* feat(serve): add bind address option
* fix(serve): validate service bind address
* fix(serve): validate bind flag values
* fix(serve): reject scoped bind addresses
* docs(serve): document bind option paths
* docs(serve): note that --mobile-pairing needs an all-interfaces bind
* fix(serve): move bind address parsing to shared so the packaged CLI can load it
The packaged CLI ships out/cli and out/shared but not out/main/orcad, so
cli/runtime/serve-bind-flag.js failed the packaged runtime-import check
(and would fail to load) when it imported main/orcad/orcad-bind-address.
Move the module, unchanged, to src/shared and point orcad, Electron serve
and the CLI at it.
* test(serve): pin down the default listener and the --bind startup paths
- Extract the Electron serve listener pin into resolveServePinnedBindHost
and test that serve without --bind stays unpinned (all interfaces, as
before), an explicit --bind pins it from either argv form, and the orcad
browser sidecar stays on loopback.
- Golden the orcad and Electron serve argv without --bind against the
pre---bind output.
- Test that the CLI passes bind only when given and refuses --bind with
--mobile-pairing before launching.
- Guard that serve-bind-flag and the shared serve validation load only CLI
and shared modules, which is what the packaged CLI ships.
---------
Co-authored-by: lw <13579959+bertramlongwuyan@user.noreply.gitee.com>
Co-authored-by: Neil <neil@stably.ai>
Co-authored-by: xuanchangfeng <210147232+xuanchangfeng@users.noreply.github.com>
Nothing in the repo reads DYLD_* variables, so the entitlement only weakens Hardened Runtime for the signed app and its inherited helpers. allow-jit stays because the terminal-host Node helper requires it.
Co-authored-by: Brijesh Vora <brijesh.vora@corelight.com>
Co-authored-by: Neil <neil@stably.ai>
* fix(hooks): run a created worktree's setup once its own content is approved
Adds worktrees:runApprovedSetup (IPC) and worktree.runApprovedSetup (RPC,
capability worktree.run-approved-setup.v1). The host re-hashes the worktree's
setup and default-tab commands, runs them only on a match, and otherwise
reports the new content. After a create mismatch the host opens only the
command-less default tabs; approving in the trust prompt starts setup and the
command tabs. Clients whose host lacks the capability keep setup skipped.
* fix(hooks): review a created branch's different setup from the workspace, not instead of it
The trust prompt now says the workspace was created and its branch's setup
differs; approving runs it for that workspace without replacing the primary
checkout's approval (which made the next create prompt twice). Declining shows
a toast whose Review setup action reopens the prompt. Repo-wide trust returns
the content's hash so the host can run it.
* fix(worktrees): import approved setup from its path after the RPC methods move
* test(e2e): wait for the shell-ready deadline after a startup exec
Since #30978 the terminal host no longer probes for a prompt when a profile
exec drops Orca's ready marker; it submits the startup command once at the
15 s shell-ready deadline. The oracle allowed only 8 s, so the headless serve
case failed unless the exec barrier happened to run long. Wait until that
deadline (measured from terminal creation) plus the recovery budget, and
route the spec from the shell-ready barrier and shell startup sources.
* ci(e2e): route the startup-exec oracle helper to its spec
The spec forged an SSH owner for a local file and relied on the read
falling back to local transport. Since #31617 a tab reads only from its
captured owner, so the forged tab correctly shows 'Couldn't verify which
host owns this file'. Use a paired client that opens the same checkout
locally and on its runtime host instead, and route editor model identity
sources to the spec.
A link whose page is already open now reuses that tab (#32907), so the paged
preview's click of the same URL never added the second tab the spec counted.
Also route CSV viewer sources to the CSV specs.
* feat(cli): attach existing worktrees to folder workspaces
worktree set accepts --parent-worktree folder:<id> (hosts advertise
worktree.set.folder-parent.v1), keeping each worktree to one parent in the
store. --parent-worktree active/current resolves to a folder workspace from
its terminal or enclosing directory, and worktree create without --repo under
a folder workspace lists the folder's Git repositories with a ready command.
* fix(cli): report folder parents and keep typed flags in folder repo choices
worktree show/set report a folder parent; the missing-repo error lists --repo
choices to add to the user's command and names the Attached worktrees panel.
Load and import repairs skip malformed rows and only touch imported worktrees,
an import keeps a worktree's existing parent, a twin-owned folder parent warns,
and folder ambiguity keeps its choices.
* fix(persistence): settle one parent per imported worktree before writing
An Orca server migration import now decides each touched worktree's parent
before any write: an explicit folder parent replaces an inferred worktree
parent, any other disagreement refuses the import. --parent-worktree active
in a folder terminal still lists the folder's repositories.
* fix(persistence): let a folder parent beat hydrate's inferred worktree parent
Load repair and import reconciliation share one rule: an inferred worktree parent loses
to a folder parent when the folder was chosen explicitly or the worktree parent carries
hydrate's orchestration stamp. Import tests now run payloads through the manifest parser
and cover a conflict that appears after staging.
* fix(cli): list folder repos when active resolves to the enclosing folder
worktree create without --repo used only the terminal's env to find the folder, so
--parent-worktree active run outside an Orca terminal fell back to the generic
missing-repo error. Use the parent create already resolved.
permission dropdown, and fresh profiles keep OpenCode on its own settings, so
the agents without Auto read Mixed after Full access -> Approve for me. Assert
that, track OpenCode in the stored-permission checks, and route the spec on
permission settings sources.
Slash and mention suggestion menus in the phone chat composer stay reachable above the on-screen keyboard while typing.
Co-authored-by: Guderlilian <73724370+qq2144@users.noreply.github.com>
Claude launches now get --plugin-dir prepended (subagent navigation, #33322),
so the stand-in read '--plugin-dir' as its marker path and the spec never saw
the detached child. Route the spec on descendant-kill and Claude launch-flag
sources.
* fix(hooks): run worktree setup only when its content matches the approved script
Clients may send approvedSetupHash with a create. After git worktree add the
host hashes the created branch's shared setup and default-tab commands with
the prompt's function; on a difference it skips them and returns
setupTrustMismatch with the branch content. Creates without a hash (older
clients, CLI, mobile, orchestration) and older hosts behave as before.
* fix(hooks): send the approved setup hash and re-prompt on a changed branch
The composer and direct work-item creates send the hash of the setup content
the user approved (empty content included; none under repo-wide trust or
local-only policy). When the host reports different branch content, show a
setup-skipped toast and the existing trust prompt with that content.
* test(hooks): hash the prompt's and the host's orca.yaml from the same fixture
Covers CRLF and BOM primary checkouts, the plain-HTTP digest, the hash a
runtime host reports to remote and mobile clients, run-both and local-only
policies, creates without a hash, and pins the digest so a format change is
deliberate.
* style(worktrees): keep the creation flow under the line limit
Since #34683 a move journal promotes the WSL hosts gate to serve-only, so startup
recovery now runs and undoes a move interrupted before blobs-transferred. The
spec restarted at source-fenced and expected the fence to survive; assert the
undo there, fence from blobs-transferred, and expect the committed refusal to
name the host (#34880). Route fence and recovery sources to the spec.
Co-authored-by: Claude <noreply@anthropic.com>
Terminal padding now insets the bottom edge too (#36789), so the terminal sits
paddingY above the pane bottom. Route pane geometry sources to the spec.
The one-time 'This page renders on your desktop' tour opens beside the address
bar of the first client-hosted page and intercepts clicks on the address-bar
suggestion. Seed it as seen for this spec via a new paired-client launch
option, and route address-bar conversion sources to the spec.
Since #31080 a client keeps one saved entry per server, so pairing the same
host a second time under another name is refused. The steady 'Direct host' is
now a different server than the interrupted one. Tooltip checks name the host
so a closing tooltip from the previous card cannot match. Route the spec from
its pairing-store and runtime-status sources.
Board drops send wakeWslDistro (#36230) and network failures toast localized
copy (#34621). Route GitHub Projects sources to the board spec so PRs run it.
* fix(browser): require a per-session token on the local CDP proxy
* fix(browser): reset the helper session when the CDP proxy refuses it
Explain a 403/404 handshake from Orca's CDP proxy and retire the session so
the next command gets a fresh proxy and helper. Drop an inherited
AGENT_BROWSER_CDP from the helper env so only Orca names the target, and pin
the env contract and minimum bundled agent-browser version in tests.
* fix(setup): show the detected command for every setup import source
* fix(setup): review imported setup scripts in a dialog instead of inline
Imported setup and archive scripts can be long, so the setup card now shows
the first command and a Review setup script button. The dialog shows the full
setup and archive scripts and is where non-package-manager imports are saved.
Package-manager suggestions stay inline and editable as before.
* fix(setup): use compact dialog title and description sizes
* fix(plugins): drop a plugin's approval when its catalog moves it to another repository
* plugins: revokePluginConsent refuses when the plugin service is not running
* cli: orca plugin install/update ask again when a plugin moves to another repository, for every plugin kind
* Support Gemini Enterprise authentication for Antigravity Chat
* Keep Antigravity account settings current across clients
* Type the mutable account snapshot fixture
* Keep account settings readable during Antigravity sign-in
* Use standard account form controls and localized placeholders
* Refresh Antigravity account settings after host reconnect
* Annotate the account form fixture without a cast
* Keep Antigravity auth settings at the ACP boundary
* Exercise host settings in the Enterprise sign-in regression
* Save visible Antigravity account settings before sign-in
* Remove obsolete account Apply copy and correct renderer test types
* Put Antigravity sign-in choices before the action
* Keep unchanged account fields editable on blur
* Preserve Antigravity sign-in status beside account choices
* Fix sign-in status predicate type narrowing
* Reject unconfirmed Antigravity account settings
* Show a shared Antigravity account failure once
* Type the Antigravity publication candidate
* Pair Antigravity settings failures with meaningful recovery
* Register account failure copy and refresh the schema catalog
* Keep the latest Antigravity save outcome ahead of delayed reads
* Type the retained account normalization fixture directly
* Keep RPC recovery coverage within its project boundary
* Keep Antigravity settings recovery in the Sign in action
* Keep unavailable Antigravity choices out of the recovery form
* Keep Antigravity settings reads stable across renders
* Retain the ACP task fixture deletion option type
* Hide unavailable Antigravity sign-in choices after a load failure
Since #33622 a confirmed HTML preview link honors openLinksInApp (default off,
so it opens the system browser) and no longer calls
openBrowserProfileTabInActiveWorkspace, so the spec's store spy never fired.
Opt the client into in-app links and observe the opened page URL directly.
Route doc-preview sources to the spec.
Orca already drops protected strings when it loads a language pack and keeps
its own English text for trust, permission and destructive dialogs.
validate:plugin and orca plugin validate still rejected the whole pack, so the
community directory refused otherwise-good packs. They now report how many
protected strings are ignored as a warning; every other pack rule stays strict.
* Redesign plugin detail pages with publisher discovery
- Show publisher provenance, related plugins, and clearer source links
- Preserve installed release attribution and README links across catalog changes
- Support navigation between plugin pages while preserving review queues
* Fix publisher link button styling
* Format resolved installer test
* Remove unrelated scratch files
* feat(mobile): sign Android release builds with a dedicated release key
Add an Expo config plugin that gives the generated app/build.gradle an
env-driven release signing config, and have the Android release workflow
require the keystore secrets, fail before building when they are missing,
and check the APK's signing certificate against a pinned fingerprint.
* fix(mobile): keep the signing setup hint intact in the workflow error
* test(mobile): type the plugin import without a cast
* feat(mobile): keep debug signing until release signing is enabled
Release signing now turns on when the ORCA_ANDROID_RELEASE_CERT_SHA256
(or ORCA_ANDROID_REQUIRE_RELEASE_SIGNING) repository variable is set.
Until then the workflow keeps today's debug-signed build and adds a
warning and job summary note; once enabled, missing secrets fail.
* feat(mobile): enable v2 and v3 APK signing for the release key
* fix(antigravity): say Antigravity Chat isn't available on Windows instead of silently opening a terminal
Google publishes no Windows agy-acp-server, so a Windows host refused
Antigravity chats with the generic 'agent' reason: launches fell back to a
terminal with no explanation and Settings showed nothing.
A Windows host now reports reason 'windows' (createSupport and published
availability), which launch receipts state as chat_unavailable_on_windows,
the new-tab picker labels 'No chat on Windows yet', and explicit-chat
failures explain. Agents and Accounts settings say so for a Windows host.
The decision is the execution host's OS, so a Windows client chatting on a
macOS or Linux host is unaffected; one predicate
(structuredChatUnavailableOnPlatform) gates it all.
Fixes#30817, STA-11245
* docs: correct chatUnavailableOnThisDesktop comment
* fix(speech): stop offering Zipformer EN 20M and move its users to Parakeet v2
The model drops the start of every utterance. A replacedBy catalog field hides
it from the desktop and phone pickers, and a saved selection loads as Parakeet
v2. Downloaded files stay listed in Settings so they can be deleted.
* fix(speech): dim the retired model row in the Voice picker
* refactor(speech): share the offered-model check between main and renderer
* fix(speech): dim only the retired model's text so its delete button stays clear
The payload-delivery suite started its first Windows PowerShell 5.1 hook
launches at the same moment as the runner's cold first powershell.exe start,
with only a 15s per-launch bound. The stdin-lifecycle suite already absorbed
that cost in a warm-up; share it so both suites pay cold start outside the
per-hook hang bound.
* test: stop unit tests from running the developer's real agent CLIs
A live relay test reached the host's real `claude --version` through the
inherited PATH. On a machine whose claude wrapper never exits, each run left
an orphaned loop behind (14 were found on one host, running for days).
- Add a unit-test setup guard that shadows agent CLI names on PATH with
stand-ins that fail the test which ran them.
- Give the live relay test its own claude/codex stubs.
- Kill the whole process tree when the WSL guest's version probe times out.
* test: keep structured-session tests off the real codex CLI
These hosts get a scripted transport, but the at-rest `/` command read
spawns the resolved codex command directly, so on a machine with codex
installed each run started a real `codex app-server`. Point them at a
missing binary, as on CI.
* test: isolate agent discovery in session fixtures
Antigravity automations refused --extra-agent-args, so unattended agy jobs
could only use the model and effort from agy's global settings.json. Allow
--model and --effort through the existing allowlist; permission, mode, and
other flags stay refused.
Part of #35234 (STA-12678).
* fix(antigravity): wait for agy's composer before pasting continuation context
The renderer's paste gate armed its 1.5s quiet window on any bracketed-paste enable, including the launching shell's own prompt, so a slow agy start received the handoff in cooked mode before its TUI ran. Wait for agy's '? for shortcuts' composer row while agy holds bracketed paste instead.
* fix(antigravity): accept agy's composer rows when a custom statusLine hides the hint
A command statusLine replaces agy's '? for shortcuts' row, so Continue in New Session waited the full 20s and pasted blind, and a CLI launch prompt burned 20s of its 60s. Also accept the rule / '>' / rule composer rows (the layout #38930 reads), still anchored to agy's own 2004h and revoked by the shell's 2004l. Checked against agy 1.5.2 captures: custom statusLine and default layouts fire; the trust dialog and signed-out startup do not.
* fix(antigravity): offer back pre-ready typed text the composer wait drops; move scanner tests
The agy composer wait has no quiet fallback, so a slow start now hits the
20s timeout and drops held type-ahead. The drop notice now carries a Copy
action with the typed text, and still shows when the user switched
worktrees, so it is never lost silently.
Move the composer-rows scanner cases into draft-paste-ready-scanner.test.ts
since main removed antigravity-readiness-transcripts.test.ts (#38973).
* chore(plugins): regenerate declarative testing bundle for antigravity config change
Antigravity Accounts rejected a WSL account location outright. agy inside
WSL keeps its login in its file store (~/.gemini/antigravity-cli/
antigravity-oauth-token; it detects WSL the same way
isAntigravityFileStorageHost does), so save/select/remove now work there:
- A WSL credential backend reads and writes that file through runWslProcess
in the selected distro (default resolved via resolveWslDefaultDistro). It
checks the distro is running before every guest command and never starts a
stopped one. Writes are a compare-and-swap inside the guest (sha256 of the
expected file), go through a private temp file and rename, and pass the
credential on stdin rather than the command line.
- runWslProcess accepts stdin input for argv-delivered commands.
- Each distro gets its own encrypted vault, so accounts never cross distros
or the Windows host.
- Accounts "Refresh usage" is enabled for WSL now that usage reads the
selected runtime (#38985), and rejects a reading tagged with another
runtime's provenance.
Fixes#25847
Linear: STA-9624
* fix(automations): run lists send byte-bounded previews; Run Details reads one full row
* fix(automations): retry a failed full run read on the next history read
* fix(automations): show a full-output read failure with Retry in Run Details
* fix(automations): Run now returns the saved pending row; quiet, titled, retained run terminals
Run now answers with the durable pending row instead of waiting behind
the automation's lane; one detached handler records a launch failure.
Existing-workspace launches no longer surface their owner, a new
workspace's startup tab carries the run title, and finished-run cleanup
keeps a terminal a user opened after its process started.
* fix(automations): keep launched runs and user-opened run terminals intact
A failed flush after a launch no longer rewrites a dispatched run as a launch failure.
A user tab open marks the pane activation shows, read after any spawn it starts.
* fix(automations): own launch-failure recording in the service; quiet pending pane notice on mobile
* Add session Build and Plan choices to native chat
* Refresh planning choices across connected composers
* Give composer catalog tests a complete transport fixture
* Preserve planning choices through fresh turns and older hosts
* Type planning delivery fixture choices explicitly
* Honor mode capability for retained queued command picks
* Update existing native mode contract expectations
* Keep proposed plans visible in completed chat turns
* Group chat mode with permissions and goal controls
* Keep chat composer controls accessible in narrow panes
* Preserve composer overflow ownership and editor focus
* Complete composer localization catalog and subscription checks
* Keep chat composer controls contained in narrow panes
* Use the button spacing for compact Save
* Place chat mode immediately before Goal
* Restore Plan status test dependencies after main sync
* Show an exit pill only while chat Plan mode is active
* Remove obsolete composer tooltip translations
* Remove unused Build toggle catalog entries
* Apply Plan commands immediately from composer picks
* Preserve picked skill origin in composer admission
* Supply phone platform to composer regression fixture
* Preserve mode picks until their writer admits them
* Narrow the phone mode command before its draft transform
* Document the option fixture mutation shape
* Match phone Plan controls to the composer mode pills
* Assert optimistic mode feedback in the phone fixture
* Retain the phone mode pick recovery regression proof
* Use the relocated observation imports after the main merge
* Restore the composer IME module import after merging
* Read the queue assertion helper from its relocated owner
* Point the mode fixture at the relocated selection controller
* Register the phone Plan pill localization keys
* fix(mobile): a failed page remounts instead of re-downloading; deadlines count only foreground time
Page deadlines now pause in the background and restart on return, so React Native firing overdue timers on resume no longer reads as a page failure. A failed document remounts the same files once, then shows the failure screen; it never deletes and re-downloads on its own. Try again after a page failure or an unreadable generation deletes the host cache first, so it downloads fresh. Replaces #34550 without its file-size check.
* fix(mobile): drop reports a remounted page's retired view already queued
A failed page remounts at once, minting the replacement's session before
React commits it. A fault or ready the old view had already queued then
landed on the replacement: a second fault showed the failure screen
before the new page loaded, and a late ready marked it ready and
cancelled its load deadline. Each report now carries the session of the
view that rendered it and is dropped once that view is retired.
#38014 deleted tests/e2e/voice-microphone-selection.spec.ts but the feature-area
route still selects it, so every PR touching dictation or src/main/speech fails
'changed e2e specs' with 'No such file or directory'.
* fix(gitlab): paginate pipeline jobs and bridges past 100 items
The Checks panel fetched one 100-item page of a pipeline's jobs and
bridges, so larger pipelines (and expanded child pipelines) silently lost
the rest. Move the header-aware page walker used for labels and members
into its own module and use it for pipeline jobs and bridges too.
Fixes STA-13623
Co-authored-by: alexcastgonz <82875882+pcbeingused333@users.noreply.github.com>
* fix(gitlab): follow every advertised page and keep fetched pages on a later failure
Review repairs for #39046: follow advancing x-next-page headers without a
page cap (the cap now bounds only guessed pages for missing or unusable
headers), keep rows already fetched when a later page fails, and stop
bridge paging once the parent job list fails.
* fix(gitlab): fail the details read on an incomplete page walk
A later-page failure now fails the MR details read, so Checks and the
dialog keep the jobs they last showed and report the failure, instead of
replacing a complete list with a prefix. Forbidden child pipelines (first
page refused) stay tolerated. Paging also stops on empty pages and
rejects page numbers past the safe-integer range so it always ends.
* fix(gitlab): treat exhausted page guesses as incomplete and translate the dialog load error
* fix(gitlab): follow only the consecutive next page and name the dialog load-error key
A next-page header that skips ahead or carries trailing text is now
treated like a missing header, so no page is skipped. The dialog's
load-error message uses an intent-named translation key.
---------
Co-authored-by: alexcastgonz <82875882+pcbeingused333@users.noreply.github.com>
* Import canonical Codex execution history
* Fix Codex history import lint warnings
* Preserve Codex imported prompt content and item lifecycle
* Preserve Codex root order after history fact eviction
* Order imported Codex turns by recorded chronology
* Preserve retained Codex lifecycle and history row placement
* feat(native-chat): group turn diffs and opened PRs under Settings › Chat UI › Turn activity
Both Hide buttons now share one toast naming what Orca stops showing, with a
link to the new Turn activity section.
* fix(native-chat): say where to turn a card kind off when Hide can't save
Neil
Brennan Benson
Jinjing
Kelvin Amoaba
Jinwoo Hong
