* fix(web): add load-more pagination for Done column on issue board
The Done column was capped at 50 issues with no way to load more.
Track doneTotal in the TQ cache and add a useLoadMoreDoneIssues hook
that fetches the next page and merges it into the unified issue cache.
The Done column now shows a "Load more" button when there are
additional items.
- shared/types/api.ts: add doneTotal to ListIssuesResponse
- core/issues/queries.ts: store doneTotal from the done-status response
- core/issues/mutations.ts: add useLoadMoreDoneIssues hook, update
create/delete mutations to maintain doneTotal
- core/issues/ws-updaters.ts: maintain doneTotal on WS events
- features/issues/components/board-column.tsx: accept optional footer
- features/issues/components/board-view.tsx: render Load more button
in Done column
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(web): address review issues in done-column load-more
1. Fix total over-counting: loadMore no longer inflates total since
the initial query already includes all done issues in total count.
2. Fix onIssueUpdated: maintain doneTotal when issue status changes
to/from done via WS events.
3. Make doneTotal optional in ListIssuesResponse since it's a
frontend-only field not returned by the backend API. All reads
now use ?? 0 fallback.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* feat: add allowedDevOrigins to Next.js config for non-localhost dev access
Wire CORS_ALLOWED_ORIGINS env var into Next.js allowedDevOrigins config
so that cross-origin HMR/webpack requests from Tailscale or other
non-localhost IPs are not blocked during development.
Fixes#317
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix: keep port in allowedDevOrigins
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* fix(daemon): update existing worktree to latest remote on reuse
When an agent receives a new task on the same issue, the execution
environment is reused and the repo worktree already exists on disk.
Previously, `multica repo checkout` would fail because `git worktree add`
cannot create a path that already exists — so the agent worked on stale
code from the prior task.
Now `CreateWorktree` detects existing worktrees and updates them:
fetch origin, reset working tree, then checkout a new branch from the
latest remote default branch. The previous task's branch is preserved.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(daemon): propagate actual branch name and use correct ref in worktree reuse
- Return (string, error) from updateExistingWorktree so collision-retried
branch name propagates to WorktreeResult
- Use baseRef directly instead of origin/baseRef — bare clone refspec maps
remote branches to local refs, so remote-tracking refs may not exist
- Remove redundant fetch (worktree shares object store with bare clone)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
actor_id identifies the user, not the browser tab. Filtering WS events
by actor_id broke multi-tab sync — other tabs of the same user would
silently miss updates. Instead, make all WS cache handlers idempotent
(dedup checks on add, no-op on duplicate merge/filter) so mutations and
WS events coexist safely without filtering.
- WSClient: pass actor_id to event handlers for future per-handler use
- use-realtime-sync: remove isSelf() gating from onAny and specific handlers
- useCreateIssue: add .some() dedup guard + onSettled invalidation
- use-issue-reactions: remove payload-level self-filter (dedup already present)
- use-issue-timeline: remove payload-level self-filter on comment:created,
reaction:added, reaction:removed (dedup already present)
- Clean up useCallback deps that no longer reference userId
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Remove align-middle from IssueMentionCard (alignment is container's job)
- Add inline align-middle wrapper span in ReadonlyContent for vertical alignment
- Add img component with max-width constraint to prevent overflow
- Issue mention clicks open in new tab (matches Tiptap behavior)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replace ContentEditor editable={false} with lightweight ReadonlyContent
in comment cards. Each comment no longer creates a full ProseMirror
instance for readonly display.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add del selector to strikethrough CSS for react-markdown compatibility
- Create ReadonlyContent using react-markdown + lowlight + content-editor.css
- Export from editor module index
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
1. Separate ReportTaskUsage endpoint (POST /api/daemon/tasks/{id}/usage)
so usage is captured independently of complete/fail — fixes usage loss
for failed/blocked tasks.
2. Add usage tracking for all four providers:
- Claude: already done (stream-json message.usage)
- OpenCode: extract from step_finish.part.tokens
- OpenClaw: extract from step_end.data token fields
- Codex: extract from turn/completed and task_complete usage fields
3. Remove usage from CompleteTask payload — all usage goes through the
dedicated endpoint now.
Extract token usage from Claude Code's stream-json output in real-time
during task execution, replacing the inaccurate global JSONL log scanner.
- New `task_usage` table: tracks (task_id, provider, model) level usage
- Agent SDK: parse `message.usage` from assistant messages, accumulate
per-model and return in Result
- Daemon: convert agent usage to entries, send with CompleteTask
- Server: store usage on task completion, expose workspace-level
aggregation APIs (GET /api/usage/daily, GET /api/usage/summary)
AppSidebar renders before workspace hydrates. useWorkspaceId() throws
when workspace is null. Fix: read workspace?.id directly from store,
use enabled guard on inbox query. This fix was in commit 030627c but
got reverted by subsequent merge with main.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
issueListOptions now makes 2 api.listIssues calls (open_only + closed page).
Tests that mock the response must return data only for the open_only call,
otherwise issues appear twice in the merged result.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Board DnD: use local pendingMove state for instant card placement,
bypassing TQ's async setQueryData notification delay
- useUpdateIssue: add list invalidation to onSettled (was only detail)
- use-realtime-sync: add isSelf check to specific issue WS handlers
(prevents redundant cache writes for own mutations)
- Clean up debug console.logs from board-view, issues-page, mutations
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
When authenticating via CLI, the login page called api.verifyCode()
directly and redirected to the CLI callback without saving the JWT
to localStorage or setting the logged-in cookie. This meant the
browser had no session after CLI login, forcing users to log in
again when visiting multica.ai.
Now the token is saved to localStorage and the cookie is set before
redirecting to the CLI callback, so both CLI and web app share the
same authentication.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* refactor: remove agent triggers config field
Remove the triggers field from agent configuration. The on_assign,
on_comment, and on_mention behaviors are now always enabled (hardcoded),
as decided in the Agentflow design discussion (MUL-372).
Changes:
- Database: migration 032 drops triggers column from agent table
- Backend: remove triggers from create/update agent APIs and response
- Backend: simplify trigger-checking logic to always-enabled
- Frontend: remove TriggersTab UI and AgentTrigger types
- Tests: remove trigger config unit tests (no longer configurable)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* refactor: also remove agent tools config field
Remove the tools field from agent configuration alongside triggers.
The tools field was a placeholder — stored in the DB and shown in the
UI but never passed to the daemon or used at runtime.
- Database: migration 032 now also drops tools column
- Backend: remove tools from create/update agent APIs and response
- Frontend: remove ToolsTab UI, AgentTool type, and tools tab
- Update landing page copy
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(test): remove tools/triggers columns from test fixtures
The test fixtures still referenced the dropped tools and triggers
columns when inserting agent rows, causing CI failures.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Devv <devv@Devvs-Mac-mini.local>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
P0: Add onSettled: invalidateQueries to 10 mutations that had onMutate
optimistic updates but no server confirmation. With staleTime: Infinity,
missing onSettled means cache could permanently drift from server state.
Mutations fixed:
- useDeleteIssue, useBatchDeleteIssues (issue list)
- useUpdateComment, useDeleteComment, useToggleCommentReaction (timeline)
- useToggleIssueReaction (reactions)
- useToggleIssueSubscriber (subscribers)
- useMarkInboxRead, useArchiveInbox, useMarkAllInboxRead (inbox)
P2: Change refetchOnReconnect from false to true as safety net
for HTTP reconnection before WS reconnection.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The merge with origin/main (fe9479d6) silently reverted all consumer-side
migrations, leaving core/ as dead code. Restored all 39 files from
pre-merge commit 6032b5df, plus main's trigger.config null fix for
agents page.
Verified: 59 @core/ imports across features/ and app/, all stores
gutted/deleted, realtime sync uses queryClient not Zustand.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add core/ layer documentation (queries, mutations, WS updaters)
- Rewrite State Management section: TQ for server state, Zustand for client-only
- Update features table: reflect gutted stores (issues, inbox, workspace)
- Add @core/* import alias examples
- Update Data Flow diagram to include TQ layer
- Restore @core/* path alias in tsconfig + vitest (lost during merge)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
AppSidebar renders outside the workspace guard in dashboard layout.
On first login, workspace hasn't hydrated yet → useWorkspaceId() throws.
Fix: read workspace?.id directly from store, use enabled guard on inbox query.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Relax ClaimAgentTask SQL constraint from per-issue to per-(issue, agent)
serialization, allowing different agents to run in parallel on the same issue
- Update GetActiveTaskForIssue API to return all active tasks (array) instead of
just the first one
- Refactor AgentLiveCard to render one card per active task, routing WebSocket
messages by task_id for independent timelines
- Fix shouldEnqueueOnComment to use per-agent dedup so a mentioned agent's
pending task doesn't block the assigned agent's on_comment trigger
Closes MUL-160
- Create core/inbox/ with queries, mutations, ws-updaters
- Migrate inbox page: useQuery + mutation hooks replace useInboxStore + api.*
- Migrate sidebar unread badge to read from TQ cache
- Delete useInboxStore (127 lines) — inbox has no client-only state
- Remove inbox deps from workspace store (hydrate + switch)
- Fix WS sync: use useQueryClient() instead of getQueryClient() singleton
to ensure WS handlers write to the same QueryClient instance that
components read from (singleton is unreliable under Next.js HMR)
- Add onInboxIssueStatusChanged for issue status sync in inbox items
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add ListOpenIssues SQL query (excludes done/cancelled, no LIMIT)
- Add CountIssues SQL query for true total count
- Backend: support open_only=true param, fix total to return real count
- Frontend: two-phase fetch in issue store (all open + first 50 closed)
- Add fetchMoreClosed action for paginated closed issue loading
- Replace all hardcoded limit:200 with store.fetch() calls
Resolves MUL-369
Co-authored-by: Devv <devv@Devvs-Mac-mini.local>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Remove mock issues[] and server state fields from useIssueStore mocks
since the store now only holds activeIssueId. Data flows through
TanStack Query (mockListIssues) not the store.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The callback page was placed under app/(auth)/callback/ — a Next.js
route group — which mapped to /callback instead of /auth/callback.
Move it to app/auth/callback/ so the URL matches the Google OAuth
redirect URI.
- Migrate issue-detail.tsx: useQuery for issue data, useUpdateIssue/useDeleteIssue
- Migrate issues-page.tsx, my-issues-page.tsx, board-card.tsx: useQuery for list
- Migrate batch-action-toolbar.tsx, create-issue.tsx: mutation hooks
- Migrate edge consumers: mention-suggestion, mention-view, agents page, issue-mention-card
- Remove Zustand writes from WS sync (TQ cache is now sole source of truth)
- Remove useIssueStore.fetch() dependency from workspace store
- Gut useIssueStore to client-only: { activeIssueId, setActiveIssue }
- Update test wrappers with QueryClientProvider
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
