- Use bundled Nous Research SVG (inverted) on the Hermes runtime card
instead of the generic ProviderIcon.
- Lock GitHub-star and Help buttons in the settings header to a uniform
26px height so the differing icon sizes don't misalign them.
- Drop the inline Delete button from the idle/terminated agent-card
states — deletion stays available from the agent detail view.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Replace process.env.HOME fallbacks with os.homedir() in dashboard usage
probes and the hermes plugin profile resolver so unset HOME no longer
yields literal "~" paths. Skip POSIX process-group semantics on Windows
in engine/merger and dashboard-tui's pgrep-based vitest killer. Add
shell: true to npx spawns in CLI skills/extension so .cmd shims resolve
on Windows, and route test:build-exe through cross-env.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- packages/dashboard/src/routes.ts: hermes/openclaw runtime metadata declares
`version` as `string | undefined`, but BUNDLED_PLUGIN_RUNTIMES requires
`string`. Coalesce to "0.0.0" so the bundled fallback list type-checks.
- packages/dashboard/app/components/__tests__/NodeStatusIndicator.test.tsx:
the file had been reduced to a stub ("placeholder") which the test runner
parsed as an undefined identifier. Restore the original 155-line suite from
c4a48491e; the underlying NodeStatusIndicator component is unchanged.
All 9639 dashboard tests pass; pnpm typecheck is clean.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Merge metadata for the three example runtime plugins bundled as workspace
dependencies of @fusion/dashboard with pluginLoader.getPluginRuntimes(), so
the NewAgentDialog "Plugin Runtime" dropdown populates without requiring
an explicit `fn plugin install`. Installed plugins override the bundled
fallback by runtimeId.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The Paperclip runtime card's Local CLI tab previously only derived an apiUrl
from the local config and then made HTTP calls itself, so the Test action and
the company/agent pickers ignored the user's onboarded CLI auth context.
Add CLI-backed variants for every Paperclip call that has a `paperclipai`
counterpart, and route through them when transport=cli — both from the
settings card and from the runtime adapter's prompt path.
- New plugin functions spawning `paperclipai … --json`:
* probePaperclipViaCli, listCompaniesViaCli, listCompanyAgentsViaCli
(settings card test + pickers)
* createIssueViaCli, getIssueViaCli, agentsMeViaCli (runtime hot path)
- New dashboard routes: /providers/paperclip/cli-status, /cli-companies,
/cli-agents (read-only façades over the plugin's CLI helpers)
- PaperclipRuntimeCard: branches on transport=cli to use the cli-* fetchers
- PaperclipRuntimeAdapter: stores transport on the session and routes
createIssue/getIssue + identity derivation through CLI variants in CLI mode;
raises a clear error when agentId is unset in CLI mode (paperclipai has no
/agents/me equivalent)
- getIssueComments / wakeAgent / getRunEvents stay on HTTP (no matching
paperclipai subcommands) and continue to use the apiKey discovered from
the local paperclipai config, so CLI mode still works end-to-end
- Tests: 9 new paperclip-client tests covering each CLI variant + 5 new
adapter tests for the Local-CLI transport branch (64/64 plugin tests pass)
- Update routes.test for the existing BUNDLED_PLUGIN_RUNTIMES fallback so
the bundled hermes/openclaw/paperclip entries are expected alongside
installed plugins
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Define and export unavailable node policy types in core settings interfaces
- Add project-level unavailable node policy default and runtime validation helper
- Add core unit coverage for unavailable node policy parsing and acceptance cases
- Guard Paperclip mint requests to include companyId only when available for type-safe payloads
- Document unavailable node policy in the settings reference
- Hermes / OpenClaw plugin index.ts now re-export `probeHermesBinary` /
`probeOpenClawBinary` and their status types so the dashboard's
`runtime-provider-probes.ts` façade can import them via the public
package entry instead of deep paths.
- Dashboard `package.json` adds `@fusion-plugin-examples/hermes-runtime`,
`…/openclaw-runtime`, `…/paperclip-runtime` as workspace deps so
pnpm symlinks them into `packages/dashboard/node_modules/`. Without
these, the new probe imports failed with "Cannot find module" during
`pnpm typecheck`.
This clears 6 of the 9 outstanding typecheck errors. The remaining 3 are
in the in-flight Hermes plugin rewrite (runtime-adapter still imports
from a deleted `./pi-module.js`; the new `index.ts` calls a factory
with the wrong arg type) and should be resolved by the same change set
that landed the rewrite.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two long-standing test failures, root-caused and fixed.
terminal-mobile-keyboard-layout.test.ts (2 tests)
CSS structure changed: the desktop terminal modal moved its initial
width/height into companion `:not([style*="width"])` /
`:not([style*="height"])` rules so persisted resize values can win
over the default. The tests still searched the base block. Updated
the helpers to extract from each rule explicitly, asserting:
width: min(1800px, …) in the :not([style*="width"]) rule
max-width: calc(100vw - …) in the base rule
85vh somewhere in the :not([style*="height"]) rule
min-height + max-height: calc(100dvh - 40px) in the base rule
Same constraints, just reflecting the current selector layout.
ModelOnboardingModal.test.tsx (2 tests)
Component refactor (commit c94e9be31) made the "what GitHub unlocks"
feature list conditional on `!isGitHubReady`, so the test that mocks
`authenticated: true` no longer sees "Import issues as tasks". And
the gh-CLI optional explanation copy was rewritten to "OAuth from
the dashboard is optional" instead of "OAuth integration in
Settings → Authentication is optional". Updated:
- "shows connected state": assert the new connected-state
sentence ("GitHub is connected — issue imports …") instead of
the now-hidden feature list line.
- "explains gh CLI auth": use getAllByText for the two-place
"GitHub CLI is already authenticated" copy and match the
current "OAuth from the dashboard is optional" sentence.
Also adds the missing `afterEach` import to silence its TS error.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Token counts on the task detail Stats panel could overflow the metric
box for large values (e.g. 10-digit comma-separated totals). Switched
each metric to a container-query inline-size context and gave the
value a fluid font-size of `clamp(13px, 13cqw, 28px)` so it grows to
fill the box up to a 28px cap and shrinks down to 13px on narrow grid
columns. Combined with `min-width: 0`, `overflow-wrap: anywhere`, and
the metric's `overflow: hidden`, long values now stay within the box
and short values render as large as they can without dominating.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- ModelOnboardingModal: native CSS resize with persisted size via
useModalResizePersist; default ~1100px wide; modal scrolls inside
the content area instead of clipping.
- Provider cards flex-wrap so the API-key form drops to a full-width
row instead of being squished into a fixed side column. Connected
badge no longer stretches the full body width.
- Step changes now reset content scrollTop so each page lands at the
top.
- GitHub step: prominent GitHub mark via ProviderIcon; when gh CLI is
authenticated the intro reads as 'you are all set', primary CTA
becomes 'Continue with gh CLI auth', and a secondary
'Connect OAuth (optional)' button is offered.
- CustomModelDropdown: highlight init runs once per open session, and
filter changes reset highlight to top + scrollTop=0, fixing the
scroll-fight when filtering models.
- TUI dashboard: at >=150 cols, Stats panel sits to the left of Logs;
bottom row drops to Utilities + Settings.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
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>
Two related dashboard fixes.
1. Card timer mismatch: the board card timer chip showed only workflow
runtime (e.g. <1m on FN-2716) while the task detail Stats panel
reported "Total execution time" of 7m+ for the same task. Cause —
the slim board listing strips `task.log` to keep payloads small, so
the card's client-side `[timing]` log scan returned 0. Now the slim
path aggregates `[timing] … in <N>ms` durations server-side into a
new `task.timedExecutionMs` field before stripping the log; the
card prefers this aggregate, falling back to the client scan when
the full log is loaded (TaskDetailModal). Wire payload stays slim.
2. View Changes diff modal: defaulted to `90vw × 80vh` and was not
user-resizable. Switched to `min(95vw, 2200px) × min(90vh, ...)`
default with `resize: both`, persisted via useModalResizePersist
(`fusion:changes-diff-modal-size`). Mobile keeps fullscreen layout.
Overlay dismiss switched to the shared `useOverlayDismiss` hook so
resize-drags that release on the overlay don't close the modal.
Updated the diff modal's regression tests to match the new constraint
shape (still asserts max-height clamps to viewport via calc()).
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 95566795e; 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>
Terminal modal and agent detail view are now resizable via the native
CSS grip with sizes persisted per-modal in localStorage. Mobile keeps
fullscreen layout (resize disabled, !important overrides any persisted
desktop dimensions).
Both modals previously dismissed when a drag-resize started inside the
modal but released over the overlay — the synthesised click event
targets the common ancestor (overlay), tripping the e.target ===
e.currentTarget dismiss check. Switched to mousedown→mouseup tracking
so dismiss only fires when both events land on the overlay.
Terminal additionally observes its own pixel box via ResizeObserver
and refits xterm on every grip drag — `resize: both` doesn't emit
window resize.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two more brand-id spellings observed in the wild that previously fell
through to the generic Cpu glyph:
fireworksai → FireworksIcon (joins fireworks / fireworks-ai)
qwen-ai → QwenIcon (joins qwen / qwen-coder / alibaba / tongyi)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Audited the pi-ai built-in model catalog provider IDs against the
ProviderIcon registry and filled every remaining gap.
New icons:
--provider-cerebras (#f1592a) → CerebrasIcon (concentric arcs)
--provider-groq (#f55036) → GroqIcon (lightning bolt)
--provider-vercel (text) → VercelIcon (triangle wordmark)
New aliases routed to existing icons:
minimax-cn → MiniMaxIcon
azure-openai-responses → AzureIcon
google-gemini-cli → GeminiIcon
google-generative-ai → GeminiIcon
opencode-go → OpencodeIcon
vercel-ai-gateway → VercelIcon
After this commit every provider id emitted by `@mariozechner/pi-ai`'s
register-builtins (anthropic, openai, openai-codex, google,
google-vertex, google-gemini-cli, google-antigravity,
amazon-bedrock, mistral, openrouter, fireworks, cerebras, groq,
huggingface, kimi-coding, minimax, minimax-cn, zai, xai, opencode,
opencode-go, github-copilot, azure-openai-responses,
vercel-ai-gateway) plus the user-curated set (qwen, lm-studio,
hugging-face, alibaba/tongyi, hf, lmstudio…) renders a brand icon
instead of falling back to the generic Cpu glyph.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds six new provider icons rendered as inline SVG, each backed by a new
brand token in styles.css:
--provider-qwen (Qwen / Tongyi)
--provider-lmstudio (LM Studio)
--provider-huggingface (Hugging Face)
--provider-mistral (Mistral AI)
--provider-azure (Microsoft Azure / Azure OpenAI)
--provider-fireworks (Fireworks AI)
Aliases registered:
qwen / qwen-coder / alibaba / tongyi
lmstudio / lm-studio
huggingface / hugging-face / hf
mistral / mistral-ai
azure / azure-openai
fireworks / fireworks-ai
Also extends the Gemini icon to cover every Google-product alias the
catalog throws at it: google-vertex, vertex, google-cloud-code,
cloud-code, antigravity (in addition to the existing google,
gemini, google-antigravity).
Tests cover each new mapping (and confirm the Gemini alias group all
resolves to the gemini icon with --provider-gemini color).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- ModelOnboardingModal: completed/skipped step indicators render as
<button> so they're clickable for review. Without explicit reset,
browsers gave them their default white-ish background. Added
background: transparent + border/padding/font-family resets to
.model-onboarding-step-indicator so it inherits cleanly from the
surrounding card on every theme.
- SettingsModal authentication: tightened spacing around providers —
group margin 16→10, group label margin 8→4, provider card margin
8→4, header padding 12 16 → 8 12. Less wasted space between rows.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The Settings, Workflow Steps, Automations (and friends) modals all carried
desktop `min-width: 480-520px` constraints from the resize work. On phones
that pushed them off-screen, and the overlay's 10vh top padding plus a
fractional `height: 80vh` left awkward gaps top and bottom.
Apply the same `@media (max-width: 768px)` full-screen-sheet treatment
that File Browser already had to every resize-aware modal: drop the
overlay padding, set width/height to `100vw`/`100dvh`, zero out the
min-* / max-* constraints, and disable `resize` (touchscreens can't grab
the corner grip anyway).
Also tightened the existing File Browser and GitHub Import overrides to
match (`min-width: 0`, `min-height: 0`, `resize: none`).
For TaskDetailModal the mobile rules go inside the existing first
`@media (max-width: 768px)` block to keep the FN-1331 detail-body
padding regression test happy (it captures from the first 768px media
query greedily).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Update-check frequency (end-to-end)
- Add `updateCheckFrequency: "manual" | "on-startup" | "daily" | "weekly"`
to GlobalSettings (default: "daily").
- Backend `performUpdateCheck` honors the frequency: TTL = day/week,
`manual` returns cache or empty without hitting npm, `on-startup`
refreshes once per process lifetime then serves cache. `/refresh`
route forces network regardless.
- Settings → Updates surfaces a working `<select>` for the cadence,
disabled when auto-checks are off entirely.
- Tests: 4 new cases covering ttlForFrequency mapping, weekly window,
manual semantics, on-startup once-per-process behavior.
TUI update notice
- Read cached update result synchronously on TUI startup. When an
update is available, render a yellow notice line on the splash and
a colored ● next to the version in the status bar.
Settings modal header polish
- Add "Star on GitHub" pill (icon + Star + cached star count from the
GitHub API, 1h localStorage TTL) and "Help" button (opens project
Discussions). Both link to the Runfusion/Fusion repo.
- Star button auto-hides after the user clicks it (intent = star),
tracked in localStorage `fusion:github-star-clicked`.
- Settings → General gets a "Show Star on GitHub button" checkbox so
users can hide it preemptively. New global setting
`showGitHubStarButton: boolean` (default true) gates rendering.
More resizable modals
- Task Detail modal: 85vh default, resize: both, persisted via
useModalResizePersist (key `fusion:task-detail-modal-size`).
- Quick Chat FAB: full 8-direction resize (4 corners + 4 edges) via
pointer-event handlers; persisted to
`fusion:quick-chat-size-<projectId>`. Each handle has the right
cursor + role="separator" for accessibility.
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>
The task card timer chip previously fell back through several metrics
(timed duration → workflow runtime → wallclock), so cards showed only a
subset of execution time. For FN-2714 this rendered <1m on the card while
the stats tab reported >2m of workflow runtime.
The chip now reports the sum of [timing]-tagged log events and workflow
step runtime (matching the new "Total execution time" metric in the stats
panel), with live elapsed for in-progress workflow steps. When neither
metric is recorded, the chip is hidden rather than falling back to
wallclock.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Move TaskCard time indicator rendering from the header back into the footer metadata row
- Render timer and files-changed metadata together in the shared footer row layout
- Update timer chip CSS to stay right-aligned with margin-left:auto and prevent shrinking
- Update TaskCard and mobile board tests to assert footer placement and alignment behavior
- Render an accessible loading status when the first agents fetch is in flight
- Keep existing agent cards visible during refresh instead of replacing the grid with a loader
- Add AgentsView CSS for the loading container with responsive sizing
- Add regression tests covering initial loading, post-load cleanup, and refresh behavior
The live-token-usage refactor (d30edfda5) merged the SSE-updated `task`
prop on top of `fullDetail` to keep tokenUsage/status fresh. SSE strips
`log` to [] (stripTaskListHeavyFields) for list payloads, so the spread
clobbered fullDetail.log and the Activity timeline rendered empty.
Carve `log` out of the merge alongside `prompt`. Adds a regression test
that simulates the SSE-stripped task prop with a populated fullDetail.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Render the countdown timer in the task card header next to metadata for clearer scanning
- Remove legacy footer timer styling and adjust TaskCard structure accordingly
- Update TaskCard unit tests to assert the new timer placement and header behavior
- Expand mobile board tests to validate timer visibility and placement across card states
When the worktree-recycle pool reassigned a path to a new task, the old
task's diff endpoints kept reading the new task's branch state — surfacing
unrelated commits as the original task's "files changed" list.
- Clear task.worktree/branch in the merger after the worktree is released
to the pool or removed, so the path no longer points anywhere.
- Validate the worktree's current branch matches task.branch in the three
worktree-backed diff endpoints; on mismatch return empty rather than
diffing against a foreign branch.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
When the worktree-recycle pool reassigned a path to a new task, the old
task's diff endpoints kept reading the new task's branch state — surfacing
unrelated commits as the original task's "files changed" list.
- Clear task.worktree/branch in the merger after the worktree is released
to the pool or removed, so the path no longer points anywhere.
- Validate the worktree's current branch matches task.branch in the three
worktree-backed diff endpoints; on mismatch return empty rather than
diffing against a foreign branch.
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
- Update PlanningModeModal styles to decrease top padding in the question scroll area
- Improve modal content density so question content starts closer to the top
- Add runtime selection controls to the NewAgentDialog custom tab flow
- Update dialog behavior to persist and apply the selected runtime when creating custom agents
- Add NewAgentDialog styling updates for the new runtime UI states
- Expand NewAgentDialog tests to cover runtime selection behavior on the custom tab
- Add TodoView controls to create todo tasks directly and optionally assign an agent
- Style the TodoView agent picker dropdown for desktop and mobile interactions
- Add TodoView tests covering todo task creation and agent assignment behavior
- Simplify dashboard TUI status grid sizing to use flexible panel layout
- Replace runtime source radio groups with tablist/tab buttons in NewAgentDialog and AgentDetailView
- Preserve runtime mode behavior by keeping model/runtime selection handlers and reset logic intact
- Align AgentDetailView runtime selector styling with tab pattern and remove obsolete radio-toggle CSS
- Update AgentDetailView tests to assert tab semantics via role and aria-selected
- 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 engine shell utility to resolve the correct shell per platform
- Update routine runner to execute commands through shared shell selection
- Update cron runner to use the same cross-platform shell behavior
- Apply shared shell handling in dashboard routes for command execution
- Add shell utility tests covering platform-specific selection behavior
- Fix mobile refresh button icon visibility in SystemStatsModal
- Add auto-refresh activity indicator with spinning refresh state
- Update modal styles for responsive refresh controls and animation states
- Expand SystemStatsModal tests to cover mobile visibility and auto-refresh indicator behavior
- Remove token usage badge rendering and compact token formatting from TaskCard
- Drop token usage fields from TaskCard memo equality checks now that the badge is gone
- Delete obsolete TaskCard token badge CSS rules and icon import
- Simplify TaskCard tests by removing token badge assertions and related fixtures
- Keep /api/system-stats process and host metrics available even if scoped project resolution throws
- Fall back task and agent aggregates to zero when scoped task or agent stores cannot be loaded
- Preserve vitest process reporting while guarding optional vitestLastAutoKillAt lookup behind scoped context success
- Add route tests for scoped project-not-found fallback behavior and include a patch changeset for @runfusion/fusion
- Add dashboard API endpoint and server route to terminate Vitest worker processes safely
- Extend legacy API client with killVitestProcess support for frontend invocation
- Add SystemStatsModal UI controls and styling for triggering Vitest kill actions, including mobile width stability fixes
- Expand route and modal tests to cover new controls and behavior
- Add changeset and architecture docs updates for the new Vitest kill capability
The black-translucent status bar overlays the page in iOS standalone PWAs,
clipping close buttons in full-screen mobile modals and the QuickChatFAB
panel. Add `padding-top: env(safe-area-inset-top, …)` so headers render
below the status bar.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Reduce Task Detail move split chevron width on mobile for a more compact control
- Add mobile padding override to TaskCard send-back button to match sibling action sizing
- Normalize task action button dimensions across card and detail contexts
Capture per-session token usage from pi-coding-agent's getSessionStats()
after each promptWithFallback in the executor and merger paths, so
task.tokenUsage populates live during runs and reflects final totals on
done tasks. Previously the executor never read session usage and only
the heartbeat path bumped agent token totals, leaving task.tokenUsage
undefined even after completion.
Stats panel and done-card timing also now reflect live state: the modal
overlays the SSE-updated task prop on top of the one-shot fullDetail
snapshot, in-progress workflow steps contribute live elapsed to the
Workflow runtime metric, and the done card uses Timed duration (matching
the stats tab) with workflow runtime as fallback. Time indicator labels
coarsened to <1m / Nm / Nh / Nd.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Reposition the show-hidden toggle into the UsageIndicator inline layout next to usage details
- Update UsageIndicator styles to support the new inline button placement and spacing
- Add UsageIndicator tests that assert the show-hidden button renders in the correct placement
- Preserve existing behavior while improving visual hierarchy for hidden-usage toggling
- Reduce horizontal spacing for archive/unarchive and send-back action buttons on TaskCard
- Replace hardcoded padding, margin, gap, and font-size values with design-token-based calculations
- Keep mobile action-button padding aligned with the updated compact sizing
- 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 token-based scrollbar styling for .detail-body across Firefox and WebKit pseudo-elements
- Scope agent log scrollbar rules to .detail-section--agent-log .agent-log-viewer-scroll to avoid selector target drift
- Add TaskDetailModal CSS regression tests that assert scrollbar declarations for modal body and agent log containers
- Introduce shared test helpers for robust selector/declaration matching in stylesheet assertions
