This merge introduces canonical mock helper modules across the dashboard and engine packages, and migrates representative test suites to use them for consistency. The changes add four new mock helpers (`mockApi.ts`, `mockLucide.ts`, `mockCore.ts`, `mockCoreEngine.ts`) and harden the API mock proxy's
Fusion-Task-Id: FN-3044
- useTasks debounced effect bailed when searchQuery transitioned back to
undefined (App passes `searchQuery || undefined`), so clearing the
input never refetched the unfiltered list. Track previous value via
ref and only skip the initial mount.
- FTS5 search used bare tokens (exact term match), so "frob" did not
match "frobnicator". Append `*` to each token (and to quoted tokens)
for prefix matching. LIKE fallback already did substring matching.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This merge adds research settings to the settings modal and dashboard (FN-2839, FN-3029), implements a dashboard font scale setting with persistence and documentation (FN-3027), refactors AgentDetailView with a new AgentLogViewer CSS module (FN-2839), and improves NewAgentDialog with custom model dr
Fusion-Task-Id: FN-3027
Adds `focus-visible` styling to action buttons in the AgentLogViewer component, improving keyboard accessibility for interactive button elements.
Fusion-Task-Id: FN-3037
The vitest child-process guard wrapped exec/execFile without preserving the
`[util.promisify.custom]` symbol, so awaited `execAsync` resolved to a raw
stdout string instead of `{stdout, stderr}`. That single regression cascaded
through ~60 "failing" tests across cli, core, engine, and dashboard whose
production code was actually correct. Also relax the AI-CLI blocklist for
cheap introspection (--version/--help/which …), give SIGTERM'd subprocesses a
brief grace period before being flagged as "left running", fix a few real
test-side bugs uncovered along the way (executor mock step transitions, iOS
last-resort keyboard path, mission SSE replay tests racing with the real AI
agent), and convert dashboard route tests' dynamic `await import("../server.js")`
to static imports so first-test timings drop from 2–5s to <200ms.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Workflow-step REVISE retries, pause→todo handoffs, and the
context-overflow fresh-session requeue were all routing tasks back to
`todo` before returning to `in-progress`. The default reopen-to-todo
path reset every step to pending and rewrote PROMPT.md checkboxes, so
each retry restarted from step 0 even when earlier steps had already
been done — the symptom seen on FN-2978, where every workflow REVISE
or pause cycle wiped the task's progress.
- Add `preserveResumeState` to `TaskStore.moveTask`. When set, skip
`resetAllStepsToPending` + `resetPromptCheckboxes` and keep
`worktree` and `executionStartedAt` so the resumed run reattaches to
the same checkout. `status`, `error`, and `blockedBy` still clear.
- Use it on the workflow-rerun bounce, the three pause-graceful
handoffs, and the context-overflow requeue. The agent-terminated
pause path still discards (it nukes worktree+branch by design).
- Context-overflow requeue clears `sessionFile` synchronously in the
awaited `updateTask` immediately before `moveTask`, so the next
dispatch cannot reopen the saturated session via a stale pointer.
- `fn_task_update` no longer silently regresses `done`/`skipped` steps
to `in-progress`, no longer captures a stale rewind checkpoint when
it does, and tells the agent honestly when a regression is ignored.
- Mobile chat keyboard: ChatView/QuickChatFAB gate layout on the new
`keyboardOpen` flag so focused-input + viewport-shrink iOS cases
still adjust when the computed overlap is zero.
Tests:
- New `preserveResumeState` coverage in store.test.ts; updated
workflow-rerun + pause-graceful assertions in executor.test.ts.
- Restructured the previously-flaky "routes exhausted prompt-mode
workflow hard failures" test to drive the bounce inline; passes in
isolation and in the wider workflow/pause/context sweep (59/59).
- Added regression tests in ChatView.test.tsx and QuickChatFAB.test.tsx
for the iOS last-resort `keyboardOpen=true, keyboardOverlap=0` case.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Three coordinated fixes for the FN-2978 incident class — auto-requeues
that orphaned committed work and watchdog kills on long verification runs.
**Auto-requeue branch reuse** (executor.ts, worktree-pool.ts)
- executor.ts:1782 now uses `task.branch || fusion/<id>` so persisted
branches are honored on requeue. Previously the hardcoded fallback
always tried to re-create the original branch, hit a conflict with
the prior run's ref, and got suffix -2/-3. Other call sites already
honor task.branch — this aligns the worktree-acquisition path.
- worktree-pool.ts:181 prepareForTask now probes existing branches with
`git rev-parse --verify` and checks them out as-is. Falls through to
suffixed creation only when the branch is genuinely in use by another
live worktree. Previously force-reset with `checkout -B`, destroying
prior commits.
- New private reconcileStepsFromGitHistory walks `git log
baseCommitSha..HEAD` for `feat(FN-X): complete Step N` commits and
marks matching steps[] as done so resumes don't redo committed work.
**Manual reset endpoint + UI** (dashboard)
- POST /api/tasks/:id/reset (requires `confirm: true`) — clears worktree,
branch, all retry counters, resets steps[] to pending, moves to todo.
Distinct from /retry which is the soft-resume path.
- Reset button alongside Retry in TaskDetailModal with confirm dialog,
wired through useTasks → AppModals → API.
**fn_run_verification tool** (run-verification-tool.ts, executor.ts)
- New custom tool wrapping test/lint/build commands with a heartbeat
callback (per-line + 60s synthetic), 200KB head+tail output cap, hard
timeout with SIGTERM→SIGKILL escalation, and auto-bootstrap detection
for missing node_modules. Prevents the inactivity watchdog from
killing sessions during long compiles.
- Cross-platform via `shell: true` (Node picks /bin/sh on POSIX,
cmd.exe on Windows). Prompt section in EXECUTOR_SYSTEM_PROMPT and
EXECUTOR_PROMPT_TEXT instructs agents to prefer package-scoped
verification first and reserve workspace-scoped runs for final
integration.
**Tests** (64 passing)
- detect-pseudo-pause.test.ts (27 tests) — covers all 7 regex patterns,
structural fallback, FN-2978 regression text.
- reconcile-step-regex.test.ts (25 tests) — pins the commit-message
regex against a wide variant set.
- run-verification-command.test.ts (12 tests) — basic execution, output
capture, heartbeat callbacks, timeout, error handling. POSIX-specific
cases (multi-cmd `;`, `>&2`, `\$USER`) gated behind itPosix.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Add attachment-first chat compose flow with paperclip picker, drag-and-drop, and paste support in ChatView input
- Render pending attachment preview chips with per-file remove actions and drag-over visual affordance
- Display sent message attachments with image/file presentation and accessible focus/interaction styling
- Update ChatView tests and icon mocks to cover attachment-only sends, file selection, and preview behavior
Fusion-Task-Id: FN-2911
- Keep active streaming responses alive when resuming an existing quick chat session instead of resetting stream state
- Update useQuickChat session-resume logic to only restart waiting indicators when appropriate
- Adjust QuickChatFAB waiting indicator behavior so it reflects real in-flight assistant activity
- Add focused hook and component tests covering same-session resume and waiting-indicator regressions
Fusion-Task-Id: FN-2904
On a fresh install useAuthOnboarding's effect ran at mount before the
setup wizard's 500ms auto-open timer fired. The one-shot ref locked,
and the resolved fetch could either stack model onboarding on top of
the wizard or never re-trigger after the wizard closed.
- Gate the trigger on projectId being set so the wizard owns the
bootstrap phase; the auth check only fires once a project exists.
- Re-check setupWizardOpen via a ref when the auth fetch resolves to
avoid stacking onboarding on top of a wizard opened mid-fetch.
- Release the one-shot in that suppressed branch so the effect retries
when the wizard closes.
Adds two regression tests: fresh-install handoff and mid-fetch wizard
suppression.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Wire onRetryTask from App through Board/Column/WorktreeGroup/ListView into TaskCard
- Add a Retry button with loading/disabled state to failed task error boxes in TaskCard
- Update failed-card styles to use design tokens and add retry button visual states
- Add TaskCard tests covering retry visibility, callback invocation, loading, and error toast behavior
- Add a patch changeset for @runfusion/fusion documenting the dashboard retry shortcut
- Add POST /api/memory/dream route that runs project and agent dream processing with AI prompt execution
- Expose triggerMemoryDreams() in dashboard legacy API client for direct dream invocation
- Switch SettingsModal and useMemoryData Dream Now actions from automation lookup to the new endpoint
- Update dashboard route, hook, and settings modal tests to cover success and error handling
- Add a patch changeset for @runfusion/fusion documenting the new endpoint and client helper
- Extend useMemoryData with triggerDreamNow and dreamRunning state by locating and invoking the Memory Dreams automation
- Add Dream Now actions to MemoryView and SettingsModal memory settings with loading UI, success/error toasts, and file refresh in MemoryView
- Expand hook and component tests to cover dream-trigger visibility and execution flows in both views
- Update terminal mobile keyboard layout CSS contract expectations for the current mobile modal width and min-height rules
Production typecheck (tsconfig.json + tsconfig.app.json) was already
clean, but a third config that includes test files surfaced 661 errors
across 60+ test files — accumulated drift between mock fixtures and
production types. Six parallel typescript-pro agents fixed every one
without touching production code.
Per-scope before/after (errors → 0):
ChatView 183
Mailbox + Agent suite (5 files) 156
Task / Modal suite (6 files) 127
App + small components (12 files) 96
Hooks + api/auth (8 files) 48
Long tail (32 files) 51
-----------------------------------------------------------------
Total 661
Major fix categories:
- Untyped state objects inferring `never[]` / `null` literals (root
cause of ~120 errors in ChatView alone — added a single
`UseChatReturn` annotation)
- Mock objects missing fields that became required: `WorkflowStep.mode`,
`ChatMessage.thinkingOutput / metadata`, `ChatSession.projectId`,
`Task.log`, `ProjectHealth` fields, `PtyTerminalSessionInfo.createdAt`,
`Agent.metadata`, `InboxResponse.total`, etc.
- Mock objects with stale fields that no longer exist:
`AgentBudgetStatus.budgetPeriod`, `truncated` on log responses,
`OutboxResponse.unreadCount`, `MergeResult.source/target/details`
- Modal props that became required (e.g. `PlanningModeModal.onTasksCreated`)
- String literals not in narrowed unions (`Column`, `WorkflowStepPhase`,
`InsightStatus`, `AgentLogType`, etc.)
- `querySelector` returning `Element` cast to `HTMLElement` for
`@testing-library/react`'s `within()`
- Vitest mock typing: `.mock.calls` access needing `vi.mocked(...)`,
zero-param tuple handling, generic `vi.fn(() => [])` inferring
`never[]`
Helpers introduced in test files (no shared infra):
- `makeSettings(overrides)` in ModelSelectorTab.test.tsx
- `makePromptOverrides(overrides)` in AgentPromptsManager.test.tsx
- `FileBrowserTestOverrides` type alias in FileBrowser.test.tsx
- `makeInboxResponse / makeOutboxResponse` in MailboxView.test.tsx
Verification:
- tsc -p tsconfig.json: exit 0
- tsc -p tsconfig.app.json: exit 0
- tsc -p tsconfig.test-check.json (new — includes test files): exit 0
- vitest run: 9639 / 9641 (2 pre-existing failures
in terminal-mobile-keyboard-layout.test.ts
unrelated to this work; verified via
`git stash` + run on clean HEAD)
Adds packages/dashboard/tsconfig.test-check.json to keep this regression
guard available locally — same as tsconfig.app.json minus the test
exclude.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
When a user drags the native CSS resize grip from inside a modal and
releases the mouse over the overlay, the synthesised click event
targets the common ancestor (the overlay) — fooling the existing
e.target === e.currentTarget dismiss check. Audited every modal with
`resize: both` and switched them to a shared mousedown→mouseup tracking
pattern (new useOverlayDismiss hook) so dismiss only fires when both
events land on the overlay.
Modals fixed: TaskDetail, Settings, FileBrowser, GitHubImport,
GitManager, ScheduledTasks, Scripts, WorkflowStepManager. (Terminal
and AgentDetail were already fixed in 56e8e7324; PlanningModeModal
already had the right pattern inline.)
Also: in Settings → Authentication, the "Anthropic via Claude CLI"
card now lives inside the Authenticated group when authenticated and
the Available group otherwise, instead of floating at the top.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Modal resize + size persistence
- Extract `useModalResizePersist` hook (ResizeObserver + localStorage),
apply to Files, Git Manager, GitHub Import, Workflow Steps, Automations
(ScheduledTasks), and Settings modals. Each gets `resize: both`, sane
min/max constraints, and a unique storage key.
- Bump default heights so the modals feel less cramped (Git: 92vh,
Workflow/Automation: 80vh, Settings: 80vh / 1100px).
Scrollbar theme
- Add a global `*::-webkit-scrollbar*` + `scrollbar-color` rule in
styles.css so chat, document, system stats, file browser, usage
indicator, etc. inherit the theme. Existing per-component overrides
(.board, .column-body, .settings-sidebar, planning modal) still win.
Document view
- Collapse "Show hidden" toggle and search input onto the same row as
the Project Files / Task Documents segmented control. Stack again
below 768px.
Mailbox / Todos
- Match Todos header treatment to the Mailbox header (typography,
padding, border).
- Add top spacing above Mailbox Inbox/Outbox/Agents tab bar so the
vertical gaps balance.
Settings
- Wider, resizable, persisted Settings modal.
- Project Models description and Authentication panel get proper
horizontal padding.
- Reorder project sidebar so "General" is first.
- Plugins page: clean margins, integrate refresh button, exclude
bundled runtimes from the "Installed Plugins" list (they were
appearing twice — once erroring, once in their own section).
- New "Updates" panel with auto-check toggle + "Check now" button
(frequency control noted as needing a backend schema field).
Background sessions
- Fix stale "AI N" / planning-icon badge: `handleDeleted` in
`useBackgroundSessions` now writes a tombstone, advances the
timestamp guard, and broadcasts completion so the cross-tab
sync store stops resurrecting the deleted session on the next
merge tick. Added regression test.
- Re-fetch list on SSE reconnect so terminal events fired during a
network blip don't get permanently lost.
System stats
- Refresh button uses correct single class (was getting both `btn`
and `btn-icon`, which conflicted on padding/border).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Sidebar lists all planning sessions for the project (any status) with
inline status badges, relative timestamps, "+ New session", and an
inline delete confirm. Mobile collapses to a single pane with a back
chevron in the header.
- Closing the modal (X / Escape / overlay) no longer cancels the server
session — it stays in the list, resumable. Only an explicit Delete
cancels + removes. Removes the now-redundant minimize button.
- Add re-sync on reopen so a session whose terminal SSE event was missed
doesn't stay stuck on a stale loading view; previously only a hard
reload recovered.
- Persist user-chosen modal width/height across opens via localStorage,
driven by a ResizeObserver on the modal element.
- Theme the modal scrollbars to match the rest of the app.
- Banner dismiss only hides the banner now — it must not delete the
underlying session, since sessions are first-class in the new sidebar.
- Refresh background sessions list on SSE reconnect so the footer "AI N"
pill never gets stuck on tombstoned sessions after a network blip.
- Drop selection + broadcast completion on Create Task / Create Tasks
so the footer count drops in lockstep instead of waiting on SSE.
- Auto-scroll the AI thinking output as new tokens stream in (only when
already pinned to the tail).
- Guard overlay-click dismissal with a mousedown-on-overlay check so
releasing a resize drag outside the modal doesn't close it.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Prevent useAuthOnboarding from opening model onboarding while the setup wizard is active
- Add app-level wiring so onboarding flow respects setup wizard visibility state
- Add regression tests covering setup wizard gating and non-stacking onboarding behavior
- Add todoView to experimental feature settings and expose it in Settings modal and docs
- Derive todosEnabled in app settings hook with regression tests for enabled/disabled behavior
- Gate App routing so Todos only renders when the flag is enabled and redirects otherwise
- Hide Todos entries in header overflow and mobile navigation when the flag is off
- Add reusable ConfirmDialog component, styles, and useConfirm hook to provide async confirmation flows
- Wire ConfirmProvider at app level and migrate confirm call sites across task, agent, roadmap, plugin, and settings UI actions
- Update modal interaction patterns to support dialog reentry and consistent destructive-action confirmations
- Expand dashboard tests with confirm dialog and hook coverage plus migrated component test assertions
- Add global updateCheckEnabled setting to core schema/types and wire dashboard command to cache update checks in the CLI
- Implement dashboard server update-check cache module plus REST routes for status and refresh behavior
- Add dashboard client hook, legacy API helpers, and UpdateAvailableBanner UI to show cached CLI update notices
- Cover update-check server routes, hook behavior, banner rendering, and route registration with focused tests
- Document update-check configuration and API behavior in architecture and settings reference docs
- Add a shared useMobileKeyboard hook that tracks visualViewport overlap on mobile browsers
- Apply keyboard overlap CSS variables to QuickChatFAB so fullscreen mobile panels resize above the virtual keyboard
- Apply the same keyboard-aware sizing to ChatView mobile thread layout and auto-scroll messages when keyboard opens
- Expand ChatView, QuickChatFAB, and hook test coverage for mobile keyboard detection and viewport-resize behavior
The X button in the footer's Background Tasks popover silently failed when
ai_sessions.lockedByTab pointed to a stale/closed tab: cancelPlanning et al.
returned 409, dismissSession set lockConflict and bailed before calling
deleteAiSession. Always proceed to delete; the dismissal tombstone already
prevents stale SSE/sync from resurrecting the session.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Track and thread usage trigger bounds from Header through modal management to UsageIndicator
- Render desktop usage details as an anchored popover while preserving existing modal wiring behavior
- Extract usage indicator styling into a dedicated UsageIndicator.css and remove planning-mode style duplication
- Expand header and usage indicator tests to cover anchor rect propagation and popover behavior
- Stabilize CLI bundle output sqlite assertion for workspace test reliability
- Refactor useFavorites callbacks to read latest state instead of closing over stale favorites values
- Ensure add/remove/toggle flows update local state and persistence from a consistent source of truth
- Add regression tests that reproduce stale-closure timing scenarios and verify favorites remain correct
- Expand hook test coverage for rapid consecutive updates to guard against future closure regressions
- Reset memory backend status to unresolved during fetches so stale capabilities are never treated as authoritative
- Gate MemoryView read-only and qmd warning states on resolved backend status and add an explicit checking state for qmd availability
- Update SettingsModal memory editor gating so writes remain enabled while backend status is still loading, then enforce resolved read-only state
- Expand dashboard tests for useMemoryBackendStatus, MemoryView, and SettingsModal to lock in loading, warning, and refresh behavior
- Treat devServer as a legacy alias of canonical devServerView in Experimental Features
- Canonicalize feature keys when rendering and saving settings to prevent duplicate Dev Server rows
- Preserve enabled-state compatibility by reading either key in useAppSettings
- Add regression tests for single-row rendering, alias normalization on save, and legacy hook behavior
- Add dismissal tombstone tracking in useBackgroundSessions to ignore stale refresh, sync, and SSE updates for dismissed session IDs
- Treat terminal session states consistently and broadcast completion on dismiss so local/session-sync state remains converged
- Keep lock-conflicted planning cancellations visible instead of falsely dismissing the session when another tab owns the lock
- Expand useBackgroundSessions regression tests to cover stale sync/SSE resurrection, refresh behavior, lock-conflict handling, and newer-authoritative restore paths
- Extend column action menus beyond Todo to include In Progress and In Review
- Add Stop All action that pauses only non-paused tasks with confirmation and success/error toasts
- Add Move All to Todo action with confirmation and partial-failure handling for bulk moves
- Thread pauseTask through useTasks, App, and Board so column menus can trigger task pausing
- Expand Column tests to cover new menu actions, disabled states, and bulk operation behavior
- Introduce a new "project-setup" onboarding step between GitHub and first-task actions
- Gate first-task navigation on project selection and add contextual setup-wizard guidance
- Centralize ordered onboarding step definitions in shared state and reuse them for navigation/progress labels
- Extract ModelOnboardingModal styles into a dedicated component CSS file and update tests/mocks for the new step order
A duplicate agent id slipping through useAgents (race between initial fetch
and an SSE refresh, or backend pagination edge case) was flooding React with
"Encountered two children with the same key" warnings. With the active panel
re-rendering on every transcript event the warning fired every few ms and
snowballed the console buffer until the page crashed with OOM.
Dedupe by id at the hook (so every consumer benefits) and again in
ActiveAgentsPanel as belt-and-braces.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The hook was prepending every SSE log entry into React state without bound.
Long-lived active agents flooded the array (hundreds of MB) and the dashboard
eventually died with an out-of-memory crash. Cap the buffer at 200 entries —
the UI only ever renders the first 20, so the cap is generous but finite.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The global SSE broadcast called res.write() without checking the return
value, so a paused or backgrounded client would silently accumulate
every store event for every entity (tasks, missions, plugins, agents,
chat, ...) into res.outputData until the dashboard process OOMed.
Add a 4 MB writableLength threshold; when exceeded, tear down the
connection so the OS releases the buffer.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
