Prefer exact linked PR lookup when safe, add stale-while-revalidate for sidebar hosted review metadata, and preserve branch discovery correctness for stale linked PR metadata.
* fix: address pr-bug-scan findings from #1444
Threaded linkedPRNumber through all ChecksPanel force-refresh paths; gated useComposerState's PR-URL-from-name recovery on selected repo slug match; wrapped main-process linkedPR fallback gh pr view i
* fix: satisfy react-hooks/exhaustive-deps in selectedRepoSlug effect
Extract selectedRepo?.path into a const so the effect's only dep is
the path string, removing the missing-dependency warning.
---------
Co-authored-by: orca-bug-scan-bot <orca-bug-scan-bot@stably.ai>
Co-authored-by: Neil <neil@nousresearch.com>
When a worktree is created from a PR via the source picker, the new
local branch differs from the PR's head ref, so the branch-keyed PR
lookup misses and the worktree card shows no PR strip. Pass the
worktree's linkedPR number through the IPC call and fall back to a
number-based lookup in the main process. Also recover linkedPR from
a PR URL pasted into the workspace name when the user skips the
source picker. PR strip now wraps the whole row as the PR link.
Co-authored-by: Orca <help@stably.ai>
* Fix new workspace composer focus restore
* Unify new workspace source selection
* WIP: selected source pill in smart workspace name field
Co-authored-by: Orca <help@stably.ai>
* fix(new-workspace): truncate source pill so it doesn't expand the dialog
Co-authored-by: Orca <help@stably.ai>
* feat(new-workspace): add open-in-browser button to source pill, fix vertical alignment
Co-authored-by: Orca <help@stably.ai>
* refactor(new-workspace): drop redundant kind suffix, distinct PR/issue icons, tooltips on pill actions
Co-authored-by: Orca <help@stably.ai>
* fix(new-workspace): type linked URL into agent input without auto-submit
Co-authored-by: Orca <help@stably.ai>
* fix(new-workspace): use bracketed-paste for draft URL injection so it actually appears in the agent input
Co-authored-by: Orca <help@stably.ai>
* feat(agents): per-agent draft injection strategy (codex slow paste, pi/opencode type-chars)
Co-authored-by: Orca <help@stably.ai>
* fix(agents): smarter TUI-ready heuristic + bracketed paste for codex/pi/opencode
Replaces per-agent strategy guesswork with a measured readiness check:
title-idle / non-shell-foreground stable for 1.5s / 2.5s minimum floor.
Verified against codex, pi, opencode, claude in a node-pty + xterm-headless
test rig — bracketed paste lands in the input buffer for all four.
Co-authored-by: Orca <help@stably.ai>
* refactor(agents): drop unused per-agent draft strategy abstraction
The TUI-ready heuristic in agent-paste-draft.ts works for every tested
agent (claude/codex/pi/opencode), so the AgentDraftInjectionStrategy
field, type-chars + bracketed-paste-slow code paths, and per-agent
overrides are dead. Keep the `agent` arg on pasteDraftWhenAgentReady
for future per-agent escape hatches without touching every call site.
Co-authored-by: Orca <help@stably.ai>
* feat(agents): skip draft URL injection for copilot + cursor-agent
Both TUIs open with a 'Do you trust this folder?' menu on first launch
that consumes keystrokes as menu input — pasting a URL there either
selects an arbitrary option or quits the session. Mark them with
skipDraftUrlInjection so the workspace still opens cleanly; the user
types/pastes the URL themselves once past the trust menu.
Co-authored-by: Orca <help@stably.ai>
* feat(agents): native --prefill for claude, trust pre-write for cursor/copilot
Replaces the empirical TUI-ready waits with two deterministic mechanisms:
1) `claude --prefill <text>` flag — Claude launches with the URL already in
its input box, no submit. Eliminates the readiness/paste race entirely
for the most common agent.
2) DECSET 2004 (`\x1b[?2004h`) detection on the PTY data stream for every
other agent. That escape is the protocol-level "input layer ready,
accepting bracketed paste" handshake — emitted by claude/codex/pi/
opencode/gemini/cursor-agent/copilot the moment the input box mounts.
We tap it via a sidecar subscription on pty-dispatcher (no interference
with the primary xterm handler) and paste as soon as it lands. The
8s budget is now an upper bound, not a target.
Cursor-agent and Copilot's "Do you trust this folder?" menus are bypassed
by writing the same trust artifacts the CLIs themselves write after the
user accepts:
- Cursor: `~/.cursor/projects/<slug>/.workspace-trusted` (slug = abs path
with leading `/` stripped, remaining `/` → `-`).
- Copilot: append cwd to `trustedFolders` in `~/.copilot/config.json`
(the same array the bundled `addTrustedFolder` writes).
Verified against the cursor-agent CLI bundle (versions/2026.04.17-787b533/
index.js: `_=".workspace-trusted"`) and the @github/copilot 1.0.32 bundle
(`isFolderTrusted` / `addTrustedFolder` both read/write `trustedFolders`).
Both check via realpath() before string-comparing, so the trust preset
canonicalizes too.
skipDraftUrlInjection is dropped — both agents now get the draft URL
paste once the trust menu is pre-resolved.
Tests: 24 passing across tui-agent-startup, agent-trust-presets,
pty-dispatcher routing.
Co-authored-by: Orca <help@stably.ai>
* fix(agents): wait for post-?2004h render burst to settle before paste
OpenCode emits DECSET 2004 at ~500ms during alt-screen setup, then runs
a 1.3s splash render with NO bytes on the PTY, then paints the actual
input box at ~1.85s. Pasting on the bare ?2004h signal lands during the
silent gap and the bytes are dropped.
The fix: take ?2004h as the necessary precondition, then wait for the
TUI's render burst to finish — defined as 1500ms of stream silence
after the most recent post-?2004h byte. This captures both the fast
TUIs (claude/pi/codex emit setup escapes in one burst then go quiet)
and the slow ones (opencode emits, sleeps for the splash, emits again,
then goes quiet).
Verified against opencode/claude/pi in a node-pty rig: paste lands on
the first try with the new strategy. The hard 8s timeout still caps
the wait when an agent fails to launch.
Co-authored-by: Orca <help@stably.ai>
* fix(agents): guard agentTrust IPC so stale preload doesn't crash launch
If the preload bundle is older than the renderer (a real situation in
electron-vite dev because preload changes only apply on full restart,
not HMR), `window.api.agentTrust` is undefined and the launch crashes
with "Cannot read properties of undefined (reading 'markTrusted')"
before the worktree even opens.
Guard the call sites in launch-work-item-direct and useComposerState
to skip the trust pre-write when the IPC isn't exposed, and wrap the
invoke in try/catch so an IPC error never blocks the launch — the user
just sees the trust menu and accepts it manually, same as before this
feature shipped.
Co-authored-by: Orca <help@stably.ai>
* feat(tasks): route 'Use' through the New Workspace dialog instead of yolo-create
The Use CTA on the Tasks page used to create+activate a worktree
synchronously, which surprised users — the worktree appeared in the
sidebar before they had a chance to confirm name / agent / setup. The
unified New Workspace dialog landed in this branch already supports
opening with a linked work item pre-filled (see openComposerForItem /
openComposerForLinearItem), so just route Use through it.
The launchWorkItemDirect helper stays exported for ProjectViewWrapper,
which has its own UX where the immediate-create flow is the right call.
Co-authored-by: Orca <help@stably.ai>
* test(agents): include `agent` field in autohand startup-plan assertion
Merging main brought in the Autohand Code agent test (PR #1382), which
predated this branch's addition of `agent` to AgentStartupPlan.
Aligning the assertion fixes the lone CI test failure on this PR.
Co-authored-by: Orca <help@stably.ai>
* refactor(agents): drop unused expectedProcess arg + snapshot sidecar set
Two minor follow-ups from self-review:
1. `pasteDraftWhenAgentReady` no longer reads `expectedProcess` — readiness
is gated on DECSET 2004 alone now, not on PTY foreground process. Drop
it from the signature and from the two callers (launch-work-item-direct,
new-workspace).
2. The pty-dispatcher's sidecar fan-out iterates the live Set, which is
safe against deleting the current element but not against a watcher
that synchronously subscribes a sibling. Snapshot via Array.from
before the loop. Cheap (Set is tiny) and removes the latent footgun.
No behavior change.
Co-authored-by: Orca <help@stably.ai>
* test(e2e): match the unified smart-name input's new placeholder
The CreateFromTab refactor in this branch replaced the separate "Workspace
name" Input with a single SmartWorkspaceNameField whose default-mode
placeholder is "Type a name, #1234, branch, GitHub or Linear URL". The
worktree-create e2e test was still anchoring on the old "Workspace name"
text and could not find the input.
Update the placeholder regex to match the new copy. Free-form text typed
into smart mode is treated as a workspace name by submitQuick — same
contract the test used before.
Verified locally: targeted e2e passes in 2.2s.
Co-authored-by: Orca <help@stably.ai>
---------
Co-authored-by: Orca <help@stably.ai>
* Render GitHub markdown safely
Co-authored-by: Orca <help@stably.ai>
* Support PR line comment ranges in diffs
Co-authored-by: Orca <help@stably.ai>
* Add GitHub item dialog review workflow
Co-authored-by: Orca <help@stably.ai>
* Make attribution tests environment-safe
Co-authored-by: Orca <help@stably.ai>
* Wire GitHubItemDialog for row clicks and bound file cache
Row clicks in the Tasks GitHub list now open the new GitHubItemDialog
instead of the legacy GitHubItemDrawer, making the PR review workflow
reachable from the main entry point. The unused drawer and its state
are removed.
Also bounds prFileContentCache at 64 entries with LRU eviction so the
module-level map can no longer grow without limit across many PR
openings in a session.
Co-authored-by: Orca <help@stably.ai>
* Convert GitHubItemDialog to a right-side drawer
Replaces the centered Dialog modal with a Sheet drawer (side=right,
max-w 960/1100/1280 across breakpoints) so the review surface slides
in from the side instead of popping over the task list.
Co-authored-by: Orca <help@stably.ai>
* Address review feedback on GitHub work-item details
- Restore full pagination in listAssignableUsers (REST /assignees --paginate)
so repos with >100 assignees no longer silently drop users from the picker.
- Drop redundant REST /users/<login> fan-out in getMentionParticipants; the
aliased GraphQL query already returns login/name/avatarUrl.
- Parallelize mention-participant lookup with checks/participants fetch.
- Avoid mutating caller-provided objects in mergeGitHubUsers.
- Type addIssueComment / addPRReviewComment(Reply) preload wrappers as
Promise<GitHubCommentResult> instead of Promise<unknown>.
- Use typeof startLine === 'number' to harden the review-comment guard.
Co-authored-by: Orca <help@stably.ai>
---------
Co-authored-by: Orca <help@stably.ai>
Co-authored-by: Jinjing <6427696+AmethystLiang@users.noreply.github.com>
* feat(sidebar): improve PR checks and comments
Co-authored-by: Orca <help@stably.ai>
* fix(sidebar): classify bot comments via GitHub user.type instead of login heuristic
Third-party review bots like qodo-ai-reviewer, coderabbitai, and sonarcloud
don't follow the [bot] suffix or "bot"/"automation" substring convention, so
the regex-based detector misclassified them as human. Plumb REST
`user.type === 'Bot'` and GraphQL `author.__typename === 'Bot'` through to the
renderer as an authoritative isBot flag; fall back to the login heuristic only
when the data source can't report it.
Co-authored-by: Orca <help@stably.ai>
* fix(sidebar): allowlist AI review services that sign in as User accounts
qodo-ai-reviewer, coderabbitai, codium-ai and similar third-party review
services register as regular GitHub user accounts, so REST `user.type` is
"User" and their logins contain no "bot"/"automation" tokens. The previous
fix relied on the GitHub-reported type, which fails for these. Add an
explicit substring allowlist of known automation services so they still
land in the Bots tab.
Co-authored-by: Orca <help@stably.ai>
---------
Co-authored-by: Orca <help@stably.ai>
Co-authored-by: Jinjing <6427696+AmethystLiang@users.noreply.github.com>
* feat: cross-repo issues view with multi-repo selection
- Add multi-repo selection via RepoMultiCombobox with persisted defaultRepoSelection setting (null = sticky-all)
- Stamp repoId on GitHubWorkItem at the renderer fetch boundary; merge items from all selected repos with per-repo failure tracking
- Extract task-query helpers (tokenize/strip/parseTaskQuery) to src/shared and add unit tests
- Refactor NewWorkspacePage row to <div role=button> to allow nested interactive elements without invalid-HTML hydration errors
- fix(repo-combobox): toggle all-repos selection on repeat click
* fix(tasks): route Use CTA to item's own repo in cross-repo view
handleUseWorkItem previously referenced a removed 'repoId' state; use item.repoId so launching from a merged cross-repo list targets the correct repo.
* test(e2e): add tasks page smoke test
Adds a + button on the tasks page that opens a dialog to create a new
GitHub issue in the selected repository. Wires createIssue through the
main/preload IPC using gh api, and refreshes the tasks list after
creation so the new issue shows up immediately.
Also clears a stray Radix pointer-events lock on GitHubItemDrawer mount
so Close/open-in-GitHub buttons remain clickable after the New Issue
dialog closes.
* New workspace page with agent catalog, composer modal, and terminal integration
* fix lint
* better blank state
* fix: resolve typecheck errors in new-workspace flow
- widen activateAndRevealWorktree's issueCommand to accept direct
command shape used by NewWorkspacePage, not just the main-process
runner-script variant
- use a ref object in pty-connection tests to dodge TS narrowing the
captured callback to never across the mock closure boundary
* fix: bound worktree name auto-suffix loop to prevent OOM in tests
The auto-suffix loop in createLocalWorktree had no termination cap.
When tests (or a misconfigured env) mocked getBranchConflictKind /
getPRForBranch to always return a collision, the loop ran forever and
the vitest worker crashed with "Ineffective mark-compacts near heap
limit".
Cap the search at 100 suffixes, and if all fail, fall back to the
original specific error messages (branch already exists / PR already
owns the name) instead of a generic failure.
Update the two tests that asserted the old throw-on-first-conflict
behavior to verify the new auto-suffix path end-to-end.
* Add richer feedback dialog
* Keep anonymous submission client-side only
* Use lowercase feedback API host
* Fallback feedback submission when api host is unavailable
The `gh api --cache 60s` flag caused stale check-run data even when
the user explicitly clicked refresh. Thread a `noCache` flag through
IPC so manual refreshes skip the gh CLI cache while polling still
benefits from it.
- Add PRConflictSummary type and derive conflict metadata (base ref,
commits behind, conflicting files) via local git merge-tree
- Fix parseUnmergedEntry to use space-separated parsing (porcelain v2
unmerged entries are not tab-separated)
- Enhance rebase detection by checking rebase-merge/ and rebase-apply/
directories in addition to REBASE_HEAD
- Show conflicting files list and commits-behind count in ChecksPanel
- Disable merge button with tooltip when PR has conflicts
- Add operation banners (merge/rebase/cherry-pick) in SourceControl
and WorktreeCard
- Reorder section list so unstaged changes appear above staged
- Extract conflict-summary, issues, and gh-utils into separate modules
- Add 10s timeout to git fetch in conflict summary derivation
- Fix test fixtures to match real porcelain v2 format
During a rebase with detached HEAD, the branch name can be empty or
resolve to empty after stripping refs/heads/. This caused gh to return
arbitrary PRs and created invalid cache keys like "path::".
Guard all branch-to-cache-key paths: getPRForBranch bails before
acquiring the semaphore, renderer components skip cache lookups,
and syncPRChecksStatus validates the normalized branch.
gh pr view only matches the current checkout, which fails for worktrees
on different branches. Switch to gh pr list --repo --head to query by
branch name directly, with a fallback to gh pr view for non-GitHub
remotes. Also fix the ChecksPanel refresh button to use the proper
handleRefresh flow and force-refresh the PR cache.
- Use gh api --cache for issues, issue lists, and PR checks endpoints so 304 Not Modified responses don't count against the rate limit
- Pass branch to getPRChecks to enable the check-runs REST endpoint with caching
- Extract mapping functions into separate mappers.ts to keep client.ts within 300-line limit
- Extract mapIssueInfo() helper to deduplicate issue mapping logic
- Remove dead re-exports from client.ts
- Fix reference instability in ChecksPanel poll dependency array
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Add comprehensive tests for worktree parsing, GitHub check status derivation,
hooks YAML parsing, persistence store, layout serialization, store cascades,
and terminal helpers. Extract pure functions from worktrees.ts into
worktree-logic.ts for better testability.
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* fix: extract path-security helpers and refactor IPC modules
- Extracted path-security helpers from filesystem.ts into filesystem-auth.ts to fix max-lines lint error (402 -> 293 lines)
- Extracted duplicated ENOENT detection into isENOENT() helper to eliminate code duplication
- Fixed missing curly braces on single-line if-return in isDescendantOrEqual (lint violation)
- Replaced any with unknown in test files to satisfy lint rules
- All tests passing (29/29)
- Lint clean (0 errors, 0 warnings)
* fix: bundle preload deps for sandbox mode and fix editor test types
The sandbox: true change in createMainWindow broke the app because
electron-vite was externalizing @electron-toolkit/preload, producing
a require() call that fails in sandboxed preload scripts. Exclude it
from externalization so it gets bundled inline.
Also fix type errors in editor.test.ts from the partial store setup.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
- Add PR checks panel to right sidebar
- Stream checks from GitHub API
- Update PR title real-time with loading state
- Add error logging and dependency fixes
- Fix linting issues (curly braces, exhaustive deps)
Show CI check status icons (success/failure/pending) next to the branch
name on worktree cards. Return 'neutral' instead of 'pending' when no
CI checks exist to avoid showing a misleading spinner. Fix lint errors
in client.ts (curly braces, array-type).
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Jinjing
Neil
Brennan BensonandOrca
Jinwoo HongandOrca
