mirror of
https://github.com/stablyai/orca.git
synced 2026-10-21 08:03:33 +00:00
42cf64cdc9bbdb125e3768fce551e2ffe0413823
880
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
  NeilandOrca
|
42cf64cdc9 |
fix(linux): stop Orca terminals from launching the GNOME Orca screen reader via bare orca (#8347)
On Linux the CLI installs as orca-ide so it never shadows /usr/bin/orca (GNOME's screen reader), but agent-facing surfaces (orca-cli skill, dispatch preambles, CLI hints) all invoke bare `orca` — so on stock Ubuntu an agent inside an Orca terminal launched the screen reader, which started speaking (#7904). Fix: prepend a userData-scoped shim dir (bare `orca` -> bundled orca-ide launcher, or the stable AppImage) to the PATH of every packaged-Linux managed PTY, mirroring the existing dev-mode cli/bin prepend. The user's own shells — and their real screen reader command — stay untouched. Also flip the orca-cli skill probe to prefer orca-ide so agents outside Orca terminals never execute the screen reader. Fixes #7904 Co-authored-by: Orca <help@stably.ai> |
||
  BingZandJinjing
|
be258e23ec |
Add Grok orchestration group routing (#8058)
* docs: design Grok orchestration group * docs: plan Grok orchestration group implementation * fix: add Grok orchestration group * test(orchestration): accept Windows skill newlines * Fix @grok orchestration group matching and remove stale planning docs - Reuse the shared buildAgentNameRe matcher in groups.ts instead of a divergent local regex, so orchestration groups honor the same Windows launcher-suffix rule (grok.exe/.cmd/.bat/.ps1) as the rest of Orca's agent-title detection. - Add test coverage for real Grok OSC title shapes (spinner-collapsed, session titles) and Windows launcher-suffix titles. - Delete the now-completed design and implementation-plan docs for the Grok orchestration group work. --------- Co-authored-by: Jinjing <6427696+AmethystLiang@users.noreply.github.com> |
||
|
|
1eace14c57 |
fix(preflight): reject Windows paths from WSL agent lookup (#7994)
* fix(preflight): reject Windows paths from WSL lookup WSL agent discovery previously treated path.win32 absolute results as valid guest paths, so a Windows absolute path like C:\spoof could be counted as a found agent. Only POSIX absolute paths are valid inside WSL. * docs(preflight): explain WSL path boundary * docs(preflight): correct WSL path rejection rationale Co-authored-by: Orca <help@stably.ai> --------- Co-authored-by: Jinjing <6427696+AmethystLiang@users.noreply.github.com> Co-authored-by: Orca <help@stably.ai> |
||
|
Jinjing
|
2802f0419d |
Fix browser annotation sends to active Windows agents (#8309)
* fix: authorize guarded sends to active ConPTY agents * Fix stale-handle rejection using controller-less status checks and contr - Distinguish `terminal_handle_stale` (PTY rebind mid-check) from `terminal_exited` so callers get an accurate reason instead of a generic `terminal_not_writable`. - Move the PTY binding assertion to always run after the fresh foreground read, even when no controller is present, closing a gap where a rebind during a controller-less check went undetected. - Call `confirmForegroundProcess` through its owning controller instance so the method keeps its `this` receiver instead of being invoked unbound. - Update reliability gates and design doc to reflect the new test coverage and corrected error semantics. * rm design doc |
||
|
Neil
|
800967278e | feat: promote APFS shared paths (#8318) | ||
|
Neil
|
3804f75f6a | fix native chat cleanup registration leak (#7663) | ||
|
NeilandOrca
|
19b63c4cb4 |
Prune detected worktree scan generation markers (#7679)
* Prune detected worktree scan generations * Invalidate pending scans in test reset hook Co-authored-by: Orca <help@stably.ai> --------- Co-authored-by: Orca <help@stably.ai> |
||
 Jinwoo HongandOrca
|
25ecf2eea2 |
fix: reconcile SSH repo rows after host re-add (#8201)
Co-authored-by: Orca <help@stably.ai> |
||
  
|
6e3ebf534c |
fix(terminal): arbitrate active desktop viewport ownership (#8252)
* fix(terminal): let remote desktop viewers own the shared PTY width
A remote (relay/shared-control) desktop viewer resizes the host source PTY to
its own width, but was never registered as a width owner. So the host's own fit
cascade (window resize, split drag, tab reveal, "+"-new-tab re-render) freely
resized the viewed PTY back to the host-local width with no signal to the
viewer. The viewer kept its narrower grid while the host streamed wider
alt-screen frames -> cell-layout garble ("porridge") until a manual resize.
The mobile "presence lock" already solves this shape for phones by suppressing
the host's pty:resize while the phone drives. This does the same for remote
desktop viewers WITHOUT joining the mobile driver state machine (a viewer needs
only resize suppression, not input lock / phone-fit / driver banners), and it
routes every PTY geometry change through the existing enqueueLayout/applyLayout
serialization path rather than resizing the PTY ad hoc.
Design:
- Registry `remoteDesktopViewers: Map<ptyId, Map<subscriptionKey, viewport>>`,
keyed per SUBSCRIPTION so duplicate streams of one client cannot release each
other. isPtyResizeDrivenRemotely() = mobile driver OR any viewer present;
pty:resize (host fit cascade) bails on it. INPUT is never locked (shared
control: host and viewers can both type), unlike mobile.
- New internal layout target { kind: 'remote-desktop' } in applyLayout: clears
terminalFitOverrides like 'desktop', resizes only when dims changed, emits no
fit-override notifications, and does not call onExternalPtyResize (so a
viewer's width never pollutes desktop restore state). enqueueLayout remains
the sole serialized writer of PTY dimensions.
- Smallest client wins: applyRemoteDesktopLayout() sizes the PTY to the SMALLEST
attached viewer, so viewers with different screen sizes never overflow the
narrowest grid.
- No snapshot/replay race: a viewport that arrives while the initial scrollback
snapshot is being serialized is buffered and applied as ONE serialized resize
before serialization, so a mid-repaint alt screen is never baked into the
snapshot.
- Reclaim on detach: when the last viewer leaves, applyRemoteDesktopLayout
resizes the PTY back to the host's own width via enqueueLayout({kind:'desktop'}),
so the host reflows to full width without a manual window jiggle. The host
width is captured at the pty:resize suppression point (the host's own resize
attempts) — a source a remote viewport never pollutes.
- Mobile coexistence: a phone outranks a viewer; when the phone leaves, the
mobile-release paths call applyRemoteDesktopLayout so a surviving viewer keeps
the PTY, else the host reclaims.
Verified live (two local desktop instances): correct render on connect, no
garble on viewer-window resize (synchronous), and host reclaims full width on
disconnect. Unit tests cover the registry lifecycle, smallest-client-wins,
reclaim-via-layout, and mobile coexistence.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-Authored-By: Codex <noreply@openai.com>
* fix(terminal): address remote-desktop PTY width review findings
Resolves four CodeRabbit findings on the remote-desktop viewer width
ownership change, plus a round-trip race surfaced while fixing the leak:
- Leak: the one-shot terminal.updateViewport RPC has no disconnect hook,
so it must never create a width floor (nothing releases it, pinning the
host at a stale width). Add refreshRemoteDesktopViewer, which only
refreshes floors the client already owns via its stream subscription
(matched by clientId), mirroring the mobile updateMobileViewport
no-op-without-subscription invariant. The one-shot handler now refreshes
instead of registering.
- Round-trip replay: with the one-shot fallback now refresh-only, a resize
landing during the subscribe round-trip (connected, stream not yet
current) was dropped. The transport now replays the latest viewport over
the stream once it becomes current.
- Reclaim: keep the host reclaim target unless the reclaim resize actually
landed (result.ok), so a failed reclaim can retry against true host
geometry instead of a stale remote width.
- Snapshot drain: drain pendingRemoteDesktopViewport in the
sendRequestedSnapshot finally block; a viewer resize parked during a
SnapshotRequest buffering window was otherwise dropped until the next
resize.
- Key scope: scope the multiplex width-floor key by connectionId so two
connections reusing the same client-local streamId can't overwrite or
release each other's floor.
Tests: refresh-never-creates-floor, reclaim-target-retained-on-failed-resize,
SnapshotRequest parked-resize drain, connection-scoped key assertions, and a
round-trip replay race test (verified to fail without the flush).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(terminal): arbitrate active desktop viewport ownership
Co-authored-by: Orca <help@stably.ai>
* test(terminal): expect viewport claim capability
Co-authored-by: Orca <help@stably.ai>
* fix(terminal): preserve input across ownership teardown
Co-authored-by: Orca <help@stably.ai>
* test(terminal): keep subscribe buffer coverage within lint limit
Co-authored-by: Orca <help@stably.ai>
---------
Co-authored-by: shady <shady2k@gmail.com>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: Orca <help@stably.ai>
|
||
|
Neil
|
7708f8e7fe |
perf: avoid blocking base-ref resolution (#8182)
* perf: avoid blocking base-ref resolution * fix(git): bound async default-base probes * fix(git): bound default-base probes during creation * test(git): cover bounded Windows and WSL probes |
||
 OrcaWin
|
c16ea9bae1 |
fix(crash-reporting): make Windows capture and submission reliable (#8242)
* fix(crash-reporting): make Windows capture and submission reliable * fix(crash-reporting): harden recovery fallbacks * test(crash-reporting): cover returned copy failures * fix(crash-reporting): keep Windows recovery responsive * fix(crash-reporting): retry transient identity lookup |
||
|
Neil
|
6c2c7e15f6 | Fix worktree warning dedupe retention (#7694) | ||
|
Neil
|
acb35ee649 | Fix git submodule path cache retention (#7687) | ||
 Brennan BensonandOrcaWin
|
4572b0a446 |
perf(worktrees): remove redundant creation probes (#7161)
Co-authored-by: OrcaWin <alpha-eng@stably.ai> |
||
|
NeilandOrca
|
4f7fe78921 |
perf(mobile): stop polling live worktree names (#8051)
* perf(mobile): replace worktree name polling with events * fix(worktrees): push rename invalidation to remote clients worktrees:updateMeta deliberately skips the renderer notifier (PR #209), but paired mobile clients no longer poll for titles, so a manual rename would never reach them. Emit the remote-only worktreesChanged client event (with resolved-cache invalidation), gated on displayName so per-click isUnread writes stay event-free. Co-authored-by: Orca <help@stably.ai> * test(worktrees): add missing runtimeStub type member for typecheck Co-authored-by: Orca <help@stably.ai> * fix(worktrees): derive rename event repoId with the shared non-throwing parser getRepoIdFromWorktreeId matches the mobile client's event filter exactly and cannot throw after the meta write already persisted. Co-authored-by: Orca <help@stably.ai> --------- Co-authored-by: Orca <help@stably.ai> |
||
|
Jinjing
|
c5044042ea |
Ai vault view log live tail (#8205)
* Add read-only View Log tabs so AI Vault agent sessions open inside Orca'
- Adds a `readOnly` flag on OpenFile that hard-blocks edits, autosave, dirty
state, rename, and drafts, and persists/restores it safely across sessions
- Wires AI Vault's View Log/Open Log actions to open logs as permanent
read-only local tabs instead of shelling out to the OS, gated to
local, single-file, non-synthetic session paths
- Registers a dedicated `jsonl` Monaco language (JSON-style coloring without
whole-document JSON validation) and forces read-only tabs to render as raw
source, bypassing markdown/mermaid/csv/notebook viewers
* Add live-tail streaming for local AI Vault View Log tabs
- Extend the fs:readFile snapshot path with byte-stable file identity so a
read-only tab can resume appending exactly where the snapshot left off.
- Add a ranged local log tail reader plus IPC (read/start/stop watch) that
streams only newly appended bytes, detects truncation/rotation, and
cleans up watchers on tab close or renderer destruction.
- Add a renderer-side UTF-8/line-boundary decoder and useLocalLogTail hook
that appends completed lines into the existing Monaco model, falling
back to a full reload on reset/rotation, keeping snapshot behavior
unchanged when liveTail is not opted in.
- Thread the new `liveTail` flag through OpenFile/PersistedOpenFile,
workspace session persistence/restore, and the AI Vault "View Log" open
path so live tail survives restarts and stays read-only-safe.
* Remove stale planning brief for the live-tail feature
The AI-VAULT-VIEW-LOG-LIVE-TAIL.md pick-up brief is no longer needed now that the live-tail streaming work has landed (
|
||
|
b902f19b0d |
fix(terminal): send CSI-u Shift+Enter to Droid on Windows (#7620) (#7668)
* fix(terminal): send CSI-u Shift+Enter to kitty TUIs (droid) on Windows (#7620) On Windows, Shift+Enter was always sent as the Alt+Enter byte ESC+CR (added in #2418 for Codex, which reads win32-input-mode and ignores CSI-u). droid speaks the kitty keyboard protocol, parses CSI-u directly, and treats ESC+CR as a plain Enter — so Shift+Enter SUBMITTED the message instead of inserting a newline. droid works in other terminals (Windows Terminal, Warp) because those honor win32-input-mode / kitty; Orca (xterm.js) withholds kitty from local Windows ConPTY panes and emits neither. Make the Windows Shift+Enter byte pane-aware: latch whether a pane's program advertised the kitty keyboard protocol (query CSI ? u, push CSI > .. u, or set CSI = .. u) and send CSI-u (\x1b[13;2u) to those panes, keeping the Codex-compatible ESC+CR for win32-input-mode-only TUIs. Non-Windows is unchanged (always CSI-u). Verified end-to-end against the real droid and Codex CLIs through the actual production functions: droid now newlines, Codex still newlines. * fix(terminal): route Windows Shift+Enter safely for Droid --------- Co-authored-by: Neil <neil@stably.ai> Co-authored-by: Neil <4138956+nwparker@users.noreply.github.com> |
||
|
Neil
|
2a01b41638 | perf(main): make worktree path dedupe linear (#8177) | ||
|
Neil
|
435de94102 | perf(main): avoid hosted-review IPC fanout (#8171) | ||
|
Neil
|
533992bdda |
fix(git): cache unsupported capabilities per host (#8109)
* fix(git): cache unsupported capabilities per host Old Git worktree, ref-search, and merge-tree fallbacks retried unsupported flags on recurring operations, flooding subprocess traces. Centralize capability probing per native, WSL, and SSH execution host, coalesce concurrent probes, and retry periodically for in-place Git upgrades. * fix(git): recognize real old-Git merge-tree rejection * test(git): enforce real binary compatibility matrix * fix(ci): preserve Git compatibility test ownership * fix(git): retain supported capability state |
||
|
Jinwoo HongandOrca
|
977960ed4a |
Make Windows workspace deletion reliable and respect Git locks (#7963)
* Handle locked worktree force deletion Co-authored-by: Orca <help@stably.ai> * Harden locked worktree removal * Recover Windows worktree deletion after Git deregisters * Require force permission for lock overrides * Respect Git worktree locks during deletion * Require structural proof before Windows deletion recovery * Align deletion tests with fail-closed recovery * Restore deletion test filesystem spies * Scope remote deletion tests to runtime host --------- Co-authored-by: Orca <help@stably.ai> |
||
|
|
e84a8ddec9 |
Terminal performance initiative: pipeline fixes + term-speed-2 revival + PTY flow control (integration branch) (#7214)
* Skip legacy hidden skip grammar assertions
* Fix hidden TUI snapshot test setup
* Fix sleep wake history test contract
* Fix hidden delivery startup gate helper
* Fix hidden Latin skip branch predicate
* Fix hidden synchronized split-boundary replay
* Stabilize remote runtime mixed subscription test
* Keep hidden startup query parser active during window
* Stabilize raw emoji golden restore width
* Stabilize raw emoji golden fixture completion
* Keep terminals responsive under agent output load
* Add frozen-terminal repro harness and silent-drop regression tests
Investigation harness for the frozen-terminal reports (Discord
#performance, issue #2836): pane shows content, shell alive, daemon
output.log flat while typing.
- e2e: renderer crash -> auto-reload recovery and three restart/restore
shapes (live daemon, SIGSTOP-wedged daemon, daemon killed between
launches), each probing input at both drop layers. Post-crash phases
drive the renderer from the main process because a crashed target
severs Playwright's CDP session even though the app recovers.
- e2e helpers: layer-discriminating probes (direct pty.write vs
transport input, plus pty:listSessions ownership-rebuild revival).
- unit repro: vendored xterm 6.1.0-beta.287 WriteBuffer permanently
wedges when a sync throw escapes a write-completion callback or a
custom parser handler (xterm-write-buffer-stall.repro.test.ts).
- unit repros for both silent input-drop layers: main drops writes for
a live PTY once ptyOwnership loses the id (revived by listSessions),
and the renderer transport stays unbound after a failed connect.
- pty.test.ts: unregister every leaked SSH provider id in afterEach so
module-level provider state cannot leak across tests.
Co-authored-by: Orca <help@stably.ai>
* Harden xterm write pipeline against sync-throw wedge that freezes panes
A synchronous exception escaping xterm's WriteBuffer loop permanently
wedges that terminal: _innerWrite has no try/catch around the parse
action or the write-completion callback, the tail re-schedule never
runs, and write() only re-arms on an empty buffer. The pane stops
rendering and, if a replay was in flight, the replay guard latches and
pty-connection's onData silently eats every keystroke — matching the
field reports (Discord #performance, issue #2836: content visible,
shell alive, daemon output.log flat). Both vectors verified against
vendored xterm 6.1.0-beta.287 in xterm-write-buffer-stall.repro.test.ts.
Three layers of defense:
- Guard every write-completion callback Orca hands xterm at the two
choke points (writeForegroundTerminalChunk, writeBackgroundTerminalChunk),
with settle and onParsed guarded separately so a WebGL/renderer
failure during viewport settle cannot starve the replay-guard release.
- Guard all throwing-capable custom parser handlers (DA1, OSC 10/11,
CSI ?h/?l mode reports, OSC 52 clipboard, OSC 7 cwd), degrading a
throw to "not handled" — same escape class as
terminal-link-provider-guard.ts.
- Replay-guard watchdog: each engagement releases exactly once, from
xterm's completion or a 10s watchdog, so a lost completion (wedged
pipeline, disposed-terminal race) cannot latch the guard on a live
pane; replayIntoTerminalAsync resolves on either path so restore
chains cannot hang. Force-releases record a crash breadcrumb.
All guard trips record rate-capped crash breadcrumbs, so the next field
occurrence names the throwing stack instead of failing silently.
Co-authored-by: Orca <help@stably.ai>
* Cap unbounded terminal output buffers in main and the foreground queue
Field evidence (Discord #performance / #2836): renderer memory climbs to
~1.5 GB and terminals freeze; a force reload does not help until memory
recovers. Two unbounded buffers matched that shape:
- Main-process pendingData grew by string concatenation without bound
while the renderer could not receive (frozen, starved, mid-reload) —
main-heap bloat a renderer reload cannot clear. Now capped at 2 MB per
PTY: past the cap the buffered bytes are dropped and the entry stays
O(1) until the renderer ACKs again, then a droppedOutput sentinel is
delivered and the pane repaints from the authoritative main-owned
buffer snapshot (existing hidden-output restore path) instead of
continuing a stream with a silent gap.
- The renderer output scheduler capped only hidden-pane backlogs; the
foreground path could queue a visible pane's flood without bound when
the drain could not keep up. The 2 MB cap now applies to every
foreground enqueue branch too, with a foreground-specific skip notice.
Verified: new main-side cap test (starve → flood → sentinel → normal
flow resumes), renderer sentinel-to-snapshot-restore test, two
foreground scheduler cap tests; full pty/terminal-pane/pane-manager
suites (1981 tests) and typecheck pass.
Co-authored-by: Orca <help@stably.ai>
* Make replay-guard stall release probe-certified instead of time-based
The previous stall watchdog blindly released the input guard after 10s.
If a replay were genuinely still parsing on a starved machine, that
early release could leak xterm's auto-replies into the shell — and into
agent TUIs, where a leaked ESC reads as the user pressing Escape.
Replace the blind release with a probe: when a completion looks
overdue, enqueue an empty write behind the replay. xterm parses writes
in order, so every outcome is provably safe:
- probe parses after the replay completion ran: normal release already
happened; probe is a no-op.
- probe parses but the replay completion never ran: all replay bytes
have parsed, no further auto-replies can exist — the completion was
genuinely lost. Release + breadcrumb.
- probe never parses (bounded wait): the pipeline is wedged, and a dead
parser can never emit auto-replies, so releasing cannot leak input.
Release + breadcrumb naming the pane as needing recovery.
While the probe is pending — a slow-but-alive replay — the guard now
HOLDS instead of releasing early; that case is pinned by a regression
test.
Co-authored-by: Orca <help@stably.ai>
* Scale output backlog caps with the scrollback setting and breadcrumb drops
The 2 MB pending-output caps were flat, which risked dropping lines a
50k-row scrollback user would have retained. Both caps (main pendingData
and the renderer output queue) now derive from one shared policy:
max(2 MB, scrollbackRows x 120 chars) — 2 MB at the 5k default, 6 MB at
the 50k max. The main side reads the setting live via getSettings; the
renderer scheduler is configured where the terminal lifecycle already
reads the scrollback setting.
Every drop now records a rate-limited crash breadcrumb with dropped and
cap sizes (terminal_output_backlog_dropped in the renderer,
terminal_pending_output_dropped in main — no pty ids, session ids can
embed workspace paths). Field drop frequency and size decide whether the
cap constants need raising, replacing theory with data (#2836, #7017).
Backlog skip notices are now cap-agnostic since the limit varies.
Co-authored-by: Orca <help@stably.ai>
* Extract breadcrumb recording into a collection-safe leaf module
Playwright loads spec imports at collection time, and e2e specs import
terminal-module constants (e.g. terminal-attention.spec.ts pulls
POST_REPLAY_MODE_RESET from layout-serialization, whose chain reaches
replay-guard). The breadcrumb import added to the terminal modules made
that chain reach crash-diagnostics.ts, whose top-level import.meta.hot
and webview-registry import crash Playwright's transform
("ReferenceError: exports is not defined in ES module scope") — every
e2e shard failed at collection before running a single test.
Move recordRendererCrashBreadcrumb into crash-breadcrumb-recorder.ts
(type-only imports, no import.meta) and point the terminal modules and
their test mocks at it; crash-diagnostics re-exports for existing
callers. Full e2e suite collects again (262 tests / 94 files); unit
suites, typecheck, lint green. No runtime behavior change.
Co-authored-by: Orca <help@stably.ai>
* Add cross-terminal pipeline benchmark (DSR-fenced throughput + latency probe)
Run inside any terminal (Orca pane, iTerm2, Ghostty, Terminal.app, VS Code)
to measure its full byte path. DSR round-trip latency at idle and under a
paced agent-TUI load, plus fenced throughput over four deterministic
fixtures. The DSR fence forces 'all bytes parsed' before the clock stops so
xterm.js-class ingest queues can't flatter the result.
First piece of the terminal performance initiative's measurement rig.
Co-authored-by: Orca <help@stably.ai>
* Add terminal performance initiative plan
Working plan for the orca-performance branch: verified architecture
findings, workstreams (baselines, #7153 validation, term-speed-2 revival
with merge-scout numbers, stall fixes, flow control, rig extensions,
utilityProcess router, telemetry), benchmark protocol, sequencing, and
baseline-relative success criteria.
Co-authored-by: Orca <help@stably.ai>
* Add cross-terminal baseline results (Orca 1.4.91 prod vs Terminal.app vs Ghostty)
Headline: Orca DSR latency under 1MB/s agent-TUI load is p50 134ms / p99 292ms
vs 0.45ms (Terminal.app) and 0.21ms (Ghostty). Idle latency is fine (0.69ms
p50) — the problem is queueing under load, not the pipeline hop. agent-tui
fenced throughput: Orca 2.0 MB/s vs Terminal.app 37 MB/s, Ghostty 78 MB/s.
Co-authored-by: Orca <help@stably.ai>
* Add pipeline-loss decomposition benches (headless xterm + daemon ingest)
Both isolate layers of the 51x agent-tui gap found in baseline-jul02:
bare @xterm/headless parses agent-tui at 103 MB/s and daemon Session
ingest (emulator + pending-output recording + fanout) at 103 MB/s —
on the byte stream the full Orca pipeline delivers at 2.0 MB/s.
Parser and daemon are exonerated; the loss is in main per-chunk
processing, delivery/ACK pacing, or renderer layers above xterm.
Co-authored-by: Orca <help@stably.ai>
* Record baseline + decomposition findings in initiative plan
Co-authored-by: Orca <help@stably.ai>
* Add dev-build orca-performance bench result (confounded: dev mode, 282-col window, 3MB fixtures)
DSR under load p50 161ms — the #7139/#7150 branch does not move the
under-load latency class. Expected in hindsight: DSR replies are ordered
within the output stream, so the metric measures output-queue depth;
cooperative drain paces input responsiveness but cannot reorder the queue.
Shrinking the queue itself (producer flow control, task 6) and raising
agent-tui throughput (task 9) are the levers for this number.
Co-authored-by: Orca <help@stably.ai>
* Record dev-build #7153 check in findings log
Co-authored-by: Orca <help@stably.ai>
* Parse-clock high-priority terminal drains instead of fixed-nap dripping
Attribution (task #9): the drain loop wrote at most 2x16KB then slept
4/16ms regardless of parse speed — an isolation bench (new
pane-terminal-output-scheduler-throughput.bench.test.ts) measures that
drip at 1.9 MB/s background / 27 MB/s foreground against xterm's
~103 MB/s parse rate, matching the baseline-jul02 end-to-end numbers
(agent-tui 2.0 MB/s in prod 1.4.91).
Fix: high-priority (visible-pane) drains now re-arm on xterm's
parse-completion callback and carry 8 writes per tick; the isolation
ceiling rises 27 -> 117.6 MB/s (parse-limited). Background cadence is
deliberately unchanged (2 MB/s drip protects the focused pane; hidden
delivery is term-speed-2's job). DRAIN_TIME_BUDGET_MS still bounds
per-tick work, preserving #7139's cooperative-drain intent.
Validation: 621 scheduler/guard/pty tests green, typecheck clean.
Co-authored-by: Orca <help@stably.ai>
* Record task #9 attribution + parse-clock fix in findings log
Co-authored-by: Orca <help@stably.ai>
* Findings: 51x loss attributed to O(tail) retained-tail redraw path in main onPtyData
Co-authored-by: Orca <help@stably.ai>
* Window the retained-tail redraw path to the cursor's reach
Attribution (findings log 2026-07-03): main's onPtyData consumed ~93% of
the event loop under an agent-TUI flood, and the dominant term was
appendNormalizedToMultilineTailBuffer + finalizeRetainedTerminalRows
materializing ~2x tail-length row objects plus a per-row trailing-space
regex on every chunk — 0.888ms/chunk at the 2,000-line cap, on every
Claude-Code-shaped frame (cursor-up + erase-below).
The multiline algorithm now runs on a suffix window sized by the chunk's
maximum upward cursor excursion (plus the inherited redraw cursor and a
safety margin); the untouched prefix is shared by reference with a cheap
last-char trailing-space check to match the reference trim. Pathological
full-height cursor-ups fall back to the unwindowed implementation, which
is kept verbatim and exported as the reference for the 500-case
differential fuzz (retained-tail-redraw-window.equivalence.test.ts).
Micro-bench at a full 2,000-line tail: 0.888 -> 0.073 ms/chunk (12x).
1,415 runtime tests green, typecheck clean.
Co-authored-by: Orca <help@stably.ai>
* Add dev bench results: parse-clock and windowed-tail fixes
Co-authored-by: Orca <help@stably.ai>
* Record windowed-tail partial win + next-cycle recipe in findings log
Co-authored-by: Orca <help@stably.ai>
* Findings: remaining whale is the per-chunk blocked-reason check (~85% of onPtyData post-fix)
Co-authored-by: Orca <help@stably.ai>
* Throttle the terminal wait-blocked check off the PTY hot path
Post-windowed-tail attribution (findings log 2026-07-03): the blocked-
reason complex — two full-tail buildTerminalWaitText builds plus
toLowerCase and multi-pattern scans per chunk, existing only to stamp
waitBlockedAt — consumed ~85% of onPtyData's remaining cost (~700-790ms/s
under an agent-TUI flood).
The check now runs at a 50ms cadence over coalesced chunks (PTY chunk
boundaries are arbitrary, so coalescing preserves semantics), with a
trailing-edge timer so burst-final state is always evaluated, and an
immediate bypass when the incoming chunk (plus a 31-char split carry)
contains a prompt keyword — so actionable-prompt stamping stays
per-chunk-immediate while keyword-free flood frames skip the complex
entirely. Previous wait text is cached per pty instead of rebuilt, and
state is cleared at both pty teardown sites.
1,415 runtime tests green (including the cross-chunk prompt test, which
exercises the keyword bypass), typecheck and lint clean.
Co-authored-by: Orca <help@stably.ai>
* Findings + results: three stacked fixes unlock the pipeline (agent-tui 16x, DSR-load p50 161->18.8ms in dev)
Co-authored-by: Orca <help@stably.ai>
* Add producer flow-control design to initiative plan
Co-authored-by: Orca <help@stably.ai>
* Findings: revival branch green but perf-gated — daemon Session ingest regressed 103->40-48 MB/s (chain emulator restructure); merge blocked until blockedfix parity
Co-authored-by: Orca <help@stably.ai>
* Pre-filter daemon OSC/mouse scanners for introducer-free chunks
Skips the scan-tail copy and full-chunk walks when a chunk cannot contain
an OSC or private-mode sequence (single native includes() checks), with
split-sequence correctness preserved via explicit tail retention. Strictly
positive micro-optimization on the daemon per-chunk path; 641 daemon tests
green (1 pre-existing WSL failure unrelated).
Co-authored-by: Orca <help@stably.ai>
* Retract confounded daemon conviction; mandate load-controlled A/B protocol for the revival merge gate
Co-authored-by: Orca <help@stably.ai>
* Record A/B gate pass in findings log; add A/B result JSONs
Co-authored-by: Orca <help@stably.ai>
* Add producer-side PTY flow control (watermarks + protocol v19)
Main now pauses the actual PTY when a pane's renderer-pending backlog
crosses the 256KB high watermark and resumes once it drains below the
32KB low watermark (wide hysteresis band so a draining queue cannot flap
pause/resume per flush slice). node-pty pause() stops the pty fd read, so
the kernel/ConPTY buffer fills and a flooding shell blocks on write —
flood-induced buffered lag becomes shell blocking instead of unbounded
main-process buffering (terminal-performance-initiative §5).
Transport: new fire-and-forget pausePty/resumePty daemon notifications
(protocol v19; 18 added to PREVIOUS_DAEMON_PROTOCOL_VERSIONS), routed
DaemonServer -> TerminalHost -> Session -> subprocess pause()/resume().
LocalPtyProvider pauses node-pty directly. Router/degraded providers
forward; IPtyProvider gains optional pauseProducer/resumeProducer.
Safety invariants:
- Lost-resume failsafe: daemon Session auto-resumes 5s after a pause with
no matching resume; main re-asserts the pause at most once per 5s while
still above the high watermark, so a lost resume can never wedge a shell
and a sustained flood stays throttled.
- Resume on every teardown path: Session kill/exit/dispose/detach; main
releases on pty exit and on window-destroyed bookkeeping wipes; the
adapter owes paused sessions a resumePty on the next connect after a
socket drop.
- Providers without support (SSH relay, legacy protocol <= v18) no-op
silently, and the scrollback-scaled pending-output cap still bounds
main memory when pause is unavailable.
- Kill switch: PRODUCER_FLOW_CONTROL_ENABLED in ipc/pty.ts flips the
whole mechanism off in one line.
daemon-errors.ts is split out of types.ts to stay under the max-lines cap.
Tests: watermark transitions/hysteresis/re-assert (controller unit),
lost-resume failsafe + resume-on-kill/exit/dispose/detach (session),
notification routing + v18 gating + reconnect owed-resume (adapter),
direct pause/resume (local provider), and a flood test asserting pause
fires once, pending stays bounded at HIGH + one chunk, and resume fires
once after drain (ipc/pty).
Co-authored-by: Orca <help@stably.ai>
* Findings: flow control merged; definition-of-done accounting; prod verification re-scoped to packaged RC
Co-authored-by: Orca <help@stably.ai>
* Fix stray brace from revival merge in long-table-scroll-restore e2e spec (broke e2e transform in CI)
Co-authored-by: Orca <help@stably.ai>
* Prod verdict: v1.4.121-rc.0 bench — DSR-load p50 134->18.6ms (7.2x), agent-tui 2.0->11.2 MB/s, idle at Terminal.app parity; pipeline now cadence-bound
Co-authored-by: Orca <help@stably.ai>
* Recover terminal output delivery after system sleep
Root cause: main gates every pty:data send on a global + per-PTY
in-flight counter that only renderer ACKs decrement. If ACKs are lost
across a system suspend, the counters pin at the cap and every PTY —
old and newly created — is silently gated forever while output piles up
in pendingData. A focus-preserving display wake also fires no renderer
focus/visibilitychange events, so terminal wake recovery (and the WebGL
context-loss latch clear) never runs. Only a renderer reload recovered.
Three fixes:
- ACK-stall watchdog (src/main/ipc/pty.ts): if sends stay gate-blocked
for 10s with zero ACK progress while the renderer webContents is
alive, warn once, reset the in-flight delivery counters, and flush
held pendingData. Armed lazily on the first gate-blocked send and
disarmed by every ACK, so it can never fire under healthy heavy load.
- Renderer lifecycle reset now also zeroes the in-flight counters — a
reload/navigation destroys the renderer dispatcher, so outstanding
ACKs can never arrive and stale counters would gate the new renderer.
- System-resume wake IPC: main relays powerMonitor 'resume' as
system:resumed to live windows (plus forceRepaint); preload exposes
ui.onSystemResumed; the terminal wake-recovery hook runs the same
recovery path as window focus/visibilitychange.
Co-authored-by: Orca <help@stably.ai>
* VS Code head-to-head: Orca beats/ties 5 of 6 metrics (16x idle, 5x styles-stress, better p99); load p50 gap attributed to ACK window + timer-clamped drain cadence
Co-authored-by: Orca <help@stably.ai>
* Schedule zero-delay terminal drains via MessageChannel
Chromium clamps nested setTimeout(0) to ~4ms, stacking dead gaps onto
every parse-clocked drain tick; the explicit 4ms high-priority re-arm
interval added more. A posted message is still a macrotask — input and
paint are serviced between posts — so cooperative yielding survives
without the clamp. Generation-tokened cancellation; vitest keeps the
timer path (fake timers can't advance channel posts) plus a real-timer
smoke test for the channel path. Standing-queue target: VS Code's ~7ms
class (measured us 18.6ms, them 7.18ms, same rig).
Co-authored-by: Orca <help@stably.ai>
* Cut daemon and main PTY batch windows 8ms -> 2ms
At 9% pipeline utilization the DSR-under-load latency is fixed batching
windows, not queue depth (proved by the MessageChannel drain lever
moving nothing). Both hops charged an expected half-window per chunk;
2ms keeps burst coalescing at negligible IPC overhead (~500 msgs/s
worst case vs MB/s payloads).
Co-authored-by: Orca <help@stably.ai>
* Findings + tests: batch windows were the DSR-load gap (19->8.0ms dev); timing tests updated to 2ms windows
Co-authored-by: Orca <help@stably.ai>
* Fix PR CI and guard resume relay during shutdown
Co-authored-by: Orca <help@stably.ai>
* Chain e2e specs 6/6 green — gate x drain validation debt paid
Co-authored-by: Orca <help@stably.ai>
* Replace ack-stall watchdog with cumulative ACKs + solicited delivery resync
Design review: the 10s blind-reset watchdog decided correctness from a
wall-clock threshold. Rework piece 1 into a deterministic two-part design
(pieces 2 and 3 — lifecycle-reset counter zeroing and powerMonitor wake
IPC — are unchanged):
- Cumulative ACKs (TCP-style): the renderer dispatcher now tracks a
monotonic per-pty total of processed chars (terminal-pty-ack-gate) and
sends it on every ACK alongside the legacy per-chunk delta. Main keeps
per-pty sentChars/ackedChars and max-merges received totals — idempotent
and reorder-tolerant, so a lost ACK self-heals when any later ACK
arrives instead of becoming permanent in-flight debt. Provider
(SSH/daemon) backpressure is credited only the derived delta, clamped,
never negative. Main tolerates both payload shapes keyed by field
presence (dev hot-reload can mix renderer/main versions); totals reset
on pty exit and renderer lifecycle reset on both sides.
- Solicited resync (replaces the blind reset): when new pty data arrives
while that pty's delivery is fully gated and no probe is outstanding,
main sends pty:requestDeliveryResync; the renderer replies with its
cumulative totals and main reconciles via max-merge, then flushes held
pendingData. Event-triggered, verified-state recovery — no wall-clock
threshold decides correctness. The only timer is a 5s request/response
hygiene timeout that clears the outstanding flag and logs one
diagnostic warn per silent streak; it never mutates counters (a
renderer that cannot answer has dead IPC — reload is the only cure).
The 10s corrective watchdog is deleted.
Co-authored-by: Orca <help@stably.ai>
* Starting point: prior agent's garble differential fuzz harness
Three files recovered (were untracked) from a prior agent killed by API
outages, plus a trivial curly-brace lint fix in the op dispatcher so the
pre-commit hook passes:
- src/shared/agent-tui-ansi-fuzz-stream.ts (seeded agent-TUI byte-stream gen)
- src/shared/terminal-restore-parity-fixture.ts (renderer-parity fixture)
- src/main/daemon/headless-emulator-fidelity.fuzz.test.ts (suite 1: differential
HeadlessEmulator vs @xterm/headless reference on identical bytes)
Co-authored-by: Orca <help@stably.ai>
* Suite 1 findings: two new serialize round-trip bugs (B bold-loss, C cursor)
Scanned seeds 1..2000. Beyond the pre-documented serialize wrap-null-cell bug
(A, 27 seeds, tolerated), the fuzz surfaced two NEW real @xterm/addon-serialize
0.15.0-beta.287 round-trip defects, both of which garble a revealed hidden pane:
- Bug B (seeds 435, 770, 1321): serializing a dim cell followed by a bold-only
cell emits \x1b[1;22m; SGR 22 clears bold too, so restored bold is lost.
Minimal repro: '\x1b[2mA\x1b[22m\x1b[1mB' -> restored 'B' loses bold.
- Bug C (seeds 454, 1696): a final content row filled to the right margin leaves
xterm wrap-pending; the serializer's relative cursor restore lands one column
short. Minimal repro: '0123456789\x1b[3;5H' at cols=10 -> cursor x=3 not x=4.
Both isolated to pure serializer replay (no Orca preamble), confirming upstream.
Parity fixture verified faithful to the renderer pane's buffer options. Each is
pinned as a standalone it.skip repro; full evidence + classification in
notes/garble-fuzz-divergences.md. Seed 113 (handoff's DECSC/DECRC case) does not
diverge on the current harness. No production code changed.
Co-authored-by: Orca <help@stably.ai>
* Add perf prerelease update check modifier
Co-authored-by: Orca <help@stably.ai>
* Suite 2: hidden-reveal seq-reconciliation fuzz + two new snapshot bugs (D, E)
Property-tests the reveal seq-reconciliation byte-stitch (getChunkDataAfterSnapshot
/ reconcileChunkAgainstRestoredSnapshot in pty-connection.ts), mirrored exactly:
N=200 seeded hide/reveal scenarios with a rich agent-TUI hidden prefix snapshot
and an append-only racing tail, chunked with seq/rawLength meta, seq-domain
restarts, unmetered chunks and droppedOutput markers. Asserts snapshot-at-S +
reconciled tail == snapshot-of-everything (seq-neutral) and == always-visible
(end-to-end). Runtime ~5s at 200; FUZZ_ITERATIONS override documented.
Two NEW real snapshot-limitation garbles found while building it, both distinct
from suite 1's serialize bugs and pinned as standalone it.skip repros:
- Bug D: the DECSC saved-cursor register is not serialized. A hidden TUI that
saves the cursor (ESC 7 / CSI s) and restores it on reveal (ESC 8 / CSI u)
lands the restore at home. Repro: 'AB\x1b7\x1b[4;10HCD' + '\x1b8X' -> 'XB' vs 'ABX'.
- Bug E: a snapshot taken mid-escape-sequence (a PTY read split an escape) drops
the partial sequence (it's parser state, not buffer), so the tail's
continuation renders literal. Repro: 'AB\x1b[3' + 'mCD' -> 'ABmCD' vs 'ABCD'.
Fired on ~24% of the corpus (tolerated + counted via prefixEndsMidSequence).
The append-only-tail design isolates seq reconciliation from these and the Bug C
cursor cascade. Full evidence + fix directions in notes. No production changes.
Co-authored-by: Orca <help@stably.ai>
* Suite 3: 25-cycle park/reveal drift e2e test
Extends terminal-hidden-view-parking.spec.ts with a deterministic 25-cycle
park->reveal test on a static rich alt-screen TUI frame (box drawing, SGR
colors, wide CJK/emoji). Baselines against the frame after the first snapshot
restore (so both sides pass through identical machinery — the alt-screen restore
correctly drops normal-buffer scrollback, which is contract not garble), then
asserts every subsequent reveal reproduces it byte-for-byte with no accumulated
drift and no hidden-skip banner. Exercises the real renderer teardown +
HeadlessEmulator snapshot restore + PTY reattach path the fuzz suites model in
isolation. Passes in ~29s (electron-headless, workers=1).
Co-authored-by: Orca <help@stably.ai>
* Fix two serialize round-trip bugs garbling hidden-terminal snapshot restore
BUG B (addon patch): @xterm/addon-serialize's SGR diff emitted bold/dim set
params before the shared intensity reset 22, so "1;22" wiped a freshly set
bold and a bare "22" dropped a still-set bold/dim. Patched via pnpm
patchedDependencies (config/patches) to diff bold+dim as one intensity
group with the clearing 22 emitted first. Other flag pairs (4/24, 3/23,
7/27, ...) have dedicated resets and were verified unaffected.
BUG C (Orca-side hardening): the addon restores the cursor with relative
moves computed from where it assumes replay leaves the cursor; a final row
filled exactly to the right margin leaves replay wrap-pending and the
restore lands one column short. New shared
serializeWithAbsoluteCursor appends an absolute CUP from the source
terminal's authoritative cursor at every restore/replay serialize site
(daemon/runtime HeadlessEmulator.getSnapshot, renderer mobile snapshot
serializer, shutdown layout capture). It skips empty snapshots and
wrap-pending sources so it never changes already-correct behavior.
Round-trip repros + non-regression coverage in
src/main/daemon/terminal-snapshot-serialize-roundtrip.test.ts (verified
failing with the fixes stashed). buildRehydrateSequences extracted to its
own module to keep headless-emulator.ts under the max-lines budget.
Co-authored-by: Orca <help@stably.ai>
* Gates: tolerate+count Bugs B/C in deep mode; drop inverse from reconciliation tail
- Fidelity suite: add snapshotHasSelfCancellingBoldReset (Bug B) and
isMarginWrapPendingCursorOffByOne (Bug C) predicates so the corpus tolerates +
counts them like Bug A. FUZZ_ITERATIONS=2000 is now green (~113s) and fails
only on genuinely new divergences; each tolerance keeps its <50% degeneracy
guard. Default 300 unchanged (~17s).
- Reconciliation suite: drop SGR 7 (inverse) from the append-only tail. Inverse
marks trailing blanks with an inverse-fg the serializer round-trips slightly
differently by capture depth — a Bug-B-class serialize nuance, not seq
reconciliation. FUZZ_ITERATIONS=1000 is now green; default 200 unchanged.
- Notes updated: every bug class is both pinned (skipped repro) and tolerated in
its corpus; combined default runtime ~19s.
Regex uses String.fromCharCode(27) to stay oxlint no-control-regex clean.
Co-authored-by: Orca <help@stably.ai>
* Keep RC update checks off perf prereleases
Co-authored-by: Orca <help@stably.ai>
* Fix snapshot DECSC register loss (Bug D) and mid-escape boundary drop (Bug E)
Bug D: the serialized screen cannot carry the VT100 DECSC saved-cursor
register, so a hidden ESC 7 followed by a post-reveal ESC 8 restored to
home and clobbered live cells. The snapshot epilogue now re-saves at the
source's saved position before the final absolute CUP
(readSavedCursorRegister + serializeWithAbsoluteCursor; the active
buffer's own register, so alt screens carry theirs). Position-only by
design; never-saved terminals are left untouched.
Bug E: a PTY read ending mid-escape leaves the sequence in the emulator's
parser, so serialize dropped it and the racing tail's continuation bytes
rendered literally after reveal (~24% of the fuzz corpus). The emulator
now tracks the unparsed trailing partial at ingest
(terminal-partial-escape-tail.ts, committed post-parse like the mouse
mirror) and ships it as TerminalSnapshot.pendingEscapeTailAnsi.
applyMainBufferSnapshot writes it LAST, after POST_REPLAY reset — any
later ESC would abort the dangling sequence. Seq accounting is unchanged:
the tail is a suffix of bytes the snapshot seq already counts, so
reconcile slicing needs no adjustment.
Fuzz suites: unskip the Bug B/C repros (fixed on this branch) and the new
D/E repros; remove the B/C/E tolerance predicates so regressions fail
loudly. Only Bug A (upstream wrap null-cell) stays tolerated + counted.
Green at FUZZ_ITERATIONS=2000 (fidelity) and 1000 (reconciliation).
Co-authored-by: Orca <help@stably.ai>
* Count suffixed RC tags (rc.N.perf) in the shared rc counter — second suffixed cut collided with the first
Co-authored-by: Orca <help@stably.ai>
* Classify suffixed rc tags (rc.N.perf) as rc telemetry identity in release builds
The build-identity guard only knew vX.Y.Z and vX.Y.Z-rc.N, so suffixed
perf RCs cut fine but every platform build refused the tag and the
releases published empty.
Co-authored-by: Orca <help@stably.ai>
* Cut the hidden-restore flood feedback loop (A) + query carve-out on drops (B)
(A) Under a foreground flood, the hidden-output-restore loop re-fetched
snapshots endlessly: each synchronous applyMainBufferSnapshot starved ACK
processing, main pinned at the in-flight cap, dropped at the pending cap,
and every droppedOutput/modelRestoreNeeded marker re-armed another
restore until the flood ended (rc.7.perf DSR timeouts).
- Restore loop: a foreground live-chunk queue overflow now abandons the
restore immediately (the stream is outrunning snapshot fetch+replay),
with a 3-iteration hard cap + lifecycle warn as backstop.
- Re-arm gate: drop markers/sentinels and reconcile seq-gaps on a visible
pane during its own in-flight/just-abandoned restore no longer re-arm;
live bytes write through and ONE deferred repaint (2s after the last
backpressure signal) heals the gap. Hidden-pane gate semantics are
unchanged.
- Query salvage: discarding queued restore bytes (overflow/refetch) now
extracts DSR/CPR/DA/OSC-color queries and replays them to xterm so
replies still flow.
(B) Main-side: dropOversizedPendingPtyData carves reply-eliciting query
sequences out of the dropped buffer (and out of post-drop latched data,
bounded) and ships them on the droppedOutput sentinel, so DSR probes
survive bulk drops. Query scanning moved to
src/shared/terminal-reply-query-extraction.ts, shared verbatim with the
renderer's hidden-startup query extraction.
Co-authored-by: Orca <help@stably.ai>
* ACK terminal output at parse-drain, not dispatcher enqueue (C)
The renderer credited main's per-PTY in-flight window the moment a
pty:data chunk entered the dispatcher, so the 512KB window meant "bytes
received", never "bytes parsed". Under flood the renderer write queue
grew unbounded behind instant ACKs; main saw no backpressure, crossed
the pending cap, and bulk-dropped output (rc.7.perf DSR timeouts).
Crediting is now parse-deferred: each delivery carries a fire-once
credit (deliverPtyDataWithDeferredAck); the pane's first scheduler write
claims it (writeTerminalOutput.ackCredit) and the output scheduler fires
it when the bytes are consumed — after terminal.write in the
parse-clocked drain, or on ANY discard path (backlog cap replacement,
discardTerminalOutput, disposed-terminal drops, flush recovery).
Deliveries that never reach the scheduler (reconcile drops, restore
queueing, pre-mount eager buffer) settle at handler return, so the
invariant holds: every delivered chunk credits exactly once, parsed or
discarded. E2E ack-gate hold/release and delivery-resync semantics are
unchanged (all crediting still routes through ackPtyData).
Main-side equilibrium: with ACKs at parse cadence, in-flight becomes
true backpressure — pendingData stays near the 256KB producer-pause
watermark, far under the >=2MB drop cap, so bulk floods block the shell
(node-pty pause) instead of dropping.
Co-authored-by: Orca <help@stably.ai>
* Synthesize salvaged query replies directly instead of replaying into xterm
The 10MB dev bench proved the write-back salvage insufficient: a
pending-cap drop always triggers a snapshot restore, whose replay guard
swallows xterm auto-replies and whose discardTerminalOutput races away
still-queued query writes — the salvaged DSR died both ways and the
fence still timed out.
Salvage now answers directly on the input path (immune to both): CPR
(CSI 6n) from the live buffer via transport.sendInput, DA1 with the
renderer's canned response, OSC color probes via the existing direct
responder. Rare queries (DECRQM, DA2) keep the best-effort xterm
replay.
Co-authored-by: Orca <help@stably.ai>
* Untrack branch-added bench result JSONs (20 files); keep numbers in the findings log
Files stay on disk; main's 7 pre-existing results are untouched.
Co-authored-by: Orca <help@stably.ai>
* Branch guide: document merge-not-rebase sync strategy and conflict pattern
Co-authored-by: Orca <help@stably.ai>
* Merge origin/main (#7316 tab-strip click-vs-drag fix); adapt #7290 recovery-reload tests to this branch's dual did-finish-load listeners
The three tests grabbed the FIRST did-finish-load listener; on this branch
the renderer delivery-gate reset registers before the orphan sweep, so the
sweep tests exercised the wrong handler (one failing, two vacuously green).
They now fire all listeners like a real reload.
Co-authored-by: Orca <help@stably.ai>
* Fix branch CI lint: split pane-interaction functions out of artificial-opencode-terminal-load.spec (815>800 lines), modernize perf-html-report script
No max-lines disable per repo rules; extracted to
artificial-opencode-pane-interactions.ts. toReversed() and
import.meta.filename replace reverse()/fileURLToPath.
Co-authored-by: Orca <help@stably.ai>
* Fix Windows update-relaunch killing the live terminal daemon
On a Windows update relaunch the daemon can be wedged past every RPC
budget (final checkpoint flush + installer/AV disk pressure), so the 3s
health check AND the 5s session-list hello both time out while sessions
are still alive - and the launcher failed closed, killing the daemon and
every terminal session it owned.
- Adopt an unresponsive daemon whose pipe still accepts a raw
connection; a new rejected health state keeps replacing daemons that
answered and refused the handshake (never adoptable).
- Give Windows pid files a real startedAtMs (daemon self-reports it in
the ready IPC message) and verify it via CIM CreationDate piggybacked
on the existing command-line query, so the pid-recycling guard is no
longer inert on win32.
- Only delete legacy daemon pid/token files when the pid-file process is
provably dead; deleting a live daemon''s token made its sessions
permanently unadoptable after a protocol bump.
- Capture agent resume records every 60s in the renderer (skipping
unchanged records) so hard kills still leave a fresh resume record.
* Heal blank terminals when main→renderer push delivery dies (renderer-pull delivery watchdog)
Field evidence (v1.4.121-rc.0 debug snapshot, 2026-07-06): a wedged window
held 530,115 un-ACKed in-flight chars — one PTY pinned at the 512KiB per-PTY
high water plus a fresh terminal's 245-char prompt that was sent and never
consumed — while the user ran the snapshot over invoke from that same window.
Main→renderer push delivery (pty:data and every sibling channel) was dead;
renderer→main→renderer invoke was alive. Upstream precedent for
one-directional IPC death: electron#37067 (suspected Mojo pipe disconnect,
stalled as need-info). Every terminal goes blank, new terminals are born
blank, and only a renderer reload recovered.
The existing recovery layers cover the OTHER variants of this bug family and
structurally cannot reach this one:
- The xterm write-pipeline sync-throw guards, output-buffer caps, and
probe-certified replay-guard release (#7150 family) run only after bytes
arrive in the renderer — here they never do. (The pending cap did work as
designed in the field: ~2.1MB pendingDroppedChars, bounded main heap.)
- Cumulative ACKs self-heal lost ACK messages and the solicited delivery
resync reconciles verified totals (4647df86a; #7260 on main) — but the
resync probe, the powerMonitor wake relay, and the droppedOutput restore
markers all ride main→renderer push, the direction that is dead. The
probe's unanswered path deliberately only logs.
This adds the missing lane, renderer-initiated and ridden entirely over
invoke — the direction the field snapshot proved alive:
- terminal-delivery-watchdog.ts: 15s heartbeat, free while output flows.
Hot-path cost is one Map upsert per received chunk; a tick does no IPC
unless the terminal plane was silent for the whole interval and a PTY
still expects delivery. Two consecutive silent ticks with main reporting
ACK-starved in-flight confirm the wedge; heals are one-shot per 60s
cooldown so a persisting wedge cannot repaint-storm.
- pty:reportRendererDeliveryState (invoke): always max-merges the renderer's
cumulative processed totals (a free extra repair lane for the lost-ACK
variant); with heal:true — and only after main has itself seen ≥10s of ACK
silence — writes off bytes the renderer provably never received
(received ≤ acked < sent; a received-but-unparsed backpressure window is
never written off), drops that PTY's pendingData (snapshot covers
everything ≤ markerSeq, hidden-drop parity), credits provider flow
control, and returns restore markers in the reply.
- The renderer re-attaches all push listeners (cures a detached-listener
variant outright; a safe no-op against a dead channel) and routes the
pulled markers through the existing pty:modelRestoreNeeded machinery —
panes repaint from the main-owned buffer snapshot with zero push
delivery involved.
- Field discrimination built in: the heal warn logs
ipcRenderer.listenerCount('pty:data') (listener detached vs channel dead)
with the full delivery snapshot, so the next occurrence names the root
cause without asking the user to run anything in a console.
Repro harness: the exposeStore-gated __terminalDeliveryWatchdog hook
blackholes pty:data ahead of the dispatcher — the field failure in
miniature (no receive count, no ACK credit, no dispatch).
terminal-push-delivery-loss-recovery.spec.ts proves the wedged output
repaints while the blackhole is still engaged and live flow resumes after
release, with no reload. Unit suites pin the watchdog state machine
(zero IPC under flow, two-tick confirm, cooldown), the dispatcher reattach
seam, and the main-side write-off semantics.
Perf: nothing added to main's send/flush path; the renderer data path gains
one integer/Map update per chunk; idle cost is one ~100-byte invoke per 15s
only during total terminal silence. Terminal perf e2e suite (typing
latency, redraw freeze, output scheduler, hidden TUI restore, artificial
opencode load) passes on this change; no watchdog activity occurs under
ack-gate pressure scenarios because receive-progress gates the heartbeat.
Co-authored-by: Orca <help@stably.ai>
* Expose the hidden-yet-visible delivery-gate contradiction in the debug snapshot
The v1.4.124-rc.2.perf blank-terminal field snapshot showed a different
state than the v1.4.121 transport wedge: no delivery gating at all
(ackGatedFlushSkipCount 0, in-flight 38KB, far under every cap) but TWO
ptys hidden-delivery-gated with 78MB dropped as hidden. The aggregate
counters cannot say whether the pane the user was staring at was one of
the gated ones — the one number that separates "normal background
dropping" from "main is starving a visible pane because the reveal
unmark never fired".
Add hiddenDeliveryGatedVisiblePtyCount / hiddenDeliveryGatedActivePtyCount
(overlap of the gate's hidden set with the renderer's visible/active
reports — a contradiction that must be zero) to the delivery debug
snapshot, and a once-per-minute warn when hidden-gated bytes are dropped
for a pty the renderer reports visible or active, with the full snapshot
attached. Zero cost outside the debug read and the already-dropping path.
Co-authored-by: Orca <help@stably.ai>
* Unlatch the hidden-delivery gate when user input disproves a stuck document.visibilityState
macOS occlusion tracking can wedge document.visibilityState at 'hidden'
after display sleep and never fire another visibilitychange. The hidden-
delivery gate then keeps dropping renderer-bound bytes for panes the user
is looking at (field snapshot 2026-07-06, v1.4.124-rc.2.perf: 78MB dropped
across 2 pane-level-visible ptys with a fully healthy transport), and every
recovery path (window focus, system-resume relay, backlog recovery) re-ran
syncHiddenRendererPtyDelivery only to recompute the same stale predicate —
nothing could ever clear the gate. The user sees a frozen terminal; typing
echo is dropped in main; only a reload recovers.
Real user input while the document claims hidden is a physical
contradiction: keystrokes and clicks only reach a focused, on-screen
window. stale-document-visibility.ts latches that proof, runs each pane's
existing visibilitychange resync (gate unhide + hidden-output snapshot
restore), and hands authority back to the occlusion tracker on the next
genuine visibilitychange. No timers; the failure bias is safe — a wrong
latch can only restore pre-gate delivery cost, never drop bytes. Hot path
unchanged: the foreground predicate still returns on the same single
comparison while the document is visible.
tests/e2e/terminal-stuck-occlusion-recovery.spec.ts pins the wedge
(visibilityState pinned hidden -> output dropped, not painted; the
hiddenDeliveryGatedVisiblePtyCount field discriminator reads >0) and the
recovery (one Shift keypress repaints the missed output from the main-owned
snapshot, no reload, while visibilityState still reads hidden). Negative
control verified: the spec fails without this fix. Typing-latency perf
gate passes; terminal-pane unit suites 366/366.
Co-authored-by: Orca <help@stably.ai>
* Add a one-paste terminal freeze report: __orcaTerminalFreezeReport()
Every field report of the frozen-terminal family so far has needed
follow-up asks (console output, main logs, second snapshots) because each
capture showed one process's counters at one instant. This makes a single
DevTools command sufficient: `await window.__orcaTerminalFreezeReport()`
returns renderer state (document.visibilityState + the stale-visibility
override, pty:data listener count, delivery-watchdog totals), main's debug
snapshot extended with a per-pty delivery table (sent/acked/pending, hidden
vs visible-set membership, last send/ACK ages, window focus flags, power
suspend/resume ages, app version), and bounded breadcrumb rings from BOTH
processes recording the transitions that matter: gate marks/unmarks,
visibilitychange and stale-visibility latches, watchdog stalls and heals,
restore markers, heal write-offs, and renderer lifecycle resets (so "user
already reloaded" is visible in the history).
Costs stay off the data path: breadcrumbs record only rare transitions into
a 100-entry ring with same-kind coalescing (a flood costs one slot per
second); the per-pty table is built only when the snapshot is read; the
per-send bookkeeping adds one Date.now() to existing accounting writes. Pty
ids are redacted to their `@@` suffix because daemon session ids embed
worktree paths. The report assembles over invoke IPC — the direction proven
alive in every observed wedge — and a failing invoke is captured as data
instead of sinking the report.
The stuck-occlusion e2e now also pins the report end-to-end: after the
wedge + keystroke recovery, the report must carry the stale-visibility
latch and gate transitions in the renderer ring, gate-mark/unmark in main's
ring, and a populated per-pty table. Suites: pty.test.ts 258, terminal-pane
1705, shared ring 5; typing-latency perf gate passes.
Co-authored-by: Orca <help@stably.ai>
* perf(daemon): keep-tail thin hidden panes' stream so agent floods never bury typing (STA multi-workspace lag)
Hidden panes are exempt from pendingData flow control (main gate-drops
their bytes after ingestion), so N background agents ran unbounded ahead
on the one shared daemon->main stream socket (measured 192MB user-space
backlog) and visible-pane echo waited FIFO behind it — typing appeared
seconds late whenever several agents burst on a loaded machine
(8x512KB/s + 12 CPU spinners: p50 293ms fix-off; 12x1MB/s: 6.1s).
Mechanism (replaces producer pacing — no reveal catch-up, ever):
- Shallow socket write gate (128KB) + per-session fairness bypass bounds
echo latency by construction; kernel-flush refill sentinel keeps held
bulk draining at full speed (drain-only refill capped at ~8MB/s).
- Backgrounded sessions' queued output is keep-tail dropped (newest
512KB kept, in-order dataGap replaces the middle); a ~2MB GLOBAL
budget shrinks per-session keep-tails (floor 64KB) so a worktree
switch never waits behind the aggregate. Reply-eliciting query bytes
(DSR/DA/OSC probes) are salvaged from dropped spans.
- Notifications are structurally lossless: the daemon runs the same
shared scanners main uses (bell/OSC 133/pr-link/2031) over every byte
BEFORE drop decisions and relays facts in byte order; ordered
background markers hand scan authority back and forth, seeded with the
emulator's partial escape tail so a sequence split across the handoff
neither phantom-fires nor goes missing. Titles/agent-status stay
main-side (kept-tail convergent).
- Main: background = hidden AND no remote view subscriber (a live
mobile/web view is never thinned); on dataGap main resets cross-chunk
parse carries, drops the headless mobile mirror, and reuses the
hidden-drop model-restore marker.
Wire: three new stream events, tolerated within protocol v19 (old mains
ignore unknown events; old daemons never see the trigger). Kill
switches: ORCA_DAEMON_BACKGROUND_STREAM_DROP=0,
ORCA_DAEMON_SHALLOW_SOCKET_GATE=0.
A/B (pnpm bench:multi-workspace-typing): 8x512KB/s + 12 CPU workers
p50 293ms/p90 647ms -> 15/21ms (= baseline); 12x1MB/s 6,146ms -> 20ms;
light loads unchanged; zero missing echoes. Latin hidden-restore e2e
green (probe-verified aggregate-drain root cause). New deterministic
repro harness: tests/e2e/terminal-multi-workspace-typing-latency.spec.ts
+ CPU pressure workers.
Co-authored-by: Orca <help@stably.ai>
* diag(terminal): breadcrumb WebGL context-loss/atlas + wake triggers into freeze report
Silent instrumentation (memory ring only, no new console lines) so the next
post-wake garble report attributes itself. Adds:
- shared/terminal-webgl-diagnostics.ts: lib-safe sink so pane-webgl-renderer
(lib) can record without importing the components-layer ring; wired to the
ring in terminal-freeze-breadcrumbs.
- webgl-context-loss crumb at onContextLoss, webgl-atlas-reset crumb at the
atlas registry reset — the pair that distinguishes 'atlas corrupted' from
'missed repaint'.
- wake-recovery:<source> crumb (focus/visibilitychange/system-resumed) with the
clearGlyphAtlases decision; source in the kind so distinct triggers don't
coalesce.
- per-pane WebGL state (getAllPaneRenderingDiagnostics) in the freeze report.
Gates: typecheck 0 errors; terminal suites 328 files pass; oxlint clean.
Co-authored-by: Orca <help@stably.ai>
* fix(lint): use Number.parseInt/parseFloat in terminal-view-attributes
oxlint unicorn(prefer-number-properties) flagged 24 global parseInt/parseFloat
calls in the terminal-view-attributes feature (
|
||
 Brennan BensonandBrennan Benson
|
e1de59aaa8 |
fix(quick-open): keep primary results when the git ignored pass fails (#8049)
The no-ripgrep git fallback ran both ls-files passes under Promise.all with killSurvivors on first rejection, so any ignored-pass failure (timeout, kill, non-zero exit) discarded the successful primary listing and Quick Open showed an error with zero files — the all-or-nothing failure called out in #7719. Directory collapse (#7842) bounds the enumeration so timeouts are far less likely, but a single-pass failure still nuked every result. Now, in the local main process and the SSH relay alike, only a primary-pass failure is fatal: an ignored-pass failure logs a warning and the listing resolves with the primary results plus any ignored entries streamed before the failure. Cancellation semantics are unchanged — an aborted scan still rejects via the primary pass or the expansion's cancellation check. Fixes #7719 Co-authored-by: Brennan Benson <brennanbenson@Brennans-MacBook-Pro.local> |
||
 
|
8976bd969c |
Fix SSH worktree delete routing after host reconnect (#7997)
* Fix SSH worktree delete routing * Harden host-scoped worktree deletion Co-authored-by: Orca <help@stably.ai> --------- Co-authored-by: Jinwoo-H <jinwoo0825@gmail.com> Co-authored-by: Orca <help@stably.ai> |
||
|
OrcaWin
|
b93a7eece1 | fix(watcher): preserve isolation in electron-vite builds (#8162) | ||
|
Jinwoo HongandOrca
|
dc468f0ded |
Fix Linux Parcel watcher crawl restart loop (#8155)
Co-authored-by: Orca <help@stably.ai> |
||
|
|
8f6e44ed53 |
Show agent session history on mobile (#6786)
* Show agent session history on mobile Bring the desktop "Agent Session History" panel to Orca Mobile as a per-worktree screen: browse past agent transcript sessions across the host with scope tabs (Workspace/Project/All), search, grouping, session cards, and tap-to-read message previews. The transcript scan previously ran only over Electron IPC, so mobile could not reach it. Expose it over the runtime RPC protocol mobile already speaks (aiVault.listSessions) so the scan runs on whichever host owns the transcripts — correct for local and SSH/remote hosts. Both the desktop IPC handler and the new RPC method share one cache, so opening the desktop panel and the mobile screen never double-scan. The pure filter/group/display logic is lifted into /shared (the renderer re-exports it) so the standalone mobile package can reuse it. Mobile narrows scoped tabs client-side by cwd path-prefix because the host scan treats scope paths as a widening union. Resume-from-mobile is intentionally a follow-up. * Fix mobile agent history list rendering and RPC authorization - Authorize aiVault.listSessions in the mobile RPC allowlist so the mobile client's call is not rejected before dispatch (without this the screen could never load sessions at runtime). - Name each SectionList section's rows `data` (the field React Native reads) instead of `cards`, fixing a type error and silent empty-section rendering. * Address review feedback on agent session history - Match quoted repo:/path: search operator values so labels and paths with spaces match (e.g. path:"/Users/ada/My Project"). - Hold a scoped tab in loading until the worktree list resolves instead of firing an unscoped fetch that briefly shows unrelated host history; proceed once loaded even if the worktree is absent (no stuck spinner). - Clear cached host capabilities on disconnect/host-switch and failed status.get so a capability-gated action can't linger for a host that doesn't support it. - Cover the real OrcaRuntimeService codex-home forwarding path and the quoted-operator parser with tests. * Hide redundant mobile current worktree badges Co-authored-by: Orca <help@stably.ai> * Resume agent sessions from mobile history (#6969) Co-authored-by: Orca <help@stably.ai> * Adapt merged seams to main's lint and reply-sender hardening Co-authored-by: Orca <help@stably.ai> * Cap mobile project-scope paths to the aiVault RPC bound Co-authored-by: Orca <help@stably.ai> * Share the aiVault scopePaths bound between the RPC schema and mobile Co-authored-by: Orca <help@stably.ai> * Guard shared AI Vault inflight cleanup against concurrent key replacement The extracted cache module's .finally() cleared inflight tracking unconditionally, dropping the if (inflightKey === key) guard its sibling outer cache kept: an older scan resolving after a different-key scan replaced the tracking would null the newer scan's dedup slot, so a re-request started a duplicate transcript rescan. Mirrors the sibling guard; the regression test flushes a macrotask so a reverted guard fails fast on the call count instead of hanging. Co-authored-by: Orca <help@stably.ai> * Harden aiVault.listSessions contract and gate mobile header entry on capability - Clamp scopePaths (64) instead of rejecting, cap limit at 2000, and make executionHostId optional so mobile can omit it; restamp per caller. - Retain successful mobile terminal-create mutation ids for 60s so resume retries dedupe after transient socket drops. - Gate the session-header Agent History action on the aiVault.v1 capability (mirrors the host-list action) so old hosts never show a dead-end entry. - Fix stale contract comments (scopePaths clamp semantics; filters move includes quoted repo:/path: operator parsing). * Add subagent field to session test fixtures after #7423 merge AiVaultSession.subagent became required on main; the five fixtures added on this branch predate it. Top-level scanned sessions carry null. --------- Co-authored-by: Orca <help@stably.ai> Co-authored-by: Brennan Benson <brennanbenson@Brennans-MacBook-Pro.local> |
||
  
|
143c36e566 |
feat(jira): group Jira issues by status (#7958)
* fix(jira): group Jira issues by status in task page Generated with [Devin](https://devin.ai) Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com> * feat(jira): order grouped statuses by agile board columns configuration * test(jira): add comprehensive tests for Jira issue grouping functionality Add test coverage for the new Jira issue grouping features including: - Grouping issues by status name - Sorting sections by agile board column configuration - Falling back to alphabetical sorting when no board config exists - Collapsed groups state management - Filtering issues based on collapsed state - Backend tests for getProjectStatuses API with various scenarios Generated with [Devin](https://devin.ai) Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(jira): add backend tests for getProjectStatuses and RPC routing Add test coverage for the new getProjectStatuses backend functionality: - Returns empty array when no clients are available - Returns statuses from project statuses API when no board configuration exists - Orders statuses by agile board column configuration when available - Handles missing status IDs gracefully by falling back to unordered list - Clears token on auth errors - Returns empty array on operational errors - RPC routing for jira.getProjectStatuses method Generated with [Devin](https://devin.ai) Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(jira): harden status grouping --------- Co-authored-by: Andres Van Reepingen <andres.vanreepingen@datacamp.com> Co-authored-by: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com> Co-authored-by: Jinjing <6427696+AmethystLiang@users.noreply.github.com> |
||
|
Brennan Benson
|
2f62292944 |
fix(terminal): reset PTY delivery accounting across renderer lifecycle resets (permanent frozen panes) (#8034)
* fix(terminal): reset PTY delivery accounting across renderer lifecycle resets (frozen panes) Panes froze permanently because main's PTY->renderer flow control leaked its unacked in-flight byte counters and pending backlog across renderer reloads/crashes: the lifecycle reset cleared only visibility hints, daemon PTYs outlive the page, and the new page never acks the dead page's bytes, so any pty with >=512KB unacked at reload time was delivery-gated forever (proven by dev repro: counter pinned at exactly 524288; one manual ackData of the leaked bytes instantly unfroze the pane). Fix: (1) zero in-flight counters + pending backlog on every main-frame renderer lifecycle reset (panes rebuild losslessly from main's authoritative snapshot on reattach); (2) hold PTY sends until the new page's pty:data listener signals ready (pty:rendererDispatcherReady), so boot-window output accrues in the capped backlog instead of being counted-but-dropped; (3) 10s self-heal watchdog + reconcile-on-handshake backstop so a missed reset (subframe-overlapped reload emits no did-start-loading) or lost handshake can never itself freeze delivery; (4) derive the active-renderer-pty report reactively so in-place rebinds keep the interactive reserve; (5) expose reset/gate state in the delivery debug snapshot for future diagnosis. Adds an experimental reliability gate (pty-delivery.renderer-lifecycle-accounting-reset) with red/green-proven regression tests for each branch of the fix. * chore(reliability-gates): link PR #8034 as the gate's motivating fix * fix(terminal): sender-guard the dispatcher-ready handshake; fix WSL gate wording CodeRabbit: the reconcile backstop destructively clears delivery accounting, so a straggler pty:rendererDispatcherReady from a dying window's webContents must not reset the new window — reuse the pty:write main-window sender check (+ regression test). Also resolve the contradictory WSL coverage claim in the reliability gate notes (WSL rides the same local/daemon pipeline; mobile/relay and SSH are separate paths). |
||
|
|
c5669cceb5 |
fix(quick-open): prune generated dirs in git fallback (#7842)
* fix(quick-open): prune generated dirs in git fallback * fix(source-control): narrow manual review provider switch * fix(quick-open): collapse git fallback directories Co-authored-by: Orca <help@stably.ai> * fix(quick-open): harden collapsed git directory expansion Co-authored-by: Orca <help@stably.ai> * fix(quick-open): close directory expansion races Co-authored-by: Orca <help@stably.ai> * perf(quick-open): collapse placeholders efficiently Co-authored-by: Orca <help@stably.ai> * fix(quick-open): preserve root and cancellation semantics Co-authored-by: Orca <help@stably.ai> --------- Co-authored-by: Jinwoo-H <jinwoo0825@gmail.com> Co-authored-by: Orca <help@stably.ai> |
||
|
Neil
|
26224196f5 |
perf(worktrees): avoid auto-maintenance in create fetches (#8039)
* perf(worktrees): avoid auto-maintenance in create fetches Git's opportunistic maintenance can keep an already-complete exact-base fetch open for seconds. Disable it per command for create-base refreshes only, leaving ordinary fetch maintenance and exact-ref freshness unchanged. * docs(worktrees): explain create fetch maintenance scope * test(worktrees): account for fetch config prefixes * fix(worktrees): cover Git 2.29 auto maintenance |
||
|
Neil
|
e4c7aab4a2 | Normalize invalid terminal line height before xterm (#8050) | ||
   |
67447fe64f |
Add AI Vault subagent display (#7423)
Co-authored-by: Orca <help@stably.ai> Co-authored-by: hmrserver <> Co-authored-by: brennanb2025 <brennankbenson@gmail.com> |
||
|
|
f238952be2 |
Agent status over WSL: guest-resident hook relay + WSL-side hook installers (STA-1515) (#7903)
* docs: full design + context for agent status over WSL (STA-1515) Why hooks don't work on Windows+WSL (loopback transport gap + WSL-side installation gap), per-client transport map, the OMP-only fixes that shipped (7642/7641) and why they don't generalize, the recommended guest-resident relay over wsl.exe stdio mirroring the SSH relay plus WSL-side hook installers, alternatives considered, validation facts and gotchas from the 2026-07-08 Windows rig run, and acceptance criteria. Co-authored-by: Orca <help@stably.ai> * feat(agent-hooks): agent status over WSL — guest relay + WSL-side hook installers (STA-1515) Agent hooks have never worked from inside WSL: under default NAT networking, WSL's 127.0.0.1 is its own loopback, so every hook POST to the Windows listener dies silently, and hook configs were only ever written to the Windows home where WSL agents never see them. Transport: a hooks-only guest relay (src/relay/wsl-agent-hook-relay.ts) runs inside the distro, binds WSL loopback on the very port the clients were already given (host-issued token; EADDRINUSE falls back to :0 with endpoint-file re-coordination, which also covers mirrored networking), and forwards parsed envelopes over its own wsl.exe stdio into agentHookServer.ingestRemote — the same shape as the SSH relay. It exits when stdin closes so a freed Windows port can never be forwarded into a dead guest listener. Installation: the unchanged SSH remote hook installers run against an SFTP-shaped adapter whose primitives are home-scoped fs RPCs served by the relay, so all 14 managed agents' hooks land in the WSL home over the already-open channel with zero per-file wsl.exe spawns. Lifecycle: per-distro manager ensured from buildPtyHostEnv on every WSL PTY spawn (covers post-restart daemon reattach re-spawns), stale-bundle reinstall via exit 42, no-node-43 cooldown, bounded retry for wsl.exe 'Catastrophic failure (E_UNEXPECTED)', breadcrumbed failures. Zero per-client transport changes; listener stays Windows-loopback-only. Co-authored-by: Orca <help@stably.ai> * fix(agent-hooks): WSL relay link-death recovery + Codex runtime-home hook install (STA-1515) Follow-ups from the first Windows-rig validation of PR #7903: Link death: a mux protocol error or keepalive timeout could kill the host<->guest link while the guest relay stayed alive returning 204s — the manager stayed 'running' and every later envelope blackholed silently (the exact observed signature: Claude hooks POST 204, store never populates). wsl-hook-relay-link.ts now guarantees exactly-once death handling from either signal (mux dispose OR child exit); the manager breadcrumbs it, kills the child, and self-restarts after a short cooldown since a live agent session produces no new PTY spawns to re-trigger ensure. ORCA_WSL_HOOK_RELAY_DEBUG=1 traces each received envelope pre-ingest. A live integration test pins the full host chain: the real esbuild bundle over real child stdio through the real manager into a real AgentHookServer.ingestRemote, exact Claude POST shape. Codex: Orca launches WSL Codex with CODEX_HOME redirected to the managed runtime home (~/.local/share/orca/codex-runtime-home/home), so hooks installed to ~/.codex were never read. installRemote now accepts an explicit codex home (flat layout), threaded from the relay manager; the config.toml trust write is deferred while the file doesn't exist (the launch path seeds it only-if-absent — creating it first would cancel the seed), and the manager re-runs the byte-equality-idempotent installers on later ensures (30s throttle) to upsert trust once the seed lands. Also: WSL test suites now run on Windows dev hosts (fs-backed suites skip with rig coverage noted; manager suite uses a fixed POSIX home). Co-authored-by: Orca <help@stably.ai> * fix(agent-hooks): renderer ownership gate treats wsl:* connection ids as local (STA-1515) Round-2 rig finding: with the link fixed, WSL hook envelopes reached ingestRemote and the durable cache, but useIpcEvents.applyAgentStatus drops any status whose stamped connectionId differs from the owning repo's — 'wsl:<distro>' !== null for a local repo, so every WSL-relayed status died before setAgentStatus and notifications. wsl:* ids are transport provenance, not ownership: the gate now normalizes them to local via isWslHookRelayConnectionId (shared contract, also used by the relay link when stamping), while still rejecting WSL-stamped events against SSH-owned repos. Provenance stays stamped — it is what made this drop diagnosable. Co-authored-by: Orca <help@stably.ai> * fix(agent-hooks): adversarial-review hardening for the WSL hook relay (STA-1515) Four independent review lenses over the branch; all confirmed findings fixed before the next rig round: Endpoint identity (4/4 reviewers): the guest endpoint dir was keyed by the EPHEMERAL Windows hook port, so a daemon-surviving agent kept sourcing the dead port-P1 file after an Orca restart — breaking the restart-resume acceptance criterion and regressing shipped OMP recovery. Now keyed by a restart-stable instance key (hash of the Windows endpoint file path, crossed via ORCA_WSL_HOOK_INSTANCE): the restarted instance's relay rewrites the SAME file, which is exactly what re-coordinates survivors. Restart policy: every failure arms the restart timer (one failed relaunch no longer ends self-recovery), and the timer probes wsl --list --running first — wsl -d BOOTS a stopped distro, so recovery must never resurrect a VM the user shut down; stopped-distro state is dropped instead. Failure counters reset only after 2min of stable uptime, so connect-then-die loops escalate to the 10-min cap instead of cycling every 10s. Timer policy extracted to wsl-hook-relay-recovery.ts with direct tests. Also: version-namespaced guest install dir (dev+prod instances no longer reinstall over each other; PID-suffixed tmp files), 30s install timeout (a wedged wsl.exe could pin the state machine at 'starting' forever), per-candidate node version probing (apt node 12 on PATH no longer masks nvm node 20 into a false no-node cooldown), WSL_UTF8=1 + NUL-stripped stderr (catastrophic-failure matcher survives UTF-16LE), ordered post-sentinel chunk handoff, port-fallback breadcrumb via the home handshake, bad home reply now fails the connect, missing-bundle warn-once, case-normalized distro keys, disposeAll wired to will-quit, one-shot 60s reinstall timer for single-spawn Codex trust catch-up, escaped + contract-derived spawn command. Co-authored-by: Orca <help@stably.ai> * docs: record round-3 rig validation status for agent status over WSL (STA-1515) Co-authored-by: Orca <help@stably.ai> * fix(agent-hooks): round-4 adversarial-review fixes for the WSL hook relay (STA-1515) - dropState identity race: recovery re-checks state identity after the distro-running probe await, and the manager's dropState only deletes the exact state it was armed for — an ensure() landing mid-probe can no longer have its fresh relay orphaned outside the map. - Distro-running probe fails CLOSED: a probe error no longer reports 'running', so recovery can never wsl-d-boot a distro the user shut down. - Relay spawns use --exec: bypasses the distro's default login shell (fish/nushell chsh) and passes argv verbatim, dropping the $-escape shim; same form as the Codex WSL login spawn. - Post-sentinel chunk handoff rides a microtask so an envelope in the trailing bytes can no longer dispatch before the link's notification handler is registered. - Guest relay mirrors the SSH relay's uncaughtException/unhandledRejection posture. - Replay cache capped at 256 panes with recency eviction (the WSL relay has no per-pane teardown signal); meta map kept in lockstep. - Launch script derives the stale-exit code from the shared contract constant; one-shot reinstall timer refuses to arm after dispose. - New oracles: sentinel unit suite, fs-bridge scoping suite, fixed-token 403/204, EADDRINUSE endpoint-file rewrite, cache-cap eviction, and the recovery/manager race regressions (verified to fail with fixes reverted). - Doc: round-4 review section + revised curl.exe stance (kept as the no-node fallback — Codex is a native binary; fresh distros ship no node). * docs: record round-4 pinned rig validation for agent status over WSL (STA-1515) --------- Co-authored-by: Orca <help@stably.ai> Co-authored-by: Brennan Benson <brennanbenson@Brennans-MacBook-Pro.local> |
||
 Doan Bac TamandBrennan Benson
|
fa2b228ad3 |
feat(rate-limits): Grok CLI OAuth weekly credit usage (#7869)
* feat(rate-limits): Grok CLI OAuth weekly credit usage in status bar Read ~/.grok/auth.json (read-only), fetch billing credits via cli-chat-proxy, and surface Grok in Settings, status bar toggles, and rate-limit polling alongside other usage providers. * fix(grok): clarify comments and address CodeRabbit review - Shorten Why comments per AGENTS.md; fix billing period end fallback. - Share GrokAccountStatus type; hash-based locale keys; reload why in Settings. * docs(grok): plain-language comments and Settings copy * feat(stats): subscription usage section with Grok in Stats & Usage Surface rate-limit weekly credits in Settings > Stats & Usage and link to Accounts for setup. * feat(stats): Grok tab in Usage Analytics dropdown * refactor(stats): drop Subscription usage block; align Grok pane with Codex * fix(grok): align settings copy and visibility tests * fix(grok): add localization catalog entries * fix(grok): avoid eager usage refresh fanout * fix(grok): harden usage refresh visibility * test(grok): cover account status privacy boundary * fix(grok): target refreshes and redact auth errors * fix(grok): hide usage UI for signed-out users and align empty states - Treat a token-less auth.json (e.g. after grok logout) as signed out instead of surfacing a permanent status-bar error. - Map billing responses without credit usage to 'unavailable' so plans with no weekly credits hide the bar like Claude API-key billing. - Gate the grok status-bar item and toggle on CLI PATH detection, matching claude/codex/gemini/kimi. - Guard an empty GROK_CLI_CHAT_PROXY_BASE_URL from producing a relative billing URL. - Drop dead minimax/kimi Stats & Usage search keywords left from the removed subscription section. - Add missing grok search keyword catalog entries and translate the English-stubbed stats keywords in es/ja/ko/zh. * test(ipc): mock grok account registrar in register-core-handlers test --------- Co-authored-by: Brennan Benson <brennanbenson@Brennans-MacBook-Pro.local> |
||
|
Jinjing
|
5d5e2c4b95 |
Add updater diagnostics and refine daemon teardown policy on quit (#7544)
* Add updater diagnostics and refine daemon teardown policy on quit
Introduce detailed logging, crash breadcrumbs, and telemetry tracing
events for the update quit-and-install lifecycle. This improves
visibility into macOS Squirrel.Mac state transitions, install guard
timeouts, and pre-quit cleanup failures.
Also, formalize and unit test the daemon teardown policy on quit to
explicitly keep the daemon alive (via disconnect) during updater and
normal quits so that daemon-backed PTYs are preserved for warm reattach.
* Remove updater lifecycle logging
Remove calls to `recordUpdaterLifecycle` across the auto-updater modules. This cleans up unused or redundant telemetry events during macOS update download, ready, and pre-quit installation phases.
* Remove redundant quit daemon teardown policy helper
Inline the daemon teardown check directly into the `will-quit` event
handler in the main process. The deleted `shouldShutdownDaemonForQuit`
helper simply returned `isDevParentShutdownRequested`, so using that
check directly simplifies the codebase and allows removing the redundant
policy and test files.
* Clarify PTY cleanup targets only local in-process PTYs
Update doc comments, test descriptions, and telemetry events to make
it explicit that only local, in-process PTYs are killed on quit.
Daemon-backed PTYs are preserved via daemon disconnect.
* Recover quit-for-update state on autoUpdater error events
- Reset update flags when the installer fails asynchronously via the
autoUpdater 'error' event (such as "no staged update" errors).
- Prevents the app from getting stuck in a half-transitioned state
where PTYs are killed and windows refuse to reopen.
- Migrate legacy updater warning logs and crash breadcrumbs to use
the structured recordUpdaterLifecycle diagnostic system.
* Simplify duplicate isMacInstallerReady() call in update-downloaded handl
Cache the result in a local variable so the deferred-status check reuses
the value from the log call instead of recomputing it.
* fix: tighten quit-and-install failure recovery
Only recover stuck quit-for-update flags after native quitAndInstall is
invoked and before install is committed. Defer PTY kill and close-listener
removal until after a successful native handoff so sync install failures keep
the session intact, and ignore post-commit updater errors for UI/mac state.
* fix: typecheck login-shell args check in shell-ready test
Avoid args.includes('-l') on a const tuple union that only admits '-i'
on one branch; compare args[0] instead so merge-base typecheck passes.
|
||
|
Jinjing
|
6166f6c47a |
feat(linear): add server-side issue filters for status and other properties (#8021)
Add attribute filters (status, priority, assignee, labels) to the Linear issues list, applied in GraphQL before pagination so hasMore matches the filtered set. Thread filters through IPC/RPC/store cache with invalidation on issue mutations, and mirror GitHub-style filter chrome on TaskPage. |
||
|
Jinwoo Hong
|
27f30c2548 | fix(windows): open WSL workspaces in VS Code remote (#7982) | ||
|
Jinwoo HongandOrca
|
6731773e4e |
Route provider account surfaces through the active Remote Orca Server (#7999)
Co-authored-by: Orca <help@stably.ai> |
||
 BingZandJinwoo Hong
|
5b1b8cc61f |
fix(preflight): resolve WSL/SSH agent paths past shell aliases (#7867)
* fix(preflight): resolve WSL/SSH agent paths past shell aliases LeanCTX and similar tools wrap claude/codex as interactive shell aliases. command -v then returns alias text, which fails absolute-path detection and hides installed agents (#7816). Prefer bash type -P, then zsh type -p, then command -v for dash/sh fallback in WSL agent discovery, WSL isCommandOnPath, and remote relay probes. * fix(preflight): harden alias-safe PATH lookup chain Require non-empty results between type -P, type -p, and command -v so bash type -p empty success cannot skip later lookups. * fix(preflight): resolve agent executables directly from PATH --------- Co-authored-by: Jinwoo Hong <73622457+Jinwoo-H@users.noreply.github.com> |
||
|
|
e7ee15f4b2 |
Improve workspace cleanup list (#7053)
* Improve workspace cleanup list Co-authored-by: Orca <help@stably.ai> * Address workspace cleanup review feedback Co-authored-by: Orca <help@stably.ai> * Fix workspace cleanup perf findings Co-authored-by: Orca <help@stably.ai> * Avoid stale cleanup progress cache Co-authored-by: Orca <help@stably.ai> * Complete workspace cleanup perf fixes Co-authored-by: Orca <help@stably.ai> * Fix worktree list option forwarding Co-authored-by: Orca <help@stably.ai> * Address workspace cleanup review nits Co-authored-by: Orca <help@stably.ai> * Fix workspace cleanup removal review findings - Fail a queued removal that now needs a force the user never approved (confirm-time approvedCandidates snapshot compared in preflight) - Reword the 120s removal timeout to say removal continues in background - Wire suppressPreservedBranchToast into cleanup removals - Stop statting a repo after the first activity metadata timeout - Document the WSL 9P best-effort stat gap; drop unused locale key Co-authored-by: Orca <help@stably.ai> * Split workspace-cleanup slice test to satisfy max-lines Rebasing onto latest main pushed the combined store-slice test over the 800-line cap. Extract shared fixtures into a test harness and split the suite into scan-progress and removal-preflight files instead of adding a forbidden max-lines suppression. --------- Co-authored-by: Orca <help@stably.ai> Co-authored-by: Brennan Benson <brennanbenson@Brennans-MacBook-Pro.local> |
||
|
Jinjing
|
69befad13e |
Harden layout validation, watcher lifecycle, and connection robustness (#7924)
* Harden layout validation, watcher lifecycle, and connection robustness - Throw instead of silently skipping when the packaged daemon-entry is missing, preventing layout regressions from passing build checks. - Terminate idle parcel-watcher processes to reclaim native handles and avoid crash-prone native node module teardowns on shutdown. - Bind the persisted WS fallback port first to prevent orphaning active mobile pairings when the preferred port becomes free again. - Cap concurrent disk reads for restored dirty tab verification at three to prevent startup connection bottlenecks on remote SSH workspaces. * Queue file IDs instead of snapshots in restored conflict scans This avoids using stale file snapshots (e.g., outdated disk signatures) if a tab is saved, closed, or re-baselined while waiting in the queue behind the concurrency limit. The live state is now fetched from the store and validated immediately before initiating the disk read. |
||
|
Jinjing
|
ee823b766c |
Move notification status helper to Contents/MacOS (#7931)
On macOS 26, UNUserNotificationCenter aborts when executables are run from Contents/Resources because bundleProxyForCurrentProcess returns nil. Moving the orca-notification-status helper to Contents/MacOS next to the main Electron executable ensures proper bundle resolution and avoids immediate crashes. |
||
|
NeilandOrca
|
06afbc4a4b |
Add optional Orca account sign-in (#7515)
* auth v1 * fable review * lint * Account menu with org membership management Default UX is a compact account menu (sign in, organization selection, sign out) that renders only when cloud auth is configured; adds an organization members dialog (invite, role, remove) gated on server-side role checks. The multi-profile switcher UI is preserved behind ORCA_MULTI_PROFILE_UI=1. Co-authored-by: Orca <help@stably.ai> * Gate the optional account sign-in UI to dev builds The account switcher stays hidden in packaged builds while the feature is in progress. Dev builds still show it when the client env vars are set, and a dev-only Settings > Dev Tools > Orca Cloud section mirrors it. Co-authored-by: Orca <help@stably.ai> --------- Co-authored-by: Orca <help@stably.ai> |
||
|
NeilandOrca
|
25ea2bbfd1 |
onboarding: seamless macOS notification permission step with live state detection (#7684)
* feat(onboarding): state-aware macOS notification permission step The Set up notifications step showed a one-size-fits-all 'Open Mac Settings' button that simultaneously fired the macOS permission prompt and opened System Settings — two competing system UIs, with System Settings unnecessary for the common fresh-install case. Electron exposes no API to read macOS notification authorization, but scheduling outcomes do reveal it: a silent probe notification's 'show' event means permission is granted, 'failed' means delivery is blocked. A new notifications:probeDelivery IPC runs that probe (cached via passive delivery evidence and a persisted confirmation flag), and the onboarding card now renders the real state: - fresh install: the probe itself pops the native Allow dialog the moment the step opens; the card flips to 'Notifications are enabled' automatically when the user clicks Allow (silent 2.5s re-probes) - blocked: amber card with an Open System Settings deep-link, which also self-heals once the user flips the toggle - granted: green confirmation card The test-notification button now feeds the same card instead of the ambiguous 'if no banner appeared…' toast during onboarding. Co-authored-by: Orca <help@stably.ai> * fix: don't log expected probe rejections while polling for permission Co-authored-by: Orca <help@stably.ai> * fix: amber warning styling + single stable dev bundle id for notifications - Blocked card now uses the app's shipped amber idiom (tinted surface with amber title/body) instead of white-on-amber-wash, which read muddy in dark mode; macOS permission card split into its own module to stay under the max-lines budget. - Dev instances previously minted a unique macOS bundle id per branch x Electron version, registering a new Notification Settings entry every time ('Orca: <branch>' rows piling up forever) and pointing the settings deep-link at ids System Settings can't resolve. All dev instances now share com.stablyai.orca.dev: one Notification Center entry, one permission grant covering every dev build. Co-authored-by: Orca <help@stably.ai> * fix: tighten macOS permission card copy Body copy was one long sentence; now a single short instruction with 'Updates automatically.' as a separate dimmer line. Also repairs locale catalog parity for keys introduced by commits rebased into this branch. Co-authored-by: Orca <help@stably.ai> * fix: drop 'Updates automatically.' line; ad-hoc sign dev app copies The extra line read as confusing filler — the cards now carry one short instruction each. Dev Electron copies had broken code signatures (the Info.plist identity edits invalidate the ad-hoc seal), which macOS punishes by refusing Notification Center registration outright: every dev notification failed with UNErrorDomain error 1, the app never appeared in System Settings > Notifications, and the settings deep-link had nothing to land on. The dev runner now ad-hoc re-signs the copied bundle after the plist edits (bundleLayoutVersion bumped so stale unsigned copies are recreated). Verified end-to-end: runner-built copy passes codesign --verify --deep, probe delivery returns delivered, the onboarding card flips green in dev, and the deep link opens the dev app's own notifications pane. Co-authored-by: Orca <help@stably.ai> * fix: drop confusing copy line; session-only permission evidence Removes the 'Updates automatically.' line from both permission cards. Also drops the persisted notificationDeliveryConfirmed flag: OS-level permission changes between sessions, and a stale positive rendered a false green card. Delivery evidence is now session-scoped only. Documented detection ceiling (verified empirically on macOS 26): while the permission dialog is unanswered — and when notifications are toggled off in System Settings after being authorized — macOS accepts requests and silently swallows them, with no public API (Notification Center delivered-history and legacy ncprefs both included) able to distinguish that from real delivery. 'failed' remains definitive for unsigned builds and dialog-level denials. Co-authored-by: Orca <help@stably.ai> * feat: real macOS notification permission readout via native helper Electron has no API for UNUserNotificationCenter authorization, and every observable fallback lies: scheduling succeeds (and getHistory lists the notification) even while macOS silently swallows display because the permission dialog is unanswered or notifications were toggled off in System Settings. The onboarding card therefore showed 'enabled' after the user disabled notifications. Adds native/notification-status-macos: a tiny Swift binary that prints the app's real authorization status. It runs from inside the app bundle (NSBundle resolves the bundle by walking up from the executable) and embeds the app's CFBundleIdentifier in a __TEXT,__info_plist section so every codesign --force pass — electron-builder's signing or the dev runner's ad-hoc deep sign — derives the identifier macOS keys notification records to. Spawning it from the app returns authorized / denied / not-determined exactly matching System Settings. notifications:probeDelivery now prefers this readout (authoritative, silent), firing at most one dialog-trigger probe per session while the decision is pending, and falls back to the previous delivery-probe heuristics when the helper is unavailable. The card polls the readout silently in every state, so toggling Allow notifications in System Settings flips the card within a poll — both directions, verified live. Test notifications also consult the readout so 'delivered' is no longer claimed for swallowed notifications. Packaged builds ship the helper via extraResources and sign it in afterPack like the computer-use helper; dev copies compile it on demand (swiftc, non-fatal when missing) with the shared dev bundle id. Co-authored-by: Orca <help@stably.ai> * feat: in-app fallback for swallowed notifications + permission card in Settings - Dispatch now consults the authorization readout before creating a native notification: when macOS would silently swallow it (denied or prompt unanswered) it returns reason 'blocked-by-system' instead of piling invisible notifications into Notification Center. The terminal notification path surfaces that as a once-per-session in-app toast with an Open System Settings action. Mobile fan-out is unaffected. - Settings > Notifications now shows the same live permission card as onboarding (moved to components/notifications/), polling the readout so System Settings changes reflect within seconds, and the test button updates it inline. - Test sends that are blocked at the OS level now show the settings-pointing failure toast instead of a generic error. Co-authored-by: Orca <help@stably.ai> * fix: hide macOS permission card while Orca notifications are disabled A green 'Notifications are enabled' card next to a disabled Enable Notifications toggle read as a contradiction — the card now renders (and the readout polls) only while Orca's own notifications setting is on. Also single-flights the authorization helper so simultaneous agent completions share one readout process. Co-authored-by: Orca <help@stably.ai> --------- Co-authored-by: Orca <help@stably.ai> |
||
|
Jinjing
|
4ded642a3a |
Detect and surface recoverable zero-turn sessions in AI Vault (#7889)
Identify Claude sessions with no saved conversation turns but possessing recoverable signals such as queued prompts or subagent transcripts. This displays them in the sidebar with a "Not saved" badge instead of filtering them out as empty, and provides detailed notices to recover them via logs while disabling standard resume actions. Additionally, extract Gemini session parsing logic into a separate module and support counting sibling subagent transcripts across local and remote SSH session scans. |
||
|
Neil
|
8dbcb2be3e |
feat(floating-terminal): amber attention dot for unacknowledged bells and completions (#7888)
Show a small amber dot on the floating-workspace launcher (both the floating-button and status-bar triggers) whenever any floating-workspace tab still has an unacknowledged terminal bell or agent completion, and a composited amber dot on the Windows tray icon when the window is minimized/hidden. Both clear through the existing show-until-interact paths — engaging with or closing the offending tab drops the dot with no stale unread state left behind. The launcher dot derives from the existing per-tab/per-pane unread maps via a new selectFloatingWorkspaceHasUnread selector (primitive boolean, empty- workspace early return, no bespoke state). The tray dot rides the notification dispatch and clears on window show/restore. |
||
|
Jinwoo Hong
|
685418d8e4 | fix(daemon): daemon cannot start in 1.4.129-rc.1 — electron require leaked into daemon-entry chunk (#7844) | ||
|
NeilandOrca
|
4a6d12959f |
fix: recover fresh local terminal spawns from a deleted saved cwd (#7847)
After #7750 removed the containment guard, the residual #7239 failure mode is a persisted/inherited startupCwd whose directory no longer exists: every spawn dies with the provider's missing-directory error. Fresh local renderer spawns now opt in (cwdFallback: 'worktree') to recover at the workspace root with a generic in-terminal notice; reattach, SSH, remote-runtime, runtime/API and mobile callers keep exact cwd semantics, and existing directories — including outside the worktree (#7685) — spawn as requested. Co-authored-by: Orca <help@stably.ai> |
||
|
Jinwoo HongandOrca
|
5a390cd60e |
Fix Claude Agent Teams detection without Claude CLI (#7834)
Co-authored-by: Orca <help@stably.ai> |