Replace all inline avatar implementations (initials divs, Bot icons,
inline img tags) with the shared ActorAvatar component for consistency.
- Extend AssigneePicker with controlled open/onOpenChange, triggerRender,
and align props to support batch toolbar and other contexts
- Replace BatchAssigneePicker (~130 lines) with shared AssigneePicker
- Replace issue-detail sidebar inline DropdownMenu with AssigneePicker
- Add canAssignAgent filtering to issue-detail more menu
- Replace inline avatars in: filter panel, members-tab, agents page,
mention-hover-card, subscribers AvatarGroup
- Add data-slot="avatar" to ActorAvatar for AvatarGroup compatibility
- Add triggerRender prop to PropertyPicker for custom trigger elements
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Keep ActorAvatar for agent list/detail avatars over dev's inline
img+initials. Also unify the new avatar upload preview to use
ActorAvatar instead of getInitials helper.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add avatar upload UI to the agent SettingsTab, matching the existing
member avatar upload pattern. Also update the agent list item and
detail header to display the uploaded avatar image.
Same issue as inbox: without a key, React reuses the AgentDetail
instance when switching agents. This causes activeTab and SettingsTab
form values (name, description, visibility, maxTasks) to persist
from the previously selected agent.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Adds a Settings tab to the agent detail panel with:
- Name and description editing
- Visibility toggle (workspace/private) matching the create dialog pattern
- Max concurrent tasks configuration
- Runtime info display (read-only)
* feat(agents): reply as thread instead of top-level comment
When an agent responds to a user comment, the reply is now nested under
the triggering comment (parent_id) instead of appearing as a separate
top-level comment. Also enables on_comment trigger by default for newly
created agents.
- Add trigger_comment_id column to agent_task_queue (migration 028)
- Pass triggering comment ID through EnqueueTaskForIssue → task → createAgentComment
- Include parent_id in WebSocket broadcast for agent comments
- Default agent creation includes both on_assign and on_comment triggers
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* feat(cli): add --parent flag to comment add for threaded replies
The agent posts comments via the CLI, so the correct fix is giving it a
--parent flag rather than wiring trigger_comment_id through the task
infrastructure. The agent reads the comment list, decides which comment
to reply to, and passes --parent <comment-id>.
- Add --parent flag to `multica issue comment add`
- Update agent runtime instructions to explain --parent usage
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* feat(daemon): pass trigger_comment_id to agent execution context
The agent now knows which comment triggered its task and gets an explicit
instruction to reply to it using --parent. The trigger_comment_id flows
from the DB through the claim response, daemon Task struct, and into
issue_context.md where the agent sees it.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(comments): agent replies to thread root, matching frontend behavior
When the triggering comment is itself a reply (has parent_id), resolve
to the thread root so the agent's reply stays in the same flat thread.
This matches the frontend where all replies share the top-level parent.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* feat(cli): show parent_id and full IDs in comment list
The table output now includes a PARENT column and shows full comment IDs
(not truncated) so agents can see thread structure and use --parent.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* feat(daemon): instruct agents to always use --output json
Agents now see explicit guidance to use --output json for all read
commands, ensuring they get structured data with full IDs and parent_id
for proper threading.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* feat(daemon): differentiate comment-trigger vs assign-trigger context
When triggered by a comment, the agent now gets clear instructions:
- Primary goal is to read and respond to the comment
- Do NOT change issue status just because you replied
- Only change status if explicitly requested
This prevents the agent from seeing "In Review" and stopping, since
it now understands the task is to reply, not to re-evaluate the issue.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(daemon): split workflow by trigger type in CLAUDE.md/AGENTS.md
The Workflow section in the agent's runtime config now shows a
comment-reply workflow when triggered by a comment (read comments,
find trigger, reply, don't change status) vs the full assignment
workflow (set in_progress, do work, set in_review).
Previously the agent always saw the assignment workflow, causing it
to check the issue status, see "In Review", and stop without reading
or replying to the triggering comment.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* refactor(daemon): remove duplicate workflow from issue_context.md
Workflow instructions now live only in CLAUDE.md/AGENTS.md (runtime_config.go).
issue_context.md keeps just the task data: issue ID, trigger type, and
triggering comment ID.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(task): skip duplicate comment on completion for comment-triggered tasks
When triggered by a comment, the agent posts its own reply via CLI
with --parent. The task completion path was also creating a comment
from the agent's stdout output, resulting in duplicates. Now only
assignment-triggered tasks auto-post output as a comment. Error
messages from FailTask are still posted regardless of trigger type.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add redact package to detect and mask secrets (AWS keys, private keys,
API tokens, bearer tokens, credentials, home paths) in agent output
before posting as comments in TaskService
- Enforce agent visibility on issue assignment: private agents can only
be assigned by their owner or workspace admins
- Add visibility picker (workspace/private) to CreateAgentDialog,
default to private
- Grey out unassignable private agents in the assignee picker with
lock icon indicator
The backend already supports on_comment triggers but the frontend was
missing the UI to configure them. Adds the "On Comment" trigger type
and "Add On Comment" button alongside the existing On Assign and
Scheduled options.
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Show issue identifier and title instead of truncated UUIDs, sort active
tasks to top, highlight running/dispatched tasks, and show contextual
timestamps.
- Split 1189-line runtimes-page.tsx into focused sub-components (list, detail, ping, usage, 5 chart files, shared UI, utils)
- Add useRuntimeStore zustand store for shared runtime state across pages (agents page now uses it too)
- Add 7d/30d/90d time range selector to usage charts
- Add full-stack runtime delete: SQL query, Go handler, API route, frontend with confirmation dialog
- Remove unused daemon:heartbeat WS listener (server never broadcasts it)
Add an `instructions` text field to the agent model, allowing users to
define each agent's role, expertise, and working style. Instructions are
injected into CLAUDE.md as an "Agent Identity" section so the agent
knows who it is on every task execution.
- Migration 021: add instructions column to agent table
- Backend: create/update/get agent handlers support instructions
- ClaimTask response includes instructions for daemon injection
- execenv: inject instructions into CLAUDE.md meta-skill
- Frontend: add Instructions tab to agent detail panel
Refactor real-time sync from per-event precise mutations to WS-as-invalidation-signal + debounced refetch.
Backend:
- Add SubscribeAll to Event Bus — auto-broadcasts ALL events, eliminates manual 25-item allEvents list
- Add skill event constants to protocol, fix skill handler string literals
- Add title_changed activity tracking
Frontend:
- WSClient: add onAny() method for wildcard event subscription
- useRealtimeSync: rewrite to refreshMap + prefix routing + 100ms debounce
- Precise handlers only for side effects: workspace:deleted, member:removed, member:added (self-check)
- Reconnect now refetches all stores (fixes missing members/skills/workspace refresh)
- Stale-while-revalidate: fetch() only shows loading spinner on initial load, not on refetch
- Remove redundant useWSEvent in agents/page.tsx and skills-page.tsx
- WSClient.disconnect() now clears all handler registrations
Inbox bugfixes:
- Unify sidebar badge count with page count via dedupedItems + unreadCount in store
- Sort by time DESC (removed severity-first ordering)
- Ellipsis on truncated detail labels
UI:
- Status/Priority pickers: replace RadioGroup with MenuItem for auto-close on selection
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Runtime page: add ResizablePanelGroup with persistent layout, fix scroll
- Agents page: replace hand-rolled dropdowns with shadcn Popover/DropdownMenu,
remove redundant wrapper div, fix header height to h-12
- Skills page: widen create dialog to sm:max-w-md, stabilize tab height
- Settings: use variant="destructive" on AlertDialogAction instead of hardcoded className
- Issues list view: rewrite with base-ui Accordion grouped by status,
show all statuses (including empty), add per-group create button,
persist expand/collapse state, apply sort settings
- Issues header: show filtered issue count next to New Issue button
- Extract shared sortIssues utility from board-column for reuse
- Remove redundant StatusIcon from ListRow (already grouped by status)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add drag-to-resize sidebar with localStorage persistence
- Rewrite issue detail page with Tiptap rich text editor, due date picker, acceptance criteria
- Redesign create-issue modal with pill-based property toolbar and expand/collapse
- Consolidate @multica/sdk and @multica/types into apps/web/shared/
- Simplify auth: remove verification codes, PATs, email service (dev-only login)
- Add 401 unauthorized handler to redirect expired sessions to login
- Fix due date format to send full RFC3339 timestamps
- Increase description editor debounce to 1500ms
- Remove arbitrary Tailwind values in create-issue modal
- Renumber migrations (inbox_actor 012→009), remove unused migrations
- UI polish across agents, settings, inbox, knowledge-base pages
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Fix global scrollbar overflow by removing h-svh from html element
- Add h-full overflow-hidden to html/body for proper app-like layout
- Fix default button variant: add shadow-sm and hover:bg-primary/90
- Update sidebar create-issue button to bg-background with shadow
- Add WorkspaceAvatar component and search/new-issue actions to sidebar header
- Improve theme provider with TooltipProvider wrapper
- Polish various page layouts, pickers, modals, and code block styling
- Clean up custom.css unused styles
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Merge origin/main which added the skills system (structured skills
with meta skill runtime injection). Resolve 4 conflicts:
- workspace/store.ts: keep both skills state + issue/inbox fetch
- types/index.ts: keep Skill types + our event exports
- handler/agent.go: merge visibility filtering + skills batch loading
- pnpm-lock.yaml: accept main's lockfile with skills deps
Also fix skill.go: migrate h.broadcast → h.publish (event bus)
to match our architecture where all WS events go through the bus.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replace agent.skills TEXT field with structured skill/skill_file/agent_skill
tables. Skills are workspace-level entities with supporting files, reusable
across agents via many-to-many bindings.
Backend: migration 008, sqlc queries, CRUD handler, agent-skill junction,
structured skill loading in task context snapshot.
Daemon: meta skill injection via runtime-native config (.claude/CLAUDE.md
for Claude, AGENTS.md for Codex) so agents discover .agent_context/ skills
through their native mechanism. Lean prompt without inlined skill content.
Frontend: Skills management page, agent Skills tab picker, SDK methods,
TypeScript types, workspace store integration.
Also removes auto-creation of init issues when creating agents.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Migrate all shadcn components into apps/web/components/ui/ so the web app
is fully independent from packages/ui for its UI layer. Update to the
latest shadcn base-nova style. Add missing semantic color variables
(success, warning, info, canvas), font-mono mapping, scrollbar styling,
and wrap Select items in SelectGroup for proper padding.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Remove tab system entirely (tab-store, tab-bar, tab-link)
- Split monolithic AuthContext into zustand auth + workspace stores
- Move issue components/config to features/issues/
- Move WebSocket provider to features/realtime/
- Move api.ts to shared/
- Migrate all consumers from useAuth() to direct store imports
- Simplify sidebar: replace hand-built dropdown with shadcn DropdownMenu,
replace custom layout wrapper with SidebarInset
- Remove unused @multica/store and @multica/hooks dependencies
- Add @/ path alias and zustand dependency
- Update CLAUDE.md with feature-based architecture conventions
Net change: +293 / -2435 lines
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add success/warning/info semantic design tokens to globals.css
- Replace all raw HTML elements (input, select, textarea, button, label)
with shadcn components (Input, Select, Textarea, Button, Label, Dialog)
across settings, issues, agents, inbox, knowledge-base, and pair pages
- Replace all hardcoded Tailwind colors with design tokens
(text-red-500 → text-destructive, text-green-600 → text-success, etc.)
- Extract shared ActorAvatar component to packages/ui/components/common
- Update status and priority configs to use semantic tokens
- Update CLAUDE.md with component organization guidelines
- Fix login page tests to use label-based queries
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Complete agents management page with create dialog, runtime device selector,
skills/tools/triggers/tasks tabs, and agent detail view
- Add AssigneePicker to issue detail page for assigning to members or agents
- Extend agent types with description, skills, tools, triggers, RuntimeDevice
- Add SDK methods for agent CRUD and task listing
- Add migration 002 for agent config columns (skills, tools, triggers)
- Update seed data with realistic agent configurations
- Use auth context as single source of truth for agents (fixes state sync)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Fix kanban board columns not adapting to available width (w-64 → flex-1)
- Fix workspace name not updating in sidebar after save in settings
- Fix comments leaking across issues when navigating between issue details
- Fix duplicate issue appearing on create (race between callback and WebSocket)
- Add real-time WebSocket listeners for agents and inbox pages
- Add `make check` one-click verification pipeline (typecheck + tests + E2E)
- Add E2E test fixtures for self-contained test data setup/teardown
- Add settings E2E test and updateWorkspace unit test
- Make `make start/setup` reuse existing PostgreSQL if already running
- Update CLAUDE.md with AI agent verification loop and E2E test patterns
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Replace all mock data with real API calls across pages (issues, agents, inbox, settings)
- Add AuthProvider context with JWT login/logout, member/agent name resolution
- Implement login page with email-based auth flow
- Add settings page with workspace editing and member list
- Wire up real-time WebSocket for live issue updates
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Left column: agent list with avatar, name, runtime badge, status dot
- Right column: agent detail with overview, skills, connected tools, tasks
- Mock data: 4 agents (2 local, 2 cloud) with varied statuses
- Inspired by OpenAI Frontier's agent management UI pattern
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
