Main inlined an OSC 133 parser with new 133;C onCommandStarted support;
our branch parses via the shared scanner so main-side side-effect facts
stay byte-identical. Resolution keeps the shared-scanner architecture and
adds 133;C to it; the foreground-agent tracker hooks into the branch's
single handleCommandFinished policy (bytes and sideEffect facts alike).
Co-authored-by: Orca <help@stably.ai>
This reverts commit a94d692cd5, restoring the
shorter "Command" label and the Play icon on the quick command entry
(PRs #5528 and #5674).
Co-authored-by: Orca <help@stably.ai>
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>
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>
(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>
sherpa-onnx cannot load model files or hotword lists from Windows
paths containing non-ASCII characters.
- Resolve speech model and hotwords paths to ASCII-safe alternatives
like ProgramData when the default userData path contains non-ASCII
- Migrate existing downloaded models from the legacy non-ASCII cache
- Ensure deleted models are removed from the legacy source folder to
prevent them from being re-migrated on subsequent launches
Fixes the PTY fallback path so Claude Fable usage carries a concrete reset timestamp and the existing status tooltip renders the Fable reset countdown.
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>
- 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>
* fix: stop Claude account loading after denial
* fix(claude-accounts): tear down whole login tree on OAuth denial
Use killChild() (not child.kill) on the denial path so the detached
POSIX process group and Windows child tree are killed, matching the
timeout/abort paths and honoring the detached-spawn comment. Strengthen
the regression test to assert the process-group teardown.
Co-authored-by: Orca <help@stably.ai>
---------
Co-authored-by: Neil <charlie-eng@stably.ai>
Co-authored-by: Orca <help@stably.ai>
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>
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>
* feat: allow custom worktree branch names
* refactor(ui): render custom branch name field unconditionally under advanced container
* fix(composer): hide manual branch field when a work-item source drives the branch
A tracked PR/issue/MR/Linear source derives the branch itself, and a linked
GitHub PR re-resolves the branch name at submit — so an override typed in the
Advanced branch field was silently ignored. Only render the field for the
typed-name and base-branch flows, where the manual override is honored.
Co-authored-by: Orca <help@stably.ai>
---------
Co-authored-by: Wolfgang Schoenberger <221313372+wolfiesch@users.noreply.github.com>
Co-authored-by: Neil <4138956+nwparker@users.noreply.github.com>
Co-authored-by: Orca <help@stably.ai>
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>
rc kind can now append an extra prerelease identifier (rc.N.perf):
sorts above its base rc.N but below rc.N+1, so perf-branch RCs are
identifiable by clients without outranking the main RC series.
Co-authored-by: Orca <help@stably.ai>
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>
The POSIX agent-hook script for every curl-based agent inlined the full
event payload via `curl --data-urlencode "payload=${payload}"`. Tool
output can be tens of KB, so the resulting process command line could be
multi-KB — which endpoint security tools (e.g. Microsoft Defender for
Endpoint) flag as an oversized/suspicious command line. That produced a
false-positive detection on Orca's own loopback (127.0.0.1) telemetry POST.
Stream the payload to an mktemp file and post it with
`--data-urlencode "payload@$payload_file"` instead. The urlencoded body on
the wire is byte-identical, so the agent-hook receiver is unchanged; the
payload simply never appears on a process command line. `trap ... EXIT`
removes the temp file on every exit path. Small bounded metadata fields
(paneKey/tabId/worktreeId/env/version) stay inline.
Applied to all curl-based agents: claude, codex, command-code, copilot,
cursor, droid, gemini, grok, antigravity. (amp/hermes/opencode post via
the HTTP request body and were never affected.) The Windows post-command
shares the same latent pattern and is flagged as follow-up.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Neil <4138956+nwparker@users.noreply.github.com>
Co-authored-by: Orca <help@stably.ai>
* fix(gitlab): port-aware self-hosted host recognition
Use the URL host (including a non-default web/API port) as the GitLab
host identity instead of the port-less hostname, and match known hosts
port-aware:
- A known-host entry without a port matches any port of the same
hostname (preserves legacy bare-host and gitlab.com recognition).
- A known-host entry WITH a port matches only that exact host:port, so
two services sharing a hostname on different ports (e.g. a GitLab and
a Gitea) are no longer conflated.
- For ssh/git remotes the port is a transport port (e.g. ssh :2222) and
is dropped; for http(s) remotes the port is the endpoint and kept.
- Also capture an optional :port in parseGlabAuthStatusHosts so a
self-hosted GitLab on a non-default port is discovered correctly.
* fix(gitlab): per-connection known-hosts cache + port-aware auth-status parsing
getGlabKnownHosts() was connection-blind and cached process-globally,
and on any failure it cached [gitlab.com] forever — so a repo on an SSH
connection never discovered its self-hosted host once a probe failed
before the tunnel was ready.
- getGlabKnownHosts(connectionId?) now caches per connection so a
connected repo's authenticated hosts don't leak into the local
context (or vice versa).
- The failure fallback (canonical default) is no longer cached, so a
later probe can re-discover the real host once auth/tunnel is ready.
- parseGlabAuthStatusHosts captures an optional :port on both the
'Logged in to <host>' and header-style lines, keeping two services on
the same hostname distinct by port.
* fix(gitlab): isolate unresolvable projects instead of cwd-fallback that hits exit 128
listIssues/getIssue fell back to an unscoped 'glab issue list' / 'glab
issue view' that infers the project from cwd. For a repo on an SSH
connection cwd is not the repo dir, so glab runs git resolution in a
non-repo dir and fails with 'git: exit status 128'. In an 'All projects'
aggregate one such failure could sink the whole issues panel.
When a projectRef cannot be resolved, return a structured, isolated
per-project result (listIssues: { items: [], error: not_found };
getIssue: null) and spawn no glab subprocess. Behavior is unchanged when
a projectRef IS resolved (the scoped '-R' / 'api projects/...' path).
* fix(gitlab): recognize modern /-/work_items/<iid> issue URLs
Modern GitLab emits issue URLs as /-/work_items/<iid> in addition to the
legacy /-/issues/<iid>. The URL classifiers only matched /-/issues/, so
work-item-form issue links went unrecognized.
Extend the gitlab-links parsers (parseGitLabIssueOrMRNumber /
parseGitLabIssueOrMRLink, which also backs isWorkItemLookupText) and
isGitLabIssueUrl to accept /-/work_items/<iid>, mapping it to an issue
work item with the same project-path + iid extraction.
* fix(gitlab): thread connectionId into getGlabKnownHosts call sites
Follow the existing connectionId-threading pattern: pass the repo's
connectionId into every getGlabKnownHosts() call (client.ts,
work-item-details.ts, orca-runtime.ts) so the per-connection known-hosts
cache is keyed correctly and self-hosted hosts are discovered against
the right glab context.
* docs(gitlab): use generic example hosts in comments
* fix(gitlab): pass self-hosted host:port via GITLAB_HOST (glab --hostname rejects ports)
* polish: satisfy oxlint curly + oxfmt on merged gitlab port-recognition code
Co-authored-by: Orca <help@stably.ai>
---------
Co-authored-by: Ptah-CT <auctor@xinfty.space>
Co-authored-by: Neil <4138956+nwparker@users.noreply.github.com>
Co-authored-by: Orca <help@stably.ai>
Grafts the sleep/wake terminal-recovery rework (PR #7260) into the RC
perf line. Resolution principle: this branch's structure wins; the
rework's delivery-accounting semantics graft in.
Kept from orca-performance:
- Producer flow control v19: PRODUCER_FLOW_CONTROL_ENABLED kill switch,
every updateProducerFlowControl call site where pendingData mutates
(256KB/32KB watermarks, pause/resume daemon notifications), and
producerFlowControl.releaseAll/release on destroy/exit. Flow control
keys off pendingData only, so it composes with the new renderer
delivery accounting untouched.
- PTY_BATCH_INTERVAL_MS = 2 with its daemon-stream-data-batcher pairing
comment; merged resync tests adapted to the 2ms window.
- Chain structure: hidden-delivery gate (drop + restore markers),
pending-output cap with droppedOutput sentinel, overflow marker set.
Grafted from fix/terminal-wake-ack-recovery:
- TCP-style cumulative delivery accounting (sentChars/ackedChars
max-merge; legacy delta payloads still accepted) replacing the
relative in-flight counter map — lost ACKs self-heal instead of
becoming permanent gate debt.
- Solicited delivery resync (probe on data-arrival-while-fully-gated,
verified-state reconcile on reply, 5s hygiene-only timeout).
- Renderer lifecycle reset zeroes the delivery gate; powerMonitor
system:resumed wake IPC (merged cleanly, no conflicts).
Both test suites pass composed: pty.test.ts 243 (flow-control +
cumulative-ack/resync), full ipc/terminal-pane/daemon gates green with
only the pre-existing WSL env-import failure.
Co-authored-by: Orca <help@stably.ai>
Jinwoo-HandOrca
Neil
Brennan BensonandOrca
Jinjing
andrewroxby
PPandping
m11yandClaude Fable 5
Soichi Sumi
haozhenfei
Jinwoo Hong
