version too, and an older
capture finishing last could overwrite a newer cached environment. On
Windows, every launch miss also reran a failed PowerShell profile load,
bypassing the 10-second limit.
- One saved shell environment per host: chat launches, their misses, the
pre-create check and AI commit messages read it and share one refresh.
- The pre-create check resolves the agent's program the way a launch does,
so only a missing program waits for the refresh; registrations declare
how to resolve their launch and check its version.
- A forced login-shell capture joins one already running for the same
shell, so captures cannot land out of order.
- On Windows, a miss within 10 s of a failed PowerShell load rereads only
the saved PATH.
* One owner for the cached shell environment
Review found two more races in the reread-on-miss machinery, both from the
host wrapper keeping its own snapshot beside the login-shell cache:
- A launch that found no program could join a reread that started before the
agent was installed, and spend its one retry on the old PATH.
- On Windows, saved-PATH rereads renewed the snapshot's age, so a failed
PowerShell profile was never retried while launches kept missing.
The login-shell cache is now the only owner of the captured environment, the
run in progress and which result wins (the newest run started; a fallback
never replaces a captured env). It takes three options instead of `force`:
- startedAfter: a lookup that found nothing waits for a run that started no
earlier than the lookup did, joining one if it has, else starting one.
- refreshAfterMs: serve the last result and rerun in the background once the
last attempt is older.
- keepCaptured: a captured env is kept for good; only a fallback is rerun
(Windows' PowerShell profile).
The host wrapper keeps only policy: the 10 s interval, on Windows the saved
PATH laid over the profile env on every read (a miss forces a fresh registry
read), and the settings overlays. Its snapshot, pending promise, staleness and
Windows "loaded" bookkeeping are gone. A lookup's start time travels with the
miss (resolveAgainAfterRefreshOnMissingProgram passes it).
* Read an injected shell environment once per runtime, as main does
The previous commit read an injected environment on every acquisition, which
broke main's 'caches shell exports but re-reads configured overrides for a
resume' contract. It is read once when the resolvers are made again.
* Model lists reread the environment on a missing program, as launches do
Live QA: the first new chat after installing Claude showed "Claude wasn't
found on the computer running this chat". The chat launch itself reread the
login shell and started the new binary, but Claude's model-list check ran
first, with the environment read before the install, and never retried:
catalog probes were deliberately built without the miss refresh.
The retry now lives in each agent's one shared lookup of its binary and
environment, which a launch and a model listing both call:
resolveCodexStructuredInvocation, resolveAcpLaunchInvocation, a new shared
resolvePiRpcInvocation, and resolveClaudeChildEnvSources (which already had
it). Each catalog probe is given the runtime's refreshAfterMissingProgram,
and the catalog context no longer strips it, so a listing and a launch
reread the same single-owner environment the same way.
* Pass optional launch lookups through directly instead of conditional spreads
* Read the host shell env through resolveLoginShellEnvironment; drop the extra wait cap
CI on the merge with main failed main's new codex-workspace-relative-path
test (#26684), which mocks the login-shell module with only
resolveLoginShellEnvironment: the wrapper called captureLoginShellEnvironment
and imported the spawn-timeout constant. The wrapper only needs the env, so
it reads through resolveLoginShellEnvironment again, as on main.
The miss wait no longer races its own 5 s timer: the shell run it waits for
is already killed at the spawn timeout, so the second cap was a guard over
the same bound. The constant goes back to private, as on main.
* Test that a hung login shell holds a launch miss for at most one shell timeout
The bound lives in the one owner: the shell run is killed at the spawn
timeout (5 s) and settles to a fallback, which never replaces a captured
snapshot. A mutation raising that timeout fails this test.
* A program found through a relative PATH entry is not missing
Main's #26684 runs Codex through a workspace-relative PATH entry (`bin`,
`node_modules/.bin`). The lookup searches such an entry against Orca's own
directory and returns the bare name; the spawn then finds the program in the
chat's directory. Counting that bare name as missing made every such launch
wait for a login-shell reread.
The retry now fires only when the lookup truly found nothing: a bare name back
from a PATH whose entries are all absolute. Each agent's one lookup tells it
which PATH it searched.
* Count a relative PATH entry as found only when it holds the program; Windows miss waits for a saved PATH read started after it looked
A bare name with a relative PATH entry is checked from the launch's working
directory, so an agent in neither entry still gets its one reread. On Windows the
miss no longer joins a registry read that began before the lookup.
* feat(windows): add administrator PowerShell terminal setting
* fix(settings): localize administrator terminal search keywords
* fix(windows): restrict elevated terminal host to Orca's runtime and requesting account
* fix(windows): harden administrator terminal routing, inventory and restore
* fix(build): build the Windows agent hook when electron-builder packages Windows
* fix(build): build the packaged Windows agent hook for the target architecture
* fix(windows): accept the development terminal runtime and test real elevation in CI
* feat(windows): show administrator status and restart instructions
* fix(settings): localize administrator mode search keywords
* test(settings): isolate the desktop permission section in shell picker tests
* Add Windows administrator restart and simplify Terminal settings
* Keep administrator restart silent and settings actions within narrow panes
* Keep Windows shell choices and permission guides usable at minimum width
* Prevent later restarts from replaying administrator terminal reset
* Fix CI route resolution and transient Windows fixture cleanup
OrcaWinandOrcaWin
Kelvin Amoaba
lucumaandKelvin Amoaba
Neil
Brennan Benson
Wooseong Kimandbbingz
