Make validated CLI cache writes idempotent and reuse the startup-scoped node worker launch across route retries, keeping the signed menu-bar app near-zero CPU while idle.\n\nCloses #124592
The computer.act v1 wire contract is gone, but the naming that survived it
still described a version split instead of the real one: screen-coordinate
execution versus window/element-scoped execution. Both are live rungs of the
same ladder.
- Extract the screen-coordinate half of the 1334-line ComputerActionService
into ComputerScreenActionExecutor (dispatch, typing, scroll, coordinate
mapping, button-hold watchdog, raw CoreGraphics primitives). Moved code is
unchanged apart from threading the queue authority check as a parameter
instead of reaching back into the queue.
- ComputerActionService keeps its name and becomes the coordinator that owns
the execution queue, the permission probe, and the shared error vocabulary.
- Rename ComputerActionServiceV2 to ComputerWindowActionExecutor, isV2Request
to isWindowScopedRequest, isComputerActV2Only to isWindowScopedOnly, and
ComputerActionError.invalidV2Request to .invalidRequest. The emitted
COMPUTER_INVALID_REQUEST: prefix is unchanged.
- cua-computer: v2-actions.ts becomes window-actions.ts, handleV2Act becomes
handleWindowAct, and the stale v1Params local in handleDesktopAct becomes
desktopParams.
- Note at the computer.act idempotency key that its v1 prefix versions the key
composition, not the wire contract.
Behavior-neutral: no logic edits, no new branches, no changed error strings.
* fix(skills): keep ClawHub search results on the source the operator picked
ClawHub search returns each result's origin under `install.reference`, but the
response model expected a flat `installRef`. That field is never present, so
every row fell through to a synthesized `@owner/slug` reference. External
skills.sh results were rewritten onto a ClawHub-native identity, dropping both
the commit-pinned source and the "not scanned by ClawHub" trust record.
Map the search wire shape explicitly and make the search contract
action-specific: `installRef` always names the result's own source, `detailRef`
appears only while ClawHub can serve a detail card for that identity, and
`trustState` travels with unscanned sources. Clients render install directly
when detail is absent instead of offering a review the Gateway must refuse.
Covers the Control UI, macOS, iOS Settings, iOS AgentPro, and Android, which
previously routed every row through review and could not install an external
skill at all.
* fix(skills): make install-only sources explicit and keep legacy review intact
Address review findings on the search identity contract:
- Replace the detail-reference capability with an explicit `installOnly` flag.
A Gateway released before this field omits it, and reading omission as
install-only made ordinary registry results skip the reviewed-version flow on
every client. Absence now means the existing review-then-install path.
- Parse closed source variants in the producer. A row whose source is unknown,
whose external reference is missing, or whose registry publisher is absent is
dropped instead of falling through to `@owner/slug`, which was the original
source swap in a different disguise.
- Carry the exact install reference alongside the canonical slug. The Gateway
already records `requestedReference`; the clients dropped it and matched
installs by slug, so a completed external install read back as unknown.
- Gate the direct-install action on admin rights. The row previously stayed
enabled for read-only operators and reached a guard that silently returned.
- Route the unscanned-source warning through the native and Control UI string
catalogs instead of a hardcoded literal.
* chore(i18n): leave generated native locale artifacts to the refresh workflow
Preflight isolates generated locale output from source changes: only the native
sources and apps/.i18n/native-source.json belong in a feature commit.
* fix(skills): satisfy Android ktlint wrapping and Swift test link construction
Extract the ClawHub result action guard into a named value so the multiline
condition follows ktlint wrapping, and pass the new requestedReference field in
the OpenClawKit installed-link fixtures.
* fix(skills): preserve external install identity across clients
* test(skills): add exact refs to recommendation fixtures
---------
Co-authored-by: Patrick Erichsen <patrick.a.erichsen@gmail.com>
* fix(computer-use): converge provider ref lifecycles
* test(computer-use): expect stale lifecycle code
* fix(computer-use): key native window refs on stable window identity
Peekaboo matched an existing window ref on the whole WindowMutationIdentity,
which embeds capturedBounds and isMinimized. Any move, resize, or minimize
therefore minted a new ref for the same live window, never evicted the old
one, and let the old ref keep resolving to a stale ServiceWindowInfo whose
identity and bounds were then used as the per-action expectations, so the
operator saw a downstream mismatch instead of COMPUTER_STALE_OBSERVATION.
Window refs now key on stable identity only (WindowServer id plus the owner
process generation that guards pid reuse) and refresh the stored target in
place on every discovery, so one live window keeps one ref for the whole
lifecycle generation and later checks compare against current data. The
one-instantiation generic reference store is folded back into the service as
plain state plus small concrete helpers.
The shared case table gains window_moved, which both providers must satisfy,
and the Swift side now drives the real ComputerActionServiceV2 with real
Peekaboo values instead of a toy store with an equality matcher. Its
in-flight generation case runs a real perform() whose lifecycle is revoked
mid-action rather than throwing the error it asserted.
* test(cua-computer): drive the real in-flight generation-rotation path
* feat(macos): show computer provider readiness
* refactor(macos): drop the status string superseded by the readiness checklist
* chore(macos): refresh i18n baseline after removing the superseded status string
* fix(ui): fork active sessions from stable history
* fix(protocol): align active fork types
* test(ui): match main-session fork routing
* fix(sessions): unify stable fork admission
* fix(native): fork active sessions from stable history
* fix(android): fork active sessions from stable history
* style(android): simplify active fork result handling
* fix(native): preserve legacy fork transport API
* test(native): complete legacy fork transport stub
---------
Co-authored-by: Peter Steinberger <steipete@gmail.com>
* fix(macos): reap app-owned child process groups
macOS-owned SSH, Codex, and node-host descendants no longer survive terminal shutdown. Codex retains EOF-first graceful exit before bounded process-group termination and reaping.
* fix(macos): correct managed cleanup wake binding
* fix(macos): preserve Codex shutdown escalation
Keep the app-owned EOF grace window while allowing abortive requests to interrupt it before process-group TERM and KILL.
Quick Chat model changes now settle in target order, the remote probe uses one deadline, and node-owned Codex child shutdown is acknowledged and bounded.
Related: #123420
* feat(browser): sync system cookies to a remote gateway profile
Add `openclaw browser cookie-sync`: decrypt allowlisted macOS Chrome-family
cookies locally and push them into a managed profile on a possibly-remote
Gateway over the existing operator channel. --watch re-syncs on cookie-DB
changes with a single Keychain prompt per session.
- New POST /cookies/set-many batch route (mirrors /cookies/set)
- Extract one canonical readSystemProfileCookies reused by import + sync
- Mandatory domain allowlist (never syncs an unrestricted cookie jar)
- Decryption stays host-local (macOS); no cookie values are logged
* feat(macos): cookie sync checkbox and configuration UI
Add an off-by-default 'Cookie sync' section (Settings > General): a toggle, an
editable domain allowlist editor, and a target-profile field, actionable only in
remote-gateway mode. CookieSyncManager supervises `openclaw browser cookie-sync
--watch` against the connected Gateway when enabled, resolving a LOCAL CLI (never
the SSH-redirect path, since decryption is host-local) and injecting gateway
URL + token/password via environment, never argv. A status row surfaces
running/stopped/error and the last sync summary.
* fix(macos): satisfy cookie sync lint gates
* chore(i18n): register cookie sync native source strings
Regenerate apps/.i18n/native-source.json baseline for the new macOS Cookie
sync settings strings (additive only). Satisfies the native:i18n:verify gate;
generated locale artifacts are refreshed separately by the locale-refresh job.
Background-only Bridge and Mac-node launches keep GUI onboarding and saved Gateway-profile Keychain state cold while preserving environment/config-owned Gateway connectivity.
* refactor(i18n): re-key native i18n artifacts to content-hash identity (v2)
The native inventory stored a write-only 'line' field per entry, so any
unrelated edit above a string rewrote apps/.i18n/native-source.json
(~half of all commits touching it were pure line-number churn). Identity
was (surface, path, source), duplicating the same string per file
(5385 entries for 4187 unique pairs) and churning IDs on file moves.
Locale artifacts were positional arrays repeating full English source
text, so one inserted string rewrote diff spans in all 21 files.
v2 artifacts: inventory entries keyed by (surface, source) with merged
per-site {path, kind} lists and pure sha256 content-hash IDs; locale
files become id-keyed sorted translation maps. Existing translations
carry over by source match with a deterministic duplicate pick; the
sticky-ID reuse machinery and positional validation are deleted.
Everything under apps/.i18n plus generated platform locale artifacts is
marked linguist-generated. ci-changed-scope gains a one-time
owner-complete migration escape mirroring the control-ui precedent.
CLI surface (baseline/check/sync/verify) and the locale-refresh
workflow are unchanged.
* ci: register run-attempt-state test in its Vitest lane
Commit e04dfd26e2 added extensions/codex/src/app-server/run-attempt-state.test.ts
without a lane owner, so the full-suite ownership audit
(test/vitest-projects-config.test.ts) fails on main. Register it in the
attempt-light shard alongside its run-attempt siblings.
* fix(mac): stop unbundled processes crashing on notification-center access
UNUserNotificationCenter.current() aborts with NSInternalInconsistencyException
(bundleProxyForCurrentProcess is nil) in processes without a bundle identity,
e.g. swift-build dev binaries. Live crash: onboarding aborted at
PermissionManager.authorizationStatus during MacNodeModeCoordinator startup.
One availability predicate on PermissionManager gates all five call sites;
unbundled invocations degrade to not-granted/skip instead of crashing.
* chore(mac): refresh native i18n inventory for guard warning
* fix(skills): keep ClawHub publisher identity from search through install
ClawHub search returns one entry per publisher, so several results can share a
slug. Every client collapsed the selection to that bare slug before calling
skills.detail and skills.install, and ClawHub answered 409 AMBIGUOUS_SKILL_SLUG
with no in-product way forward.
searchClawHubSkills now records the publisher-qualified reference once, on the
result that carries it, and the Gateway protocol documents it. skills.detail
parses the same reference grammar skills.install already accepted, so review and
install cannot resolve to different publishers. Control UI carries that one
reference through row actions, detail, busy state, and acknowledgement retries,
and shows it so otherwise identical rows are distinguishable.
Fixes#117633
* fix(apps): send the ClawHub publisher reference from native skill browsers
macOS, iOS, and Android read the qualified reference from search results and use
it for skills.detail, install, busy state, installed matching, and list identity,
so two publishers sharing a slug stay distinct instead of collapsing into one
ambiguous request.
* fix(skills): refuse external-source skill detail instead of reading a same-slug skill
ClawHub has no source-qualified read endpoint, so a skills-sh reference parsed
down to its bare slug would have returned a registry skill's card while install
resolved the external artifact. Review and install could name different skills.
skills.detail now fails closed on any reference that carries a source, and the
macOS and AgentPro rows show the publisher reference next to the summary instead
of only when a summary is missing, so same-slug rows stay distinguishable.
* chore(apps): refresh native i18n source baseline for the skill row references
* refactor(skills): drop the unread search-result ownerHandle field
installRef is the one reference clients send back, and no client reads the
publisher handle separately, so the protocol and Control UI carry one field
instead of two.
* fix(skills): name the next step when external skill detail is refused
Clients that gate install behind a successful review would otherwise see only a
refusal, so the error names the direct install path and the CLI equivalent.
* fix(macos): use a doc comment on the ClawHub row subtitle
swift-format's docComments rule requires doc comments on declarations; the
subtitle property carried a regular comment and failed macos-swift.
* fix(skills): carry ClawHub trust state to clients that can install
Forwarding installRef let clients install the exact publisher the operator
picked, including external skills-sh sources. It did not forward the trust
state that says ClawHub never scanned that source, so iOS AgentPro — the one
surface that installs in a single tap with no review step — could install an
unscanned artifact with nothing on screen saying so. The CLI already labels
these (docs/clawhub/cli.md, docs/cli/skills.md); native clients could not,
because trustState was never on the wire.
trustState becomes an optional field on SkillsSearchResultSchema. It is purely
additive: older clients ignore an unknown key and the field is absent for
registry results, so downgraded readers are unaffected and no protocol version
moves.
Every client that renders a search row now shows "Not scanned by ClawHub",
matching the CLI wording exactly: iOS AgentPro in the row above the install
button, macOS and Android beside the review action, and Control UI on the row
that explains why review is refused for these sources.
Covered by a wire assertion that the state reaches clients for an external
source and stays absent for registry rows, plus decode-and-label tests on the
shared Swift kit and the Android parser, and a Control UI render assertion.
* fix(ui): size the ClawHub detail dialog to a refusal message
Refusing detail for an external source made an error-only dialog reachable.
The shared preview panel reserves a tall reader height for skill documents, so
a two-line refusal rendered in a mostly empty dialog and read as broken rather
than deliberate. Found by inspecting the review captures.
* revert(ui,apps): drop the ClawHub trust label layer
Maintainer product decision: skills.sh runs its own scanners, so OpenClaw does
not add a second alert layer in the apps. Removes the label from Control UI,
iOS, macOS and Android, and drops the trustState wire field that nothing would
render. The CLI keeps its existing label; changing that is a separate call.
Publisher identity, the fail-closed detail refusal, and the message-only dialog
are unchanged. Splits the oversized skills view test file to satisfy max-lines
without a suppression.
* test(ui): fix ClawHub skill fixture checks
* chore(plugin-sdk): refresh API baseline
---------
Co-authored-by: Patrick Erichsen <patrick.a.erichsen@gmail.com>
* feat(gateway): recover channels and health promptly after host sleep
A dependency-free thaw detector rides the existing 30s maintenance tick:
when the process resumes after being frozen >=45s beyond cadence (laptop
sleep, VM pause, SIGSTOP), the gateway restarts running channel accounts
(dead sockets otherwise take up to ~35 minutes to notice), refreshes
health/presence, and resets the event-loop histogram so the freeze does
not read as degradation. Admission is rechecked before every recovery
side effect; a suspension beginning mid-recovery re-pends the thaw, and
timed-out channel stops complete their two-call restart in one pass.
The macOS app cooperates: NSWorkspace sleep/wake observers in
GatewayConnectivityCoordinator best-effort prepare a local gateway
suspension before sleep and resume it on wake, never blocking sleep.
The lease is bound to the route that prepared it and always cleared on
wake; route or mode changes across sleep drop it to self-expiry.
Live proof: SIGSTOP 85s on an isolated dev gateway -> 'host thaw
detected: process was frozen ~57683ms', channels restarted, health ok,
eventLoop degraded=false after thaw.
* fix(macos): resume a sleep lease whose prepare response arrives after wake
A prepare completing after didWake previously discarded the lease id,
fencing the gateway until the two-minute expiry after micro-sleeps; the
late response now resumes immediately. Document the conservative
route-token drift tradeoff.
* fix(macos): retry wake resume after refreshing the dead post-sleep transport
After real sleep the WebSocket is usually dead exactly when resume runs;
refresh the endpoint first, then attempt resume up to three times with
bounded delays, clearing the lease only on success or exhaustion. A new
sleep cycle aborts in-flight retries.
* fix(gateway): bound plugin stopAccount so channel stops cannot wedge recovery
stopChannel awaited plugin stopAccount unbounded; a never-settling stop
hung the thaw restart (and health-monitor sweeps) and held the
single-flight recovery guard forever. Race it against the existing
5s stop timeout; the timed-out path flows into the established
recoveryStopTimedOut two-call restart contract. Regression wedges
pre-fix.
* refactor(gateway): move thaw channel restart off ChannelManager and fence mid-pass
restartRunningChannelAccounts is a standalone helper over the public
manager surface with a shouldContinue probe checked before every stop
and start, so a suspension committing while an account stop is awaited
leaves later accounts untouched. Regression covers the mid-pass close.
* fix(gateway): sanitize late writes from an abandoned stopAccount
An abandoned (timed-out) stopAccount can settle after its replacement
started; route its late setStatus writes through the existing
stale-task sanitizer so they cannot repaint or tear down the
replacement. Regression fails pre-fix.
* fix(macos): make the whole AI candidate row clickable and clear stale exhausted verdicts
Live-testing pick-during-testing on 2026-08-11 showed clicks on a candidate
row's blank stretch (between the title/subtitle texts or over the spacer)
silently doing nothing: plain-style buttons only hit-test opaque label
pixels. A user trying to pick Claude Code while Codex auto-tests can click
the visually highlighted row and get no outcome. .contentShape(Rectangle())
makes the full row hit-test.
Also, after auto-candidates exhaust, a user-picked retest left the stale
"None of the found options worked" card up while the new test visibly ran;
userSelect now clears exhaustedAutoCandidates when a fresh attempt begins.
* test(tooling): drop stateless poll tests from stateful-helper fixtures
Main went red when #121923 rewrote the outbound poll tests to be
order-independent and removed their stateful helper import/file; this
lane-config test hardcoded both as stateful-helper classification
fixtures. Folded into this PR per red-main landing policy.
* fix(mac): scope managed CLI install and detection to the active app profile
Under an active OPENCLAW_PROFILE the onboarding ready page detected the
host-global managed CLI (~/.openclaw/bin/openclaw) and its update/install
flow rewrote the operator's real managed copy. The managed install prefix
now follows the profile state directory (~/.openclaw-<name>), preferred
paths exclude other profiles' managed trees (including stale validated
executables and inherited shell PATH entries), and external CLIs stay
detectable read-only. Default-profile behavior is unchanged.
* chore(mac): refresh native i18n inventory for shifted source lines
Peter Steinberger