mirror of
https://github.com/stablyai/orca.git
synced 2026-11-19 04:23:57 +00:00
c04582a4414df573ebefc12beb0e51d893d8564e
155
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
|
|
a7b5c0a8f2 |
Run supported Mac chats and background agents under Orca Agents (#41917)
* Name terminal agent hosts separately from Orca * Account for one startup process naming write * Use the Node runtime for the Linux naming contract * Run supported local agents under the independent host * Preserve retirement type narrowing and valid test plans * Align restart warnings and satisfy changed-file lint rules * Declare the current daemon protocol in build compatibility metadata * Keep hosted agent output complete through process exit * Keep the output regression focused on catalog failure and cleanup * Keep output regression fixtures compatible with quality checks * Publish Claude exits after hosted process absence is proven * Pass session fences in the hosted Claude regression * Fall back to direct chat launch when the host image is unavailable * Keep the hosted Claude recovery fixture independent of OS signing * Settle refused agent launches and preserve exit observation wakes * Keep provider population fixture defaults independent of array indexes * Document direct launch when agent host connection checks fail * Retain native provider no-start outcomes before refusal * Keep chats working when the agent host cannot read their folder * Preserve native missing-folder coverage after the access preflight * Keep desktop permission notices at the desktop launch boundary * Bound hosted launch setup and keep hosted chats out of host process growth A host still starting no longer holds a chat past its bootstrap deadline, the host's code signature is read once per host process, and the native owners the host starts for the app do not count as its terminal work. The hosted launch test runs under Node, whose destroy closes only the host's own socket copy. * Keep hosted owner pipes blocking, recheck the host before dispatch, and keep launch-folder errors The native owner clears O_NONBLOCK that a host's sockets carry, so its gate read and the CLI's stdio behave as on a direct launch. A reconnect to another host before dispatch launches here instead of on an unprobed host, and a host spawn error keeps its code and path so a deleted folder is still reported as missing. * Skip hosted owner subtrees in the growth walk and expire the host image read The growth sample now stops at hosted owners instead of walking their subtrees again, so it never does more work than before. A host's code-signature read is reused only for 10 s, so a bundle removed while the host runs is noticed on a later launch. * Satisfy changed-code quality for the hosted launch wait and its test * List provider requests in the daemon router's exhaustive switch * Launch a chat here when the host's start fails before attach The native owner runs its CLI only after this app opens its gate, which attaching does. A host spawn error, a lost reply or pipes that never arrive therefore leave no CLI running and no input sent, so the chat now launches here instead of failing. Once attached, a failure still ends the launch. Tests cover a host that can't read or can't start in the chat's folder, a host that never answers the folder probe, a lost spawn reply, and the folder-access notice verdicts. * Keep fallback time inside the owner's startup deadline A hosted start could spend 10 s on the spawn reply and 10 s on its pipes, past the owner's 15 s bootstrap deadline, so a late failure ended the chat instead of launching it here. The hosted attempt now has one 8 s budget from the start of the launch. A folder denial whose app-side recheck finished after a newer readable verdict restored a cleared notice; the newest verdict for the same helper and folder class now wins. * Fit the hosted start inside the owner's bootstrap and proof budget The hosted attempt's budget is now the bootstrap deadline minus the ownership proof window, so a hosted owner or a direct fallback still has time to prove ownership. A timer bounds the spawn request too, so a stalled event loop can't let the bootstrap deadline fire first. Only a newer readable verdict from the same helper supersedes a pending folder denial, so a later denial in another folder no longer hides a real notice, and the record holds one entry per folder class. * Track folder re-allows per helper so a replaced helper can't erase a notice Keying the re-allow marker by folder class alone forgot an old helper's re-allow once a new helper read that class, so the old helper's late denial overwrote the new helper's notice. Marks are per helper and class again, capped to the most recent pairs. * Supersede pending folder denials directly instead of keeping re-allow marks Capped marks could evict a re-allow while its denial was still being read, reviving the stale denial. A readable verdict now flags the same helper's pending denials for that folder class, and only denials still being read are kept. * Keep folder verdicts off blocking reads and bound the whole hosted start Native-chat folder verdicts now reach the shared verdict telemetry, so its spawner-path check and pid-file read are now asynchronous instead of blocking the main process. The extracted backlog code only builds the sampler, so the server keeps calling the existing probe directly. Lowered the blocking-helper baseline rows that shrank. Connecting to the host and checking its signature and the folder now share the hosted start budget, so a slow host still leaves a direct launch time to prove ownership. * Lower the sync-fs baseline row the async telemetry checks emptied * Drop folder verdicts from host checks that outlived the start budget A host check still running when the direct fallback took over could report a late denial and bring back a notice a newer readable verdict had cleared. Checks past the budget now count as abandoned and report nothing. The integration test now ends the CLI's input only after it starts, since its owner allows only a short grace once input closes. |
||
|
|
a6918ecb97 | Merge in branches | ||
|
|
fa700641aa |
test: remove low-value native input, renderer and SQLite tests (#43596)
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
7668161d34 |
test: remove low-value mobile, native and runtime tests (#43586)
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
a9e6b80aa4 | fix: guard macOS 27 dead keys before native menu routing (#43264) | ||
|
|
c7f83f4e19 |
fix(computer): stop macOS permission status probes from leaking suspended helpers (#43172)
Spawn the status helper directly instead of through LaunchServices, and have it re-exec itself as its own responsible process so TCC still checks the helper's grant. A timeout now stops the probe by PID, so a late LaunchServices launch can no longer leave a suspended helper behind. Startup kills suspended probes left by older builds. Co-authored-by: Mikel Vallejo <58292517+VallejoM@users.noreply.github.com> |
||
|
|
98fadae4d8 |
fix(computer): keep the agent's window id as the target window on Windows (#43162)
The Windows helper checked an agent's --window-id against Process.MainWindowHandle, a z-order guess that moves to a tooltip or another window of the same app. The window id is now the target: a live, visible top-level window of the target process, carried through the tree read, frame, screenshot, raise, keyboard input and the observe after the action. Co-authored-by: toodeceptive <137078450+toodeceptive@users.noreply.github.com> |
||
|
|
980352d897 |
Reduce native chat memory by replacing per-chat Electron supervisors (#42760)
* Launch provider CLIs without a per-chat runtime supervisor * Initialize process-owner fixtures with an isolated app environment * Retire provider output transports and failed ownership drafts * Finish provider stops when the session group empties * Complete native cleanup from verified watcher exit and cancel held launches * Let native watcher finish one-shot provider exits * Retire late ownership records after canceled provider launches settle * Reap each provider subtree in an isolated native parent * Preserve provider outcomes and compatibility with native parents * Publish actual provider identities and retire native parent observers * Retire numeric writer PIDs when their group is proven absent * Retire provider transports and accept proven group absence * Fence retired process identities across asynchronous cleanup * Use native timer types and readonly process fixtures * Reuse bounded timeout for compatibility output drain * Release launch scopes held by late transport error handlers * Keep processless launch failures out of watcher waits * Align lifecycle validation with final ownership contracts * Remove duplicate helper preparation from the dev launcher * Keep legacy fallback in place and simplify native ownership handoff * Defer background provider maintenance until the first window * Release startup hook maintenance for terminal launches * Use relocated artifact module in native prebuild scripts * Preserve agent failure outcomes and clean tools before ownership is lost * Budget ownership capture for its bounded identity reads * Keep lifecycle regressions at their behavior-owning seam * Explain the fake-clock process-read fixture * Capture tool ownership before restart recovery removes the CLI * Skip descendant inventories for absent provider roots * Make final output drain regression test deterministic * Keep output drain fixture assertions type-safe with IPC * Use existing provider fixtures across Node and Bun * Prepare native provider helper for Codex contract CI * Verify provider ownership using its recorded start precision * Preserve exact provider ownership through reader failures * Retire exhausted recovery records without blocking watcher cleanup * Distinguish exited descendants from unreaped zombies * Preserve group retirement without trusting an exited root * Keep CLI input owned through native parent loss * Keep watcher cleanup proof through lost-parent retirement * Guard retirement against late watcher cleanup proof * Contain provider launches that fail before stdio setup * fix(claude): preserve startup diagnostics after input pipe failures * fix(provider): capture lost outcomes before native completion cleanup * fix(provider): retire speculative captures after a valid outcome * fix(provider): share spawn absence classification through its existing boundary |
||
|
|
49ac2f82d1 |
fix(wsl): start the distro holder without a visible console window (#41126)
The WSL distro holder was spawned with DETACHED_PROCESS, so it had no console. System32's wsl.exe then starts the WSL package's wsl.exe, a console program, and Windows gave it a new visible console: an empty terminal window on every Orca start with a saved WSL host, which also killed the holder when closed. Use CREATE_NO_WINDOW in place of DETACHED_PROCESS so the holder owns a hidden console that its children inherit. The flags cannot be combined (Windows ignores CREATE_NO_WINDOW with DETACHED_PROCESS). The new process group, job breakaway retry and NUL-only handle list are unchanged. Fixes #41125 Co-authored-by: Neil <neil@stably.ai> |
||
|
|
f6a212287e |
test: remove nine low-value suites and trim twelve others (#42976)
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
36bdeabc1d |
test: remove 55 low-value suites (#42931)
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
7ade4f96da | chore(dx): silence pre-commit lint-staged chatter and pnpm install noise (#42915) | ||
|
|
126e59e1ad |
test: remove 55 low-value suites (#42916)
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
2dadb11039 |
test: remove 46 low-value suites (#42874)
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
8221969006 |
fix(computer-use): read SwiftUI labels and name custom actions on macOS (#42829)
SwiftUI controls advertise only AXAttributedDescription but answer
AXDescription, so treat AXDescription as readable in that case and give
Calculator keys and System Settings sidebar buttons their labels.
Render custom actions ("Name:X\nTarget:..\nSelector:..") as just their
name, and match perform-secondary-action on the exact pretty name first so
a custom "Delete" is not shadowed by AXDelete.
Refs #42828
Co-authored-by: Neil <neil@stably.ai>
|
||
|
|
b6173eff04 |
fix(computer): click macOS windows that refuse focus when --restore-window is allowed (#42824)
* fix(computer): click macOS windows that refuse focus with --restore-window A raised window that never becomes the focused window (Archicad's CEF start window) failed every --restore-window click with window_not_focused, because the click fence required the target to be focused as well as topmost at the point. A real click reaches the topmost window whether it is focused or not, so when the target stays unfocused the click now goes ahead while the target is topmost at the point, and is refused before anything is sent when another window covers it. Afterwards the app the user had in front comes back if the raise or the click left the target app in front without the target window. * fix(computer): restore the front app only after a click on a window that refused focus * fix(computer): keep the target app in front when an unfocused-window click opened something * fix(computer): count small popups and re-read the front app last before restoring it |
||
|
|
8e693efda4 |
feat(computer): add flags for smaller screenshots on macOS, Windows and Linux (#42801)
* feat(computer): add opt-in flags for smaller computer-use screenshots Every observe and action screenshot came back as a full-resolution PNG, so a Retina window cost about four times the pixels it needed. --screenshot-scale caps pixels per window point, --screenshot-max-edge caps the longer side, and --screenshot-format jpeg re-encodes it, on macOS, Windows and Linux. Defaults are unchanged. * fix(computer): treat requested screenshot sizes as ceilings and skip repeat encodes |
||
|
|
28fd7052ad |
feat(computer): list macOS apps without a Dock icon while they show a window (#42786)
* feat(computer): list macOS apps without a Dock icon while they show a window list-apps on macOS only listed regular apps, so menu bar and other accessory apps could not be found or targeted by name. Accessory apps are now listed while they own an on-screen window between the normal and status bar levels, which keeps the dozens of windowless system agents out, and each listed app says whether it is an accessory app. * fix(computer): list only normal-level accessory windows and resolve hidden accessory apps * fix(computer): prefer a same-named regular app when resolving, not in the list order |
||
|
|
07073db317 |
test: remove 31 low-value suites from combined audit (#42666)
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
3e123c6a23 |
fix(computer-use): turn on full macOS accessibility for Orca and Gecko apps (#42652)
The helper asked Chromium for its full accessibility tree only for a fixed list of bundle ids. Orca's own builds were missing, so they got only the basic tree the role read turns on, where AXPress on web content does nothing. Gecko apps such as Zotero and Firefox were missing too and showed only the window frame. Add Orca's bundle ids and any app that ships Gecko's Contents/MacOS/XUL. Stop asking on every snapshot: each request restarted Chromium's 2 s delay, so a click right after a snapshot hit a page that had not switched yet. Ask on first contact and wait out the delay, then again every 30 s, since turning VoiceOver off drops the mode. The list stays a list: the mode also turns on screen-reader mode, which VS Code and its forks react to. Fixes #23233 Fixes #24843 Co-authored-by: BeamNawapat <39422342+BeamNawapat@users.noreply.github.com> |
||
|
|
d6ca3aae25 |
fix(computer): report an unreachable Linux accessibility bus instead of crashing (#42642)
* test(computer): run the Linux bridge against an unreachable accessibility bus * fix(computer): report an unreachable AT-SPI bus instead of aborting the Linux bridge * test(computer): read the bridge response past the registry banner * test(computer): keep the healthy-bus check off the desktop's X display |
||
|
|
c53ba40801 |
fix(computer-use): put the pointer back after macOS --restore-window clicks, scrolls and drags (#42643)
* fix(computer-use): put the macOS pointer back after --restore-window clicks, scrolls and drags HID gestures left the user's pointer at the target. Save it first, and after a 200 ms settle warp it back unless it no longer sits where the last posted event left it, which means the user moved it. The HID idle counter can't tell: posting to the HID tap resets it too. * fix(computer-use): keep the pointer where the user moved it during a macOS HID gesture * fix(computer-use): save the macOS pointer right before the first HID event |
||
|
|
502d2bddb0 |
test: remove 58 low-value suites (#42603)
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
7684d51d9f |
test: remove 54 low-value suites (#42600)
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
a103d4af08 |
test: remove 50 low-value suites (#42593)
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
e5905509e4 | fix(notifications): respect Windows Focus Assist for custom audio (#42539) | ||
|
|
8301505d2d |
Keep original chat journals until a later update (#42479)
* Keep original native chat journals until a later update * Preserve empty legacy chat epochs on the reader worker * Remove unused legacy retirement remnants |
||
|
|
e173296f14 |
test: remove Linux screenshot refusal-only classes (#42405)
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
88a266e003 |
Make SQLite the source of truth for native chats (#39256)
* Retire projection adapter tests and use real SQL journal fixtures
* Persist bounded history detail and conversation outline projections
* Select rewind cuts by provider indexes and delete uniform rows as a set
* Cache bounded conversation outlines after reader projection
* Remove retired journal fold and message adapters
* Add the 4a main-vs-candidate native chat comparison harness
One entry (native-chat-4a-comparison.ts) seeds the 55k-history/310-chat
profile once through the seed arm, clones it per arm at a fixed path,
launches each built checkout in the background with an isolated home,
and drives the same stub provider events through the host RPC surface
both builds serve. Envelopes come from each arm's own source bundle.
Rows cover send/queue, 1/8/16 MiB streams with a competing send,
maintenance windows, open/reopen, history/detail, outline/fork,
rewind, record operations, R3 retained-history admission and, reported
separately, first-launch conversion (optional 20 GiB source). Output is
one JSON and one markdown table keyed by both head SHAs.
* Fail the 4a harness launch fast on a non-e2e build and stop the app it launched
* Shim tsx's __name in 4a harness pages and install the probe inside the guarded launch
* Read the native chat working flag from the 4a harness's visible root itself
* Retry transient main-process evaluates in the 4a harness host-work probe
* Use the history page's itemId in the 4a harness and name what a visible-state wait timed out on
* Activate 4a harness chat tabs through their tab group and seed enough smoke filler chats for the unopened-chat row
* WIP: migrate native chat tests to SQL journal readers
* Treat refusals carried inside a successful 4a harness RPC result as refusals
* Give the 4a stub's user echo the uuid its transcript row records
* Remove retired journal writer test machinery
* Chain the 4a stub's resumed transcript to the last message, past a fork copy's title row
* Fork the 4a harness's large chat through its latest response, not its latest prompt
* Port persisted chat tests to SQL fixtures (red WIP)
* Add the 4a main-vs-candidate native chat comparison harness
One entry (native-chat-4a-comparison.ts) seeds the 55k-history/310-chat
profile once through the seed arm, clones it per arm at a fixed path,
launches each built checkout in the background with an isolated home,
and drives the same stub provider events through the host RPC surface
both builds serve. Envelopes come from each arm's own source bundle.
Rows cover send/queue, 1/8/16 MiB streams with a competing send,
maintenance windows, open/reopen, history/detail, outline/fork,
rewind, record operations, R3 retained-history admission and, reported
separately, first-launch conversion (optional 20 GiB source). Output is
one JSON and one markdown table keyed by both head SHAs.
* Fail the 4a harness launch fast on a non-e2e build and stop the app it launched
* Shim tsx's __name in 4a harness pages and install the probe inside the guarded launch
* Read the native chat working flag from the 4a harness's visible root itself
* Retry transient main-process evaluates in the 4a harness host-work probe
* Use the history page's itemId in the 4a harness and name what a visible-state wait timed out on
* Activate 4a harness chat tabs through their tab group and seed enough smoke filler chats for the unopened-chat row
* Treat refusals carried inside a successful 4a harness RPC result as refusals
* Give the 4a stub's user echo the uuid its transcript row records
* Chain the 4a stub's resumed transcript to the last message, past a fork copy's title row
* Fork the 4a harness's large chat through its latest response, not its latest prompt
* Port ACP tests to SQL journal named reads
* Port Claude tests to SQL journal named reads
* Port Codex tests to SQL journals and drop the legacy row rollback test
* Port failure and persisted-read fixtures to the active chat database
* Remove obsolete internal-work counting from chat inventory tests
* test(native-chat): repair SQL epoch fixtures and sender closure case
* fix(native-chat): own async option preparation and port SQL runtime fixtures
* test(native-chat): encode the full founding identity
* fix(native-chat): recover saved live work on SQL open
* fix(native-chat): select retained image sources before history paging
* fix(native-chat): preserve catch-up authority and scoped journal teardown
* Cover the 4a conversion, history-shape and maintenance-debt cases through outside observation
Conversion: seed a medium chat (310 chats on the full profile) and refuse a
cached seed from another recipe. Observe the candidate's existing
epoch-installed ready boundary with a read-only poll of journal_chat_state.
An idle seed copy yields launch to every chat ready and the per-size
ready-mark intervals; the families copy verifies an unconverted first open and
waits for every chat ready before any steady-state family.
History/detail: the stub emits ordinary Claude subagent frames (interleaved
and sparse-root corpora) and a shapes turn (huge completed tool originals, one
above the detail reply cap, an oversized ordinary message). Each named shape is
verified through the real history/detail route before its samples count; diff
items stay explicit missing evidence because the Claude adapter writes none.
Maintenance: write and rewind large completed rows on a disposable seeded chat
under a held outside reader, then measure other-chat sends during the hold,
after release and while freed pages are reclaimed, from read-only page/WAL
state. Retired-epoch pruning is skipped unless such rows are observed.
* fix(native-chat): preserve stopped result disclosure and async fixture ownership
* Resolve lost-session warning identity at the journal writer
* Cover the 4a conversion, history-shape and maintenance-debt cases through outside observation
Conversion: seed a medium chat (310 chats on the full profile) and refuse a
cached seed from another recipe. Observe the candidate's existing
epoch-installed ready boundary with a read-only poll of journal_chat_state.
An idle seed copy yields launch to every chat ready and the per-size
ready-mark intervals; the families copy verifies an unconverted first open and
waits for every chat ready before any steady-state family.
History/detail: the stub emits ordinary Claude subagent frames (interleaved
and sparse-root corpora) and a shapes turn (huge completed tool originals, one
above the detail reply cap, an oversized ordinary message). Each named shape is
verified through the real history/detail route before its samples count; diff
items stay explicit missing evidence because the Claude adapter writes none.
Maintenance: write and rewind large completed rows on a disposable seeded chat
under a held outside reader, then measure other-chat sends during the hold,
after release and while freed pages are reclaimed, from read-only page/WAL
state. Retired-epoch pruning is skipped unless such rows are observed.
* Remove the retired ACP warning-reader fixture field
* Validate warning writer authority before duplicate selection
* Wake queued delivery when pending requests change
* Validate authority before all resolved item selection
* Use current chat test fixtures and name the handover journal owner
* Keep warning authority validation on the semantic selection branch
* test: lose unsaved ACP creations before a successful resume
* refactor: tighten SQL journal types and publication reads
* test: use SQL chat writers and async journal observations
* Replace casts and empty type imports in the 4a harness scripts with annotations and guards
Renderer times and tab ids from waitForFunction handles go through typeof
guards, the visibility probe is a declared Window global, the runtime call
keeps its own typed response, and cached manifests, Electron paths, profiles
and families are narrowed by guards. The empty `import type {}` blocks name
AppStore and use it. Three casts remain, each with a line-specific SAFETY:
the generic RPC result and Node's performance inside main-process callbacks.
* refactor: type handover owners and finite recovery snapshots
* Name the 4a arm's bundled envelope builders by their tracked source files
The protocol bundle imported from './src/shared/agent-session/...', a folder
the path-literal ratchet reads as an unresolved import. Each module is now a
tracked-file constant joined onto the arm's own checkout, the same form
runtime-file-watcher-resource-probe.mjs uses; the bundle and its exports are
unchanged.
* Replace casts and empty type imports in the 4a harness scripts with annotations and guards
Renderer times and tab ids from waitForFunction handles go through typeof
guards, the visibility probe is a declared Window global, the runtime call
keeps its own typed response, and cached manifests, Electron paths, profiles
and families are narrowed by guards. The empty `import type {}` blocks name
AppStore and use it. Three casts remain, each with a line-specific SAFETY:
the generic RPC result and Node's performance inside main-process callbacks.
* Name the 4a arm's bundled envelope builders by their tracked source files
The protocol bundle imported from './src/shared/agent-session/...', a folder
the path-literal ratchet reads as an unresolved import. Each module is now a
tracked-file constant joined onto the arm's own checkout, the same form
runtime-file-watcher-resource-probe.mjs uses; the bundle and its exports are
unchanged.
* Port cross-version wire tests to SQL journal reads
Hosts take the shared domain database, reads go through named async
journal reads and catch-up history pages, and released-build fixtures
type the release's own synchronous journal API.
* Align cross-version downgrade tests with one-way conversion
A downgraded build sees the chat and records as converted and its
later writes are not carried forward. Drop tests whose subject was
this build's legacy row and outcome writes; seed newer legacy rows
before the first conversion.
* Port cross-version wire tests to SQL journal reads
Hosts take the shared domain database, reads go through named async
journal reads and catch-up history pages, and released-build fixtures
type the release's own synchronous journal API.
* Align cross-version downgrade tests with one-way conversion
A downgraded build sees the chat and records as converted and its
later writes are not carried forward. Drop tests whose subject was
this build's legacy row and outcome writes; seed newer legacy rows
before the first conversion.
* test: open runtime fixtures with SQL journal workers
* test: combine runtime fixture imports
* test: prune resolved compatibility timeout finding
* test: use bundled SQL workers for runtime admission fixtures
* test: await SQL reader delivery and complete real rewind cuts
* test: include the medium chat in comparison profile metadata
* test: cover snapshot holds after replacement rollback and read failure
* test: keep clear history assertions on saved and rendered results
* Inject the deletion scenarios' record removal at the SQL record store's reads
The SQL record store has no transact or record-deletion command, so the
authoritative-deletion cases make the store's public record reads answer as
for a missing record and restore them before teardown. Assertions unchanged.
* Align cross-version fixtures with async SQL publication
Wait for frames that follow committed reads, await the async
completion feed, route the outline and item calls to their current
host methods, and expect the restart unreconciled latch on loaded
records.
* test: delete automation records with their command receipts
* Seed completed diff originals through the baseline writer and measure them
The 4a stub speaks Claude stream-json, whose adapter writes no diff items, so
completed diff originals were a skipped row. The seed now appends known
diffs once into its last filler chat through the seed arm's own original
journal writer, after that arm stops and before the seed is renamed or
copied, so every arm starts from the same rows and item IDs. Both arms verify
kind, original length, a short page head and the exact patch from detail
before timing the tail page and the detail read, and the conversion sweep
reports the diff chat apart from the small chats.
* Pin each 4a arm to a full SHA and prove its build came from it
Each --arm now names name=<checkout>@<full sha>. The harness refuses an arm whose HEAD
differs, whose product sources are dirty, whose renderer build-commit stamp is not that
HEAD, or whose sources changed after the build, and records that provenance in every
result. A cached seed from another seed-arm commit is refused, and report-only refuses to
mix results from different pins, seeds or recipes. Tables name full SHAs and disclose that
writer BEGIN-to-COMMIT spans are not observed.
* Establish restart recovery debt in the record writer
* Add active-chat resident memory comparison to 4a harness
* Commit complete provider dispatches and preserve Stop sources
* Coalesce pending publications across provider dispatches
* Capture CI watch actions at SQL writer publication
* Restore Claude task linkage in SQL acquisition facts
* Lift restarted queue holds through SQL admission
* Forward brief clear through Claude close
* Isolate SQL test worker lifetimes and prove fork durability
* Preserve saved queue order and recover restart custody on SQL open
* Keep saved queue contracts explicit and type fixtures accurately
* Run startup chat workspace conversion through its SQL owner
* Make journal command classifications exhaustive and remove unchecked casts
* Temporarily reverse quality cleanup to isolate translation regression
* Restore verified behavior-neutral journal quality cleanup
* Initialize real saved snapshot routes in Claude translation tests
* Repair native chat campaign readiness and contention measurements
* Measure fork contention under isolated retired history pressure
* Verify contention still targets a live stream at dispatch
* Run chat maintenance within the record-store lifetime
* Bound chat activity reads to work transitions
* test: adopt saved snapshots in direct Claude sinks
* Cover maintenance activity during registered starts
* test: initialize saved facts for Claude permission and resume fixtures
* test: await durable Claude resume and chain publications
* Complete provider dispatch identities in maintenance fixtures
* Preserve journal body admission and text close finality
* Scope optional permission backfill failure to its record change
* test: seed SQL reader fixtures through their owning contract
* test: assert saved Codex outcomes through SQL readers
* test: seed replay-budget withdrawals through command turns
* fix: refuse chat access on newer journal schemas
* test: migrate Claude echo wait; fails on known product finding
The retained Working assertion still fails after awaiting initial publication and accepted echo. The product finding is handed to the author.
* test: await SQL transcript and command publications
* test: migrate public rewind cursors and teardown warning scopes
* test: await child-work and status re-entry publications
* test: bind operation fixtures to SQL current-work reads
* test: establish recovery fixtures through owner restart reconciliation
* Keep child work owned across null status publications
* test: await subscriber and late-proof SQL observations
* test: resolve the profile transfer fixture from the repository root
* test: assert restart recovery capsules instead of internal call counts
* test: migrate Claude echo wait; fails on known product finding
The retained Working assertion still fails after awaiting initial publication and accepted echo. The product finding is handed to the author.
* test: await runtime session publications and final writer acknowledgement
* test: migrate coordinator mail SQL APIs; known product failures remain
* test: persist fork sources and use SQL RPC fixture contracts
* test: observe conversation publications in worker abandonment
* test: admit downgrade fixture owners through the SQL store
* test: await ordered attention prompt relay observations
* Keep status observer failures outside provider ingress
* Order provider facts before late delivery receipts
* test: submit replay fixture turns through the general writer command
* test: await durable worker delivery and lifecycle outcomes
* test: follow current provider ownership in coordinator mail
* fix(chat): settle mail custody from writer outcomes
* fix(perf): format retired fixture row count explicitly
* fix(chat): admit automatic worker close from current work
* Require live tool contention witness at send dispatch
* Await attention fixture updates and prune the fixed echo wait baseline
* Preserve the Pi startup exit verdict after recording process identity
* Check restored image deletion through the transcript view
* Adapt history pressure fixture and probes to current item storage
* Repair CI import paths and asynchronous lint findings
* Keep journal storage and projections acyclic
* Preserve proven startup facts across acquisition settlement
* Wait for the owning chat boundary in recovery and delivery tests
* Observe admitted delivery work in native chat fixtures
* Make native chat comparison measurements fail closed
* Isolate SQL worker fixtures across test files
* fix(sqlite): avoid retaining large bindings in cached statements
* perf(native-chat): reuse the captured queue publication
* test(orchestration): await refused mail settlement before recovery
* fix(orchestration): reconcile owned mail before consuming checks
* fix(orchestration): preserve each mail check response
* test(orchestration): diagnose surviving timers through public SQL reads
* perf(native-chat): place item metadata before large forms
* test(native-chat): require the octet-length column flag
* fix(native-chat): version persisted item presentations
* fix(native-chat): bound queued-image history admission
* test(orchestration): update shared mail source support
* fix(native-chat): preserve Stop transitions and admitted forms
* perf(native-chat): select the existing latest-turn index
* perf(native-chat): reuse history admission facts
* perf(native-chat): reuse shortened history wire forms
* perf(native-chat): batch bounded fork metadata reads
* fix(native-chat): synchronize CI fixtures and type mailbox receipts
* perf(native-chat): bound fork scratch presentation work
* test(native-chat): await subscriber catch-up completion
* fix(native-chat): retire readers after fork setup failure
* perf(native-chat): batch bounded fork metadata and output reads
* Reuse validated journal publication inputs on demand
* fix(chat): cancel reader requests on requested shutdown
* perf(native-chat): reuse narrow SQLite work reads
Read canonical current work and queue eligibility in one snapshot, and reuse validated work through the existing journal observation entry. Preserve full status publication, live authority validation, optional queue failures and final writer claims.
Frozen700 matches the reviewed 27-file source manifest and measured tree
|
||
|
|
856c91bba2 |
test: remove low-value native, extension, and editor suites (#42333)
* test: remove native policy and wrapper test suites * test: remove low-value extension and editor fixture suites --------- Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
120e55a818 |
test: remove low-value removal and lifecycle examples (#42329)
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
a92db9d4d0 |
test: remove low-value native, cache, and trust examples (#42321)
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
aaa61af50a |
fix(computer): report a locked screen instead of missing Accessibility (#42277)
* fix(computer): report a locked screen instead of missing Accessibility Behind the lock screen (display sleep, screen saver or a manual lock) macOS reports Accessibility as untrusted. Orca read that as a missing permission, and its recovery hint led agents to open System Settings. Check the session lock first and return screen_locked; permission setup refuses while locked. * fix(computer): recheck the lock under the provider lock and at setup launch * fix(computer): read the lock fresh for setup and keep locked status reads out of the cache * refactor(computer): drop the extra lock read at setup; the status and setup helper already check it |
||
|
|
d06458173d |
test: remove low-value relay, workspace and UI suites (#42288)
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
8ee811174f |
test: remove mock-only suites and native text-entry harness (#42283)
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
4d005cf8b8 | fix(computer): read wait --enabled state from the host, not label text (#42260) | ||
|
|
a406364654 |
fix(computer): click, set and scroll Windows buttons, text boxes, combo boxes and lists without bringing the window forward (#42237)
* fix(computer): press, set and scroll Windows native controls without activating the window Measured on Windows CI: UI Automation's Invoke, Toggle, ValuePattern and ScrollPattern focus a background WinForms control first, which activates its window. Native controls now go through their own messages instead, each owner-checked: BM_CLICK for buttons, check boxes and radio buttons, select-all plus EM_REPLACESEL for set-value on edits, CB_SETCURSEL with CBN_SELCHANGE for combo boxes (exact-text match, read back after the app's handler runs) and LB_SETTOPINDEX for list box pages. * fix(computer): refuse read-only Windows documents before typing keys, document the caret fallback * fix(computer): keep Windows native routes to measured controls and refuse partial writes BM_CLICK only for WinForms buttons, multi-column lists stay on ScrollPattern, a list already at its end reports success, edits refuse text past their limit, the combo's owner is checked right before the selection, a cleared selection reads as a mismatch and a disabled control is element_not_enabled. * fix(computer): hold Windows edit limits without a readback, require a complete combo search * fix(computer): read a long Windows edit's length from the field for its limit check * test(computer): share the Windows fixture action helpers |
||
|
|
f49eb51f95 |
feat(computer): launch-app starts desktop apps without taking the foreground (#42197)
* feat(computer): add launch-app to start apps without taking the foreground * fix(computer): leave running apps alone and keep the launch env clean * fix(computer): recheck aliases, keep focused launches visible, refine launch hints * chore(computer): document the launch result casts * test(computer): check launch-app reports where the app ended up on Windows CI * docs(computer): point app_not_found at launch-app * test(computer): format the merged Windows background e2e * fix(computer): match launch paths to the running file and word background launch as an attempt |
||
|
|
e2fb873c0a |
fix(computer): type into Windows text boxes without bringing the window forward (#42165)
* fix(computer): type into Windows text fields without raising the window Adds type-text --element-index on Windows through EM_REPLACESEL for user32, RichEdit and WinForms edits (owner-checked, read back with WM_GETTEXT), or a ValuePattern write merged with the TextPattern selection. Keyboard verbs on a background window now refuse with background_unavailable before sending anything and point at type-text --element-index. Foreground typing uses one SendInput KEYEVENTF_UNICODE batch instead of WM_CHAR posted to the top-level window, and get-app-state shows multiline edit text again. * fix(computer): refuse typing into Windows list items, keep the caret on a confirmed prefix |
||
|
|
62b22aeab0 |
fix(computer-use): read minimized, hidden and other-Space macOS windows without --restore-window (#42182)
* fix(computer-use): observe minimized, hidden and other-Space windows on macOS without a restore macOS only enumerated on-screen windows and dropped minimized AX windows, so an app whose window was minimized, hidden or on another Space (any full-screen app) could not be observed, and agents fell back to --restore-window, which activated every window of the app and switched the user's Space. Enumerate every window, keep the off-screen ones accessibility can still read (other-Space windows by remote token, closed windows dropped by their Space membership), and report their real minimized and off-screen state. Background pointer and key input on such a window refuses before sending anything. --restore-window now raises only the target window, once per request, and delivery is measured from whether the target became frontmost. * fix(computer-use): pin the restored window and prefer visible windows by default * fix(computer-use): keep window-index cache keys and stale-element checks across a restore * fix(computer-use): restore only this app's resolved window and refuse a closed target before raising * docs(computer-use): say macOS reads minimized, hidden and other-Space windows |
||
|
|
cc1e3740da |
feat(computer-use): scroll and select list rows in background Linux windows, and aim GTK 4 clicks (#42201)
* feat(computer-use): scroll and select rows in background Linux windows, and aim GTK 4 pointer input Scroll moves the nearest scroll bar's value by whole viewports, handing the scroll outward at a view's end, or pages a web document with Component.ScrollTo, before falling back to the pointer. A click on a list row with no action selects it through the list's Selection when the index is proven to name that row. GTK 4 rectangles come from WINDOW extents plus the X11 client origin, since its SCREEN extents read [0, 0, w, h]. * fix(computer-use): keep Linux background scroll and select honest under partial and slow results Wait for ScrollTo to move before planning the next page, report pages already scrolled instead of falling back, let refused scroll bars fall through, clear a multi-select list before a plain-click select, accept only an InvalidArgs refusal as index proof, bound child walks, and read GTK 4 origins lazily. * fix(computer-use): scroll the innermost Linux view that overflows, and bound the scroll search ScrollTo now runs on each ancestor that overflows its own box, innermost first, so an inner scroller moves before the page; coordinate scrolls start from the deepest node under the point. Cap the scroll bar search to small containers, report a document at its end as a no-op, keep looking past an unusable spanning child, and look up GTK 4 window names once per operation. |
||
|
|
513aef5e7d |
fix(computer): ask the user before retrying with --restore-window (#42181)
* fix(computer): ask the user before retrying with --restore-window Error hints, recovery steps, the skill guide and the CLI follow-up command told agents to retry with --restore-window on their own, which raises the target app and takes the user's screen. They now say to ask the user first, stop recommending a restore just to observe, and fix stale per-platform background claims. * docs(computer): name the Linux keyboard gate and the Wayland limit |
||
|
|
1f56231043 |
fix(computer-use): keep Linux actions in the background and refuse ones that need the foreground (#42157)
* fix(computer-use): keep Linux actions in the background unless the user allows a raise Pointer fallbacks, scroll and drag no longer raise the target window and warp the pointer without --restore-window; they refuse with background_unavailable before sending anything. With the flag, the target point is validated before the raise, a window on another workspace is refused instead of switching the user's view, and the pointer is put back afterwards. Also refuse background clicks that open menus, keyboard-grabbing popups or dialogs (ported from agent-desktop's foreground classifier), refuse synthetic input on Wayland where it never lands, press Return through a GTK entry's activate action, re-check focus and held modifiers while typing, and report action.delivery from what happened. * fix(computer-use): check the workspace before every Linux raise and release held buttons Run the other-workspace check before GrabFocus too, raise before a user-approved menu or dialog opener, validate a pointer target before asking for the flag, count newly active app windows as foreground, release synthetic buttons when input is interrupted, and stop typing without an exception exit so the user's held keys are left alone. * fix(computer-use): keep held modifiers from other X screens and settle remapped keys on early stop * fix(computer-use): refuse a Linux raise when the target's workspace can't be read |
||
|
|
57c9991c98 |
fix(computer-use): refuse macOS background clicks that open a menu or dialog, and stop refusing plain AX presses (#42158)
* fix(computer-use): refuse macOS background actions that open a menu or dialog over the user * fix(computer-use): let web comboboxes click in the background and stop probing a hung app * fix(computer-use): refuse background drags that start on a menu owner * docs(computer-use): note background drags on menu owners are refused * fix(computer-use): classify the element under a drag's start and check AXConfirm/AXOpen secondary actions |
||
|
|
10302c46a8 |
fix(computer-use): stop Windows clicks, scrolls and drags from taking the screen without --restore-window (#42161)
* fix(computer-use): refuse Windows pointer fallbacks without --restore-window Coordinate, right, middle, modified and multi-clicks, clicks on elements with no primary pattern, unscrollable wheel scrolls and drags raised the window and moved the real cursor without being asked. They now refuse with background_unavailable before anything is sent. Coordinate and fractional scrolls go through ScrollPattern first. With --restore-window only the target window is raised (never a cloaked one), with a UIA activation fallback and a second attempt, points are validated first and the cursor is put back behind the input. A failing UIA pattern or disabled element is an error instead of a real click, and delivery is measured after the post-action observe. * fix(computer-use): settle Windows clicks before restoring the cursor and keep scroll writes honest * fix(computer-use): restore the Windows cursor with SetCursorPos and hit-test coordinate scrolls * test(computer-use): assert Windows raise outcomes, not attempt order * fix(computer-use): report a Windows page scroll the provider ignores as not scrolled |
||
|
|
a36d1ddafd |
test: remove low-value bridge and native checks (#42105)
Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
c6b9ac9549 |
test: remove duplicate native policy checks (#42011)
* test: remove duplicate native policy checks * test: trim repeated clipboard reads --------- Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
2f9b122201 |
test: remove native smoke and duplicate input checks (#42007)
* test: remove repeated focused Linux text insertion * test: remove repeated foreground TextEdit replacement * test: remove Notepad smoke and duplicate paste coverage * test: drop repeated numeric row and scroll-work counters --------- Co-authored-by: OrcaWin <293788423+OrcaWin@users.noreply.github.com> |
||
|
|
c5657bd5a7 |
feat(computer): type into a named text field without focusing it on Linux and macOS (#41421)
* feat(computer): type into a named text field without focusing it on Linux and macOS * fix(computer): report unconfirmed Linux text readback and cover element typing * test(computer): send an element with benchmark text only for targeted typing * test(computer): pick gedit's editor, not its hidden overview map, in Linux e2e * docs(computer): confirm focus before the unindexed type-text fallback * fix(computer): report unreadable element text readback as readback_unsupported |
||
|
|
c794f75e4e |
fix(computer-use): use UI Automation for Windows scroll, expand and slider actions (#41429)
* fix(computer-use): scroll, expand and set sliders on Windows without raising the window * fix(computer-use): wait for Windows scroll and slider writes to settle before verifying * fix(computer-use): reject NaN for Windows slider set-value * test(computer-use): wait on the Windows fixture state with vi.waitFor and a schema * test(computer-use): check activation per action against a holder window in the Windows fixture * test(computer-use): log fixture window events for activation failures * fix(computer-use): report foreground when a Windows app raises itself for a UI Automation action WPF activates its own window for Scroll, ExpandCollapse, RangeValue and SelectionItem calls from any UIA client, so the runtime now compares the foreground window before and after a pattern action instead of assuming background. The Windows e2e checks delivery against the fixture's own activation count, with a holder window in another process. * test(computer-use): skip an unstarted holder when closing the Windows fixture * fix(computer-use): bound Windows pattern scroll paging and tighten slider read-back * fix(computer-use): cap Windows slider read-back tolerance and run the UIA e2e only on CI |