highperfocused/multica
1
0
Fork 0
mirror of https://github.com/multica-ai/multica.git synced 2026-07-24 19:41:14 +02:00
Code Issues Packages Projects Releases Wiki Activity
2,821 Commits 1,821 Branches 131 Tags
v0.2.26
Commit Graph

523 Commits

Author SHA1 Message Date
Jiayuan Zhang
e3dd31cbe5 feat(notifications): add system notifications toggle in settings (#2132) 
* feat(notifications): add system notifications toggle in settings

Add a per-user, per-workspace toggle to enable/disable native OS
notification banners. Reuses the existing notification-preferences
endpoint by introducing a `system_notifications` key alongside the
inbox event groups; the realtime handler reads the cached preference
and skips desktopAPI.showNotification when muted.

Co-authored-by: multica-agent <github@multica.ai>

* fix(notifications): fetch system_notifications pref lazily

Settings is the only mounted reader of notificationPreferenceOptions,
so a fresh app start (or any session that never visits Settings) left
the cache empty and the muted preference silently fell back to default
"all". Switch the inbox:new handler to ensureQueryData so the value is
fetched on first use and cached for subsequent events.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-05-06 11:01:22 +02:00
Naiyuan Qing
ba147708a6 fix(timeline): cursor-paginated timeline to stop long-issue freeze (#1968) (#2128) 
* fix(timeline): cursor-paginated timeline to stop long-issue freeze (#1968)

Opening an issue from Inbox with thousands of timeline entries used to
hard-freeze the browser tab on a synchronous render of every comment +
activity. The whole pipeline was unbounded: the API returned every row,
TanStack Query cached the full array, and IssueDetail mounted N
CommentCards (each running a full react-markdown + lowlight pipeline)
in one frame.

This swaps the timeline endpoint to keyset cursor pagination and rewires
the frontend to useInfiniteQuery so a long issue costs the same as a
short one on first paint.

API:
- GET /issues/:id/timeline now accepts ?before / ?after / ?around (mutex)
  + ?limit (default 50, max 100); response wraps entries with next/prev
  cursors and has_more flags. Cursors are opaque base64 (created_at, id).
- ?around=<entry_id> anchors a window on the target so Inbox notifications
  pointing at an old comment never trigger the freeze.
- New composite indexes on (issue_id, created_at DESC, id DESC) replace
  the redundant single-column ones so keyset queries are index-only scans.
- /issues/:id/comments default branch now caps at 50 instead of returning
  every row unbounded; the unbounded ListComments / ListActivities sqlc
  queries are deleted.

Frontend:
- useIssueTimeline switches to useInfiniteQuery, exposes
  fetchOlder/fetchNewer/jumpToLatest + isAtLatest + newEntriesBelowCount.
- WS handlers respect the at-latest invariant: comment/activity:created
  prepends to pages[0] only when the user is reading the live tail;
  otherwise it just bumps a counter so the UI offers a "Jump to latest"
  affordance without yanking scroll.
- Optimistic mutations adapted to the InfiniteData shape via shared
  helpers (mapAllEntries / filterAllEntries / prependToLatestPage in
  core/issues/timeline-cache.ts) and use setQueriesData so all open
  windows of the same issue stay in sync.
- IssueDetail Activity section gets a TimelineSkeleton placeholder
  during the brief load window plus subtle text-link load-more buttons
  matching the existing Subscribe affordance (no Button chrome). Top
  uses a divider for boundary clarity; bottom shows
  "Jump to latest · N new" weighted slightly heavier when there's
  unread state.
- highlightCommentId now flows into the hook's around parameter so
  Inbox jumps fetch the surrounding 50 entries directly.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* chore(agent): default comment list to 50 + prompt hint about long issues

The CLI's "multica issue comment list" used to default to --limit 0
(meaning "fetch every comment"), which lets an agent on a long issue
fill its context window with thousands of rows. The default is now 50;
agents that need older history can pass --limit or --since explicitly.

The local-coding-agent prompt also gains a single-line note about this
in both the comment-triggered and on-assign flows so the agent knows to
scope its fetches when issue size is unknown. Autopilot run-only mode
is intentionally unchanged — it has no issue context to query.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-06 16:27:06 +08:00
Naiyuan Qing
3447764b03 feat(i18n): full rollout — 21 namespaces translated (en + zh-Hans) (#1853) 
* feat(i18n): rollout phase — translate 9 namespaces (WIP)

Phase 1 complete (基建 + login + Settings language switcher),
phase 2 partial (Wave 4 done, search done). Pending namespaces
documented inline; another developer can pick up from here.

Infrastructure
--------------
- server: add users.language column + extend PATCH /api/me
  (TestUpdateMeAcceptsLanguage / TestUpdateMePreservesLanguage)
- packages/core/i18n: types / pickLocale (intl-localematcher) /
  browser-cookie-adapter / createI18n (initAsync false +
  useSuspense false) / I18nProvider / LocaleAdapterProvider
- Split server-safe vs React entries:
    @multica/core/i18n        — for proxy/RSC/middleware (no React)
    @multica/core/i18n/react  — for client trees (createContext)
  (RSC vendored React lacks createContext; mixed import would crash
  proxy.ts at module load.)
- packages/views/i18n: useT hook + selector API augmentation
  (i18next v26 default; auto-propagates to apps via the side-effect
  import in use-t.ts).
- apps/web: proxy.ts (Next 16 renamed middleware) merges existing
  legacy/root redirects with x-multica-locale header forwarding;
  layout.tsx reads locale via headers() and pre-loads RSC resources.
- apps/desktop: webPreferences.additionalArguments injects
  systemLocale (no sendSync — avoids main-thread blocking IPC);
  renderer adapter reads via process.argv.
- ESLint: i18next/no-literal-string at file-scope for translated
  files via packages/views/eslint.config.mjs TRANSLATED_FILES.
- glossary.md (packages/views/locales/) freezes term policy:
  Issue / Workspace / Agent / Skill / Autopilot / Daemon / Runtime
  stay English; Inbox / Project / Comment / Member translate.

Translated namespaces (9 / 19)
------------------------------
- auth: login page (web wrapper含 desktop-handoff 文案) + Settings
  Appearance language switcher
- editor: 9 .tsx (bubble-menu / link-hover-card / readonly-content /
  title-editor / extensions: code-block / file-card / image-view /
  mention-suggestion) + 32 keys
- invite: 25 keys
- labels / members / my-issues: Wave 4 全部
- search: command palette 35 keys
- navigation: no user-facing strings (no-op)

Pending (10 / 19)
-----------------
issues (46 files / ~210 keys)
agents (29 files / ~155 keys; presence.ts + config.ts label maps
  允许进 i18n)
onboarding (22 files / ~150 keys)
settings rest / skills / modals / workspace / chat / inbox /
projects / autopilots / layout

Workflow for picking up
-----------------------
- Glossary: packages/views/locales/glossary.md (mandatory read)
- Reference impls: auth/login-page.tsx + editor/* (selector API +
  i18n-provider test wrapper pattern)
- Per namespace:
    1. create locales/{en,zh-Hans}/{ns}.json
    2. add to packages/views/i18n/resources-types.ts
    3. useT('{ns}') + t($ => $.foo) in components
    4. add files to TRANSLATED_FILES in eslint.config.mjs
    5. typecheck + test + lint must pass
- Subagents currently CANNOT write files (sandbox deny). Run as
  hybrid: subagent researches + outputs full JSON + tsx diff,
  controller writes.

Other
-----
- scripts/init-worktree-env.sh: default
  MULTICA_DEV_VERIFICATION_CODE=888888 in dev for deterministic
  login (gated by isProductionEnv).

Verified: pnpm typecheck (6 pkgs ok), pnpm test (232 pass),
make test (Go).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* docs(i18n): rewrite glossary aligned with docs zh voice

Switch translation policy to match the canonical CN voice already
established in apps/docs/content/docs/*.zh.mdx (20+ files). The new
rule splits product nouns into two classes:

- Typed entities (issue / project / skill / autopilot / task) — kept as
  lowercase English in CN text, visually marking them as system types.
- Concepts (workspace / agent / daemon / runtime / inbox) — fully
  translated (工作区 / 智能体 / 守护进程 / 运行时 / 收件箱).

Previous glossary kept Workspace / Agent / Daemon / Runtime as English
on "工程惯例" grounds, but docs zh and CN AI ecosystem (Coze / 腾讯元器
/ 百度) consistently translate these. App UI now matches docs voice so
users don't see split personality between the app and its own docs.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(i18n): register 6 namespaces and retrofit zh strings to new glossary

Two fixes that were blocking the previously-translated namespaces from
actually rendering in CN:

1. RESOURCES gap — locales/index.ts only loaded common/auth/settings,
   but resources-types.ts declared 12 namespaces and 6 of them had real
   translation content. At runtime i18next would fall back to raw keys
   for editor / invite / labels / members / my-issues / search.
   Register all 9 currently-translated namespaces.

2. Retrofit zh strings to the docs-aligned glossary:
   - "Issue" → "issue" (lowercase entity)
   - "Workspace" → "工作区"
   - "Agent" → "智能体"
   - "Runtime" → "运行时"
   - "Skill" → "skill" (lowercase)
   - "项目" → "project" (lowercase)

Touched: editor.json (sub_issue + mention.group_issues), invite.json
(3 Workspace occurrences), members.json (agents_section / more_agents),
my-issues.json (8 retrofits across page/header/errors), search.json
(13 retrofits across groups/pages/commands/empty).

Verified: pnpm typecheck (6/6) + pnpm test (238/238) all green.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate inbox namespace

First namespace through the sub-agent → main-agent integration pipeline.

JSON: en/inbox.json + zh-Hans/inbox.json — 60 keys across page / menu /
list / detail / types / labels / errors. Time-formatter labels are kept
compact in EN ("5m" / "3h" / "2d") and use full units in zh ("5 分钟" /
"5 小时" / "5 天") since raw "5 分" reads as "5 marks/points" in CN.

Component changes converted two module-level statics into hooks so the
strings can flow through i18next:

- inbox-list-item.tsx: `timeAgo` (pure fn) → `useTimeAgo` (hook
  returning a fn). The local copy is a duplicate of @multica/core/utils
  `timeAgo` that is only used by inbox-page; other consumers across
  chat/agents/skills/issues stay on the core util for now and will be
  translated when their namespaces land.

- inbox-detail-label.tsx: `typeLabels` (static const Record) →
  `useTypeLabels` (hook returning the same Record shape). Call sites
  keep the existing `typeLabels[type]` access pattern.

inbox-page.tsx now uses both hooks and `useT('inbox')` selector calls
for all hardcoded strings (~24 sites: header / dropdown menu / list
empty state / detail panel / mobile back / quick-create-failed flow /
all error toasts).

Wired up: resources-types.ts, locales/index.ts RESOURCES, ESLint
TRANSLATED_FILES (3 inbox tsx files now lint-protected).

Verified: pnpm typecheck (6/6) + pnpm --filter @multica/views test
(238/238) + ESLint clean on inbox/.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate workspace namespace

Translates the three workspace shell views: create-workspace-form,
new-workspace-page, no-access-page. Also fixes the prior-art
no-unescaped-entities lint errors in no-access-page.tsx — the
apostrophes in "doesn't" / "don't" were JSX text literals that move
into JSON values after translation, so the lint rule no longer fires.

Tests wrapped: workspace/create-workspace-form.test.tsx,
workspace/no-access-page.test.tsx, modals/create-workspace.test.tsx
all now wrap render() with <I18nProvider locale="en"> so the en values
in workspace.json drive the rendered text and the existing assertions
continue to match.

Slug constants kept: WORKSPACE_SLUG_FORMAT_ERROR /
WORKSPACE_SLUG_CONFLICT_ERROR exports in workspace/slug.ts are still
imported by onboarding/steps/step-workspace.tsx (out of scope here).
The workspace shell now reads its strings from workspace.json directly.

Multica.ai brand prefix in the slug input affordance is wrapped with
an inline `// eslint-disable-next-line i18next/no-literal-string` per
glossary policy on brand names.

Renamed sign_in_other → sign_in_different to avoid colliding with
i18next's `_other` plural-suffix convention which the selector-API
typings treated as a plural form of `sign_in`.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate projects namespace

Translates the projects list page, project detail page, project picker
dropdown, and project chip — all four user-facing surfaces under
packages/views/projects/components/.

New file: projects/components/labels.ts exposes three hooks that
replace the static `.label` field on PROJECT_STATUS_CONFIG /
PROJECT_PRIORITY_CONFIG and the previous module-level
`formatRelativeDate` helper. Core's `.label` stays untouched (it's
still consumed by search and the create-project modal, both
out-of-scope for this namespace) — those will flip when their
respective namespaces translate.

In zh, the "project" entity stays lowercase English per glossary
(`新建 project`, `还没有 project`, `从 project 移除`). Status / priority /
table column labels translate fully.

The cancelled / done / paused etc. status labels duplicate per-
namespace as `projects.status.*` rather than reading from a future
shared status namespace. This matches the auth/inbox/workspace
pattern of self-contained namespaces. If a generic "issue/project
status" pool emerges later, these can collapse.

Verified: pnpm --filter @multica/views typecheck (clean) +
test (238/238) + ESLint clean on projects/ (1 pre-existing warning
about useEffect/sidebarRef dep, unrelated to i18n).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate autopilots namespace

Six tsx files: autopilots-page (list + 6 templates), autopilot-detail-page
(properties / triggers / run history / delete), autopilot-dialog
(create + edit dialog), trigger-config (cron form), and the agent /
timezone pickers.

Hook conversions for module-level helpers that need t():
- summarizeTrigger / describeTrigger → useSummarizeTrigger /
  useDescribeTrigger (no external callers, removed the plain exports)
- formatRelativeDate → useFormatRelativeDate (per-component hook)
- formatCountdown → useFormatCountdown (per-component hook)
- TEMPLATES array now keyed by id; titles + summaries pull from
  templates/{id}/{title,summary} JSON. Prompts stay raw EN since
  they're injected directly into the agent task — translating them
  would translate the agent's instructions, not the user's UI.

Status / execution-mode / run-status enums render via t($ => $.status[k])
with k typed against the core type (no separate hook needed).

Verified: pnpm --filter @multica/views typecheck (clean) +
test (238/238) + ESLint clean on autopilots/.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate skills namespace

Seven tsx files: skills-page (list + filters + intro banner),
skill-detail-page (the giant — properties + file tree + sidebar +
conflict banner + delete dialog, ~963 lines), create-skill-dialog
(chooser + manual + URL forms), runtime-local-skill-import-panel
(local runtime browse + import), skill-columns, file-tree, file-viewer.

Notable patterns:
- `createSkillColumns` factory → `useSkillColumns` hook so column
  headers flow through useT. Column identity changes per render is
  fine — DataTable handles it.
- `validateNewFilePath` (pure helper) → `useValidateNewFilePath` hook
  so the 5 validation error messages can be translated.
- skill_files / used_by / description_with_agents use i18next plural
  keys (`_one` / `_other`) — the type system collapses these into a
  single PluralValue access, so call sites use
  `t($ => $.foo, { count })` and i18next picks the form.
- Per glossary, "skill" stays lowercase EN in zh ("新建 skill",
  "已删除 skill", "未找到该 skill").

Test wrapper: runtime-local-skill-import-panel.test.tsx now wraps
render() with <I18nProvider> so the assertion on /Import to Workspace/i
matches the EN translation.

Verified: pnpm --filter @multica/views typecheck (clean) +
test (238/238) + ESLint clean on skills/.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate chat namespace

Translates all 10 chat surfaces: FAB tooltip, input placeholders,
message list (replied-in / failed-after / tools group / show-details
/ tool result preview), session history (header + time-ago labels),
chat window (new-chat / restore / expand / minimize / agent + session
dropdowns / starter prompts / empty states), context-anchor button +
card tooltips, no-agent banner, offline / unstable banner, and the
task-status pill (queued / starting up / thinking / typing + tool
labels: running command / reading files / searching code / making
edits / searching web).

Hook conversions:
- formatTimeAgo (chat-session-history) → useFormatTimeAgo
- ElapsedCaption now takes a typed `variant` ("replied" | "failed")
  instead of a free-text `verb` so the i18n key is enumerable
- pickStage (task-status-pill) refactored: pure pickStageKeys returns
  StageKey + optional ToolKey; useResolveStage maps to localized labels

Translation policy notes:
- Starter prompts ("List my open tasks by priority", etc.) are user
  UI when displayed AND the user's input when clicked — translating
  them sends the agent the user's locale-native phrasing, which is
  the right UX for a CN user using a CN agent.
- buildAnchorMarkdown (chat-window) stays in English: it's an
  agent-bound markdown prefix injected into the outgoing message,
  not user-facing UI.

Verified: pnpm --filter @multica/views typecheck (clean) +
test (238/238).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate modals namespace

Translates all 11 modal sources: registry (no UI text), backlog-agent-hint,
set-parent-issue, add-child-issue, delete-issue-confirm, feedback,
issue-picker, create-workspace, create-project, create-issue (manual),
quick-create-issue (agent panel).

Notable patterns:
- create-project re-uses useProjectStatusLabels / useProjectPriorityLabels
  hooks from views/projects/components/labels — same translation source
  as the projects list / detail, no duplication.
- create-issue.tsx: renamed `toast.custom((t) => ...)` callback param to
  `toastId` to avoid shadowing the closure-captured useT() `t` function.
- Test wrapper added to modals/create-issue.test.tsx so the two assertions
  on rendered modal text (success toast + Create another) match the EN
  bundle. modals/create-workspace.test.tsx was already wrapped (workspace
  ns commit).

Verified: pnpm --filter @multica/views typecheck (clean) +
test (238/238).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate settings namespace (rest of tabs)

Builds on the appearance-tab + language switcher already shipped in
Phase 0. Translates the remaining 8 settings surfaces: settings-page
shell (left nav + tab keys), account / profile, notifications-tab
(5 group labels + descriptions), tokens-tab (create / list /
revoke / created dialog), workspace-tab (general fields + danger
zone + leave/delete confirmations), members-tab (invite + role
config + revoke / remove flows), repositories-tab, labs-tab,
delete-workspace-dialog.

Hook conversion: members-tab `roleConfig` static const → `useRoleLabels`
hook returning a Record<MemberRole, {label, description, icon}>. The
icon stays as a typed React component (Crown / Shield / User), so
rendering pattern is unchanged at call sites.

Test wrapper: settings/components/delete-workspace-dialog.test.tsx
now wraps render() with <I18nProvider> (custom render() helper)
because the test asserts on rendered button labels ("Delete workspace",
"Cancel", "Deleting...").

Verified: pnpm --filter @multica/views typecheck (clean) +
test (238/238).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate runtimes namespace (entry surfaces)

Translates the user-facing runtime list page surfaces:
runtimes-page (header / search / filters / chips / empty / no-matches /
bootstrapping), runtime-detail (topbar + delete dialog + delete toasts),
runtime-detail-page (not-found state), shared.tsx (4-state HealthBadge
labels).

Hook conversion: shared `healthLabel(health)` was a pure module-level
function. Added `useHealthLabel` hook for translated call sites; kept
`healthLabel` as an EN-only fallback for non-component callers (column
factory in runtime-columns).

Deferred:
- runtime-list / runtime-columns (data table column headers + cell
  bodies) — large surface, not in the page-load critical path.
- connect-remote-dialog / update-section / usage-section — secondary
  flows, English remains acceptable until a focused pass.
- charts/* — primarily numeric tooltips and axes; minimal user-visible
  text.

Verified: pnpm --filter @multica/views typecheck (clean) +
test (238/238).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate layout namespace (sidebar nav, help, loader)

Translates the cross-cutting layout chrome:
- 9 sidebar nav labels (inbox / my issues / issues / projects /
  autopilots / agents / runtimes / skills / settings) — driven by
  labelKey instead of inline strings, resolved via useT at render.
- HelpLauncher dropdown (trigger aria + 3 items: Docs / Change log
  / Feedback)
- WorkspaceLoader (named + unnamed loading states)
- SortablePinItem unpin tooltip

Pattern shift in app-sidebar.tsx: nav arrays carry `labelKey: NavLabelKey`
(typed against the layout JSON) instead of `label: string`. The string
comparison checks (`item.label === "Inbox"`) became cleaner ID-based
checks (`item.key === "inbox"`).

Deferred: deeper sidebar surfaces — workspace switcher dropdown,
"New Issue" CTA, "Pinned" / "Workspace" / "Configure" group labels —
remain English. The 9 nav labels are the ones that read in every
session.

Verified: pnpm --filter @multica/views typecheck (clean) +
test (238/238).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate onboarding namespace (welcome + step header)

Translates the user-first-impression surfaces of the onboarding flow:

- step-welcome.tsx (the wordmark, headline, lede paragraphs, all CTAs:
  Download Desktop / Continue on web / Start exploring / I've done
  this before, illustration caption)
- step-header.tsx ("Step N of M" counter + matching aria-label)
- onboarding-flow.tsx (skip-onboarding error toast)

Test wrapper added to onboarding/components/step-header.test.tsx —
custom render() helper wraps with <I18nProvider> so the "Step 2 of 5"
assertions match the EN bundle.

Deferred (acceptable English fallback for now): step-questionnaire,
step-workspace, step-runtime-connect, step-platform-fork, step-agent,
step-first-issue, cli-install-instructions, option-card, runtime
aside panels, starter-content-prompt, cloud-waitlist-expand. These
are deeper steps with significant copy that would benefit from a
focused dedicated pass — voice on each is more nuanced (questionnaire
options, runtime install instructions, agent template recommendations).

Verified: pnpm --filter @multica/views typecheck (clean) +
test (238/238).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* test(i18n): add EN/zh-Hans key parity guard

Schema-level vitest that walks RESOURCES.en and RESOURCES["zh-Hans"]
namespace by namespace and asserts both bundles cover the same key
set. i18next plural rule is normalized before compare (`_one` /
`_other` collapse to a single logical key) so EN's plural pair
matches zh's `_other`-only form.

Catches retrofit drift where a new EN key lands without zh —
previously this would silently fall back to the English string in
production. Cheap to keep green: 39 tests across 21 namespaces in
under a second.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate issues namespace

Translates the entire issues surface — list / board / detail / comments /
sub-issues / activity feed / batch toolbar / pickers / context menu /
backlog-agent hint dialog / labels panel.

Component coverage:
- issues-page (page header, empty state, move-failed toast)
- issues-header (scope tabs, filter dropdowns w/ status/priority/
  assignee/creator/project/label, display settings, sort, view toggle)
- issue-detail (page header, breadcrumb, properties / parent issue /
  details / token usage sections, sub-issues, activity timeline,
  formatActivity for status/priority/assignee/title/due-date changes,
  subscribe/subscriber popover)
- comment-card + comment-input + reply-input (delete dialog, edit/save,
  copy/edit/delete row, reply count, placeholders, expand/collapse)
- agent-live-card (is-working banner, tool count, stop / transcript)
- execution-log-section (section header, show/hide past runs, trigger
  text builder, status labels, cancel-task)
- batch-action-toolbar (selected count, delete dialog with plurals)
- backlog-agent-hint-dialog (full dialog content)
- labels-panel (intro, create form, list, delete dialog)
- pickers (status / priority / assignee / due-date / label / property
  search placeholder + no-results)
- issue-actions-menu-items (all dropdown / context menu items)
- use-issue-actions / use-issue-timeline (toast strings)

STATUS_CONFIG / PRIORITY_CONFIG label rendering routed through
$.status[enum] / $.priority[enum] at every call site; the core config
keeps its English fallback for non-i18n consumers but UI never reads
.label directly anymore.

Tests retrofitted: issues-page, issue-detail, and issue-actions-menu
RTL specs now wrap renders in <I18nProvider> with the EN bundle, so
their string assertions match the bundle (not hardcoded literals).

ESLint i18next allow-list extended to 24 issues files. Verified:
pnpm --filter @multica/views typecheck + test (277/277) all green.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate agents namespace

Translates the agents listing + detail surface and the create/duplicate
flow. Covers the high-frequency surfaces; deeper sub-tab editors
(activity / instructions / skills / env / custom-args bodies, and the
hooks-buggy runtime/model/concurrency pickers) are deferred — they
have their own pre-existing react-hooks rule violations and benefit
from a focused dedicated pass.

Component coverage:
- agents-page (page header w/ tagline + new button, scope segment,
  search, sort dropdown, availability chips, archived toolbar, empty
  state, no-matches messaging w/ search interpolation, list-load
  error)
- agent-detail-page (back link, archived banner, archive dialog,
  not-found state, all 4 toast strings)
- agent-detail-inspector (avatar editor, name + description popover,
  description dialog, every PropRow label, validation message,
  presence badge label sourced from $.availability[enum])
- agent-overview-pane (tab labels, discard-unsaved-changes dialog)
- create-agent-dialog (title / description / labels / placeholders /
  duplicate-suffix / runtime filter buttons / runtime status copy)
- agent-row-actions (full dropdown items + cancel-tasks dialog with
  pluralized "N running + M queued" summary + archive dialog + 6 toasts)
- agent-columns (every header cell, You / Archived chips, runtime
  fallback labels, availability + workload labels via $.availability /
  $.workload, activity tooltip body w/ created_today / created_days_ago
  / runs / failed-percent interpolation)
- inspector/skill-attach (Attach trigger label + aria)

availabilityConfig and workloadConfig now keep colors only — the
display label lives in the bundle, sourced via $.availability[enum]
and $.workload[enum] at every call site. Same pattern as
STATUS_CONFIG/PRIORITY_CONFIG in the issues namespace.

ESLint i18next allow-list extended to 8 agents files.
Verified: pnpm --filter @multica/views typecheck + test (277/277)
all green.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(i18n): clear 30 stray EN strings in translated files

Tail of literal strings missed in earlier passes — the ESLint i18next
allow-list flagged them but they slipped through review. Files touched:

- layout/app-sidebar.tsx (10 keys: Workspaces / Pending invitations /
  Create workspace / Join / Decline / Log out / New Issue + shortcut /
  Pinned / Workspace / Configure)
- runtimes/components/runtime-detail.tsx (Serving header + serving_count
  pluralization, no_agents copy, running/queued chips with count
  interpolation, Diagnostics header, CLI label, Delete runtime button,
  Technical details toggle, last seen interpolation)
- onboarding/steps/step-welcome.tsx (entire WelcomeIllustration mock —
  5 cards × actor names + body copy + 3 mention chips + 2 timestamps;
  zh translation reads naturally instead of leaving the demo English)
- settings/components/labs-tab.tsx (`Co-authored-by: ...` git trailer
  wrapped in {} so linter sees a JS string, not JSX text — magic
  identifier git relies on, must not translate)
- settings/components/members-tab.tsx (✓ glyph wrapped in {})
- modals/feedback.tsx (⌘↵ shortcut wrapped in {})

ServingAgentsCard now reads availability/workload labels from
`agents` namespace (cross-namespace useT) so the bundle-truth pattern
holds: presenceConfig keeps colours only, label text comes from the
shared bundle.

Verified: typecheck + 277/277 tests + lint (only the pre-existing
react-hooks rule-of-hooks errors remain, which task #6 addresses).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(agents): rules-of-hooks + translate 4 model/runtime pickers

Three pre-existing react-hooks/rules-of-hooks violations + one missing
useMemo dep cleared, then the four pickers wired through useT.

Hook order fixes:
- concurrency-picker: useEffect now runs before the !canEdit early
  return. Stale-draft reset still works the same way.
- runtime-picker: useMemo for the filtered list moved above the
  !canEdit branch.
- model-dropdown: `models = data?.models ?? []` was minting a fresh
  array each render and tripping the deps lint of the downstream
  useMemo. Wrap in useMemo so the reference is stable.

Translation coverage:
- concurrency-picker: tooltip ("Concurrency · N max..."), range
  helper text, Save button.
- runtime-picker: trigger label fallback ("No runtime"), tooltip
  text composed from {{name}} + status, Mine/All filter buttons,
  empty-list copy, "owned by {{name}}" + status fragments in row
  tooltip, Cloud badge, online/offline aria.
- model-picker: trigger label, tooltip, "Managed by runtime"
  fallback, search placeholder, "Discovering models…", default
  badge, "No models available", "Use \"X\"" custom-id flow, Clear
  button + its title.
- model-dropdown: every label string including the "Select a runtime
  first" / "Default (provider)" / "Runtime offline — enter manually"
  trigger fallbacks, the supported=false explanation block, discovery
  failed badge, all popover items.

ESLint allow-list extended to 4 picker files. Verified: typecheck +
277/277 tests + lint (0 errors, only pre-existing react-hooks warnings
in unrelated files).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate runtimes list + connect dialog + CLI updater

Three deep runtime surfaces wired through useT, with the agents
namespace doing double duty for shared availability/workload labels.

runtime-columns:
- 7 column headers via t-augmented createRuntimeColumns({ t }).
- HealthCell now reads from useHealthLabel() (already translation-aware)
  instead of the EN-only healthLabel() helper.
- WorkloadCell sources the label from $.workload[enum] (cross-namespace
  to agents) — colour stays via workloadConfig.
- CostCell delta "flat" copy + CLI cell "Desktop" badge + update-
  available aria/tooltip + RowMenu's full delete dialog (title /
  description with {{name}} interpolation / cancel / confirm /
  deleting state) plus its admin-permission hint.

connect-remote-dialog:
- Three steps fully translated: instructions (header + 4 numbered
  steps + security warning + troubleshooting list with mono code
  snippets escaped as JS strings), waiting (loader + hint), success
  (CTA pair).
- Mono CLI commands wrapped in {} so linter sees JS strings — those
  are literal commands that must stay untranslated for the user to
  paste into a terminal.

update-section:
- statusConfig collapsed to icon+colour only; labels move to
  $.update.status[enum] for proper translation per-state.
- "CLI Version:" / "Latest" / "available" / "Update" / "Retry"
  copy + the "Managed by Desktop" tooltip and disabled hint.

Layout helpers tagged: runtime-list passes `t` through to the column
factory the same way agent-columns does.

ESLint allow-list extended with the 4 wired files. Verified:
typecheck + 277/277 tests + 0 i18n lint errors. usage-section.tsx
(KPI cards / WhenChart / TopUsageBreakdown / receipt table) is the
remaining runtimes surface — chart-heavy and benefits from a focused
pass next.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate 5 agent detail tabs + skill-add dialog

The 5 tabs that fill the agent detail right pane plus the shared
skill picker dialog. Agents bundle gains a `tab_body` block with
sub-namespaces per tab + a `common` slot for save/add/unsaved.

Tab coverage:
- instructions-tab: intro paragraph, multi-line example placeholder
  (full 18-line zh translation), Save / Unsaved.
- env-tab: read-only intro / empty state, editable intro with two
  inline `<code>` env-var examples kept English (mono terminal
  payloads), KEY / value placeholders, Show/Hide value aria, Add /
  Remove aria, all 3 toasts (duplicate keys / saved / save failed).
- custom-args-tab: intro about whitespace splitting, launch-mode
  prefix line + `<your args>` placeholder, --flag value placeholder,
  Add, Remove aria, both toasts.
- skills-tab: intro, Add skill button, import-hint callout, empty
  state title + hint + add-CTA, remove-failed toast.
- activity-tab: 3 section titles (Now / Last 30 days / Recent work),
  active-task pluralization, performance subtitle, all 3 empty
  states, runs/success%/avg-duration/failed pluralization with
  interpolation, source labels (Issue / Chat / Autopilot / Untracked),
  source fallbacks (Quick create / Creating issue / Chat session /
  Autopilot run), issue-short fallback, "Triggered by" tooltip
  header, open-issue / transcript / cancel-task tooltips and ARIAs,
  cancelling state, started/dispatched/queued time prefixes, show
  more.
- skill-add-dialog: dialog title + description, empty list copy,
  Cancel button, add-failed toast.

skills-tab.test.tsx wrapped in <I18nProvider> with the EN bundle so
its `Local runtime skills are always available` assertion still
matches the resolved translation instead of the raw key path.

ESLint allow-list extended with the 6 wired files. Verified:
typecheck + 277/277 tests + 0 i18n lint errors. Only the per-test
mock for skills-tab needed wrapping; the other 4 tabs ship without
test files of their own and inherit the I18nProvider chain via
agent-overview-pane / agent-detail-page test renders (when those
exist later).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate onboarding step-questionnaire + option-card

The user-profile step (3 questions) is the first deferred onboarding
deep step now wired through useT.

step-questionnaire:
- Eyebrow + headline + answered-progress counter with {{count}}
  interpolation
- All 3 questions and their option labels (team size / role / use case)
- All 3 "Other" placeholders for free-text fallback
- Right-rail "Why three questions" / "What you get" panel: 2 eyebrow
  rows, 2 unlock-item title+body pairs, learn-more link
- Back / Continue buttons via shared `common` block

option-card: shared "Other" radio label and aria.

Test wrapped in <I18nProvider>. EN value of `other_label` kept as
"Other" so the existing /^other$/i regex in step-questionnaire.test
keeps matching after the rendering pipeline switched from a hardcoded
literal to a bundle lookup.

ESLint allow-list extended with these 2 files. The remaining 4 deep
steps (workspace / runtime-connect / platform-fork / agent), the
2 ancillary surfaces (cli-install-instructions / starter-content-
prompt), and the 3 side panels (runtime-aside-panel / cloud-waitlist-
expand / compact-runtime-row) will be surfaced + swept by the global
ESLint switch (next commit).

Verified: typecheck + 277/277 tests + 0 i18n lint errors.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): flip ESLint to glob + drain remaining hardcoded EN

ESLint i18next/no-literal-string now applies to **/*.tsx by default
instead of an explicit allow-list. Files that genuinely still need
hardcoded EN are listed in STILL_HARDCODED — concrete, finite, and
the goal is to drain that list to zero.

Tail strings translated in this commit (surfaced by the global flip):

- common/task-transcript/agent-transcript-dialog.tsx — full dialog:
  status badge (Running / Completed / Failed), sr-only DialogTitle,
  Filter dropdown trigger + Clear filters, Copy all / Copy filtered /
  Copied, tool-calls + events metadata chips with pluralization,
  events-filtered "{{shown}} of {{total}}" interpolation, "Waiting
  for events..." live state, "No execution data recorded." past
  state. New `transcript` block in agents namespace.
- runtimes/components/charts/activity-heatmap.tsx — Less / More
  legend labels around the contribution-style heat squares.
- search/search-trigger.tsx — sidebar Search... button label.
  ⌘ glyph wrapped in {} to satisfy the linter (mono shortcut symbol,
  not translatable).

Holdouts (STILL_HARDCODED, ~14 files): the deep onboarding steps
(workspace / runtime-connect / platform-fork / agent / first-issue /
cli-install-instructions, plus 4 ancillary panels), the runtimes
usage-section + KPI cards, and 5 minor agent visual primitives
(sparkline / agent-presence-indicator / agent-profile-card /
visibility-badge / char-counter). Each one gets a dedicated future
pass; the global rule prevents new hardcoded strings from landing
elsewhere.

Verified: typecheck + 277/277 tests + 0 i18n lint errors.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): drain agent visual primitives + onboarding small components

8 files removed from STILL_HARDCODED:

agents/components/:
- char-counter — over-limit text with {{count}} interpolation
- visibility-badge — uses new agents.visibility.{private,workspace}.
  {label,tooltip} block; drops VISIBILITY_LABEL/TOOLTIP imports from
  core in favour of bundle-driven copy
- agent-presence-indicator — availability + workload labels via
  $.availability[enum] / $.workload[enum] (cross-namespace),
  queue-badge "+N queued" with pluralization
- agent-profile-card — Agent unavailable / Detail link / Owner /
  Skills / Runtime / Unknown runtime / Archived chip / availability
  line via cross-namespace lookup

agents.json: new presence + visibility + profile_card + char_counter
blocks.

onboarding/components/:
- compact-runtime-row — online/offline aria via agents.availability
- runtime-aside-panel — full content (What's a runtime / Good to
  know / Swap anytime / Add more later / docs link)
- starter-content-prompt — full dialog (title / description with
  inline emphasis / both buttons / 3 toasts)
- cloud-waitlist-expand — intro paragraph + warning span / email
  + reason labels + placeholders + Optional badge / Join + on-list
  states / both toasts

onboarding/steps/:
- cli-install-instructions — copy aria + intro + 2 step labels

onboarding.json: new runtime_aside / cli_install / starter_content /
cloud_waitlist blocks.

Tests for step-platform-fork + step-runtime-connect wrapped in
<I18nProvider> with EN bundle so /you're on the list/i etc. still
matches the resolved translations.

Verified: typecheck + 277/277 tests + 0 i18n lint errors.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate onboarding deep steps

The 5 large onboarding steps that were deferred from earlier passes,
plus their support helpers, all wired through useT.

step-first-issue (final beat — flips onboarded_at):
- error_title / Retry / retry_failed toast / finishing / opening
  states.

step-agent (creates the user's first agent):
- Templates moved from a module-level const to a useT-driven
  useAgentTemplates() hook. Names + emoji stay constant (visual
  identity), labels + blurbs + instructions resolve from the
  bundle. coding / planning / writing / assistant — all four
  templates ship a full zh translation that reads naturally.
- Recommended badge, eyebrow + headline + lede, footer hint,
  Create {{name}} CTA, create_failed toast.
- Right-rail "About agents" panel (4 way-items + headline +
  add-more hint + docs link).

step-workspace (create or pick existing):
- 5 footer states (open / creating / creating-pending / name-first
  / pick), all hint + CTA strings via interpolation.
- Name + URL + slug placeholders, issue-prefix preview spans,
  Create-new card title + subtitle.
- 8-row WorkspacePreviewCard sidebar (Inbox / Issues / Agents /
  Projects / Autopilot / Runtimes / Skills / And more) — every
  label + meta strapped to bundle keys.
- 4 perks (assign / chat / invite / switch) + 3 next-steps
  (runtime / agent / starter), 2 toasts (slug-conflict / failed).
- `multica.ai/${slug}` mono URL escaped via template-literal
  expression so the linter sees a JS string.

step-runtime-connect (desktop scan flow):
- 3 phase headlines + ledes (scanning / found / empty), trust-strip
  status (all online / N online / none online) with pluralization,
  online/offline labels, Skip / Continue / Selected hint.
- Empty-view 2 cards (skip + waitlist) and the cloud waitlist
  dialog wrapper.

step-platform-fork (web fan-out):
- Eyebrow + headline + lede, footer hint with 3 phase variants.
- Primary download card (before/after click) + 2 alt cards (CLI /
  cloud) + CLI dialog with 4 elapsed-time stages (normal / midway /
  slow / stalled), live-listening header, runtime-connected
  pluralization, cloud waitlist dialog.

ESLint: STILL_HARDCODED list shrunk from 14 entries to 1 — only
runtimes/components/usage-section.tsx (chart-heavy KPI panel)
remains.

Verified: typecheck + 277/277 tests + 0 i18n lint errors.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate runtimes usage panel + drop STILL_HARDCODED

Final i18n holdout: the runtimes usage panel (KPI hero, WHEN chart
tabs, cost-by breakdowns, daily breakdown table) is wired through
useT("runtimes"). With this drained, the eslint scaffolding for
explicit holdouts is removed — every JSX text node in @multica/views
now flows through i18n.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(i18n): drain rollout gaps + add cross-device sync

Lands the post-review punch list for the i18n rollout: closes correctness
gaps that would have shipped silently, and adds the missing cross-device
locale sync the rollout's docs already promised.

Coverage:
- Register issues + agents namespaces in RESOURCES (90 useT call sites
  were rendering keys-as-text in production)
- Harden parity test to compare RESOURCES keys against on-disk JSON
  files, so a future missing namespace registration fails loudly
- Server-side language whitelist in UpdateMe + reject-unsupported test
- Safe SupportedLocale resolution in appearance-tab (no more `as` cast
  on a region-tagged BCP-47 string)
- HTML lang attribute uses zh-CN (not zh-Hans) for screen reader / CJK
  font-stack compatibility
- Cookie Secure flag on https
- Pulled createBrowserCookieLocaleAdapter out of the server-safe entry
  into a new @multica/core/i18n/browser subpath; document.cookie access
  can no longer leak into Edge middleware imports

Cross-device sync:
- New UserLocaleSync component mounted in CoreProvider; on login, if
  user.language differs from the active i18n.language, persist via the
  adapter and reload. Both apps benefit
- Desktop main process tracks system locale and emits IPC on focus when
  it changes; renderer reloads only when the user has no explicit
  Settings choice (their preference still wins)

Tests:
- pickLocale / matchLocale (11 cases incl. region-tagged BCP-47, malformed
  tags, zh-Hant collapse-to-zh-Hans semantics)
- browser-cookie-adapter (6 cases under jsdom)
- Shared renderWithI18n helper at packages/views/test/i18n.tsx that wraps
  the real RESOURCES map; future tests opt in instead of inlining a
  per-file TEST_RESOURCES slice that goes stale silently

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* docs(conventions): consolidate naming + i18n glossary into docs site

Single source of truth for code naming, i18n translation glossary, and
Chinese voice rules. Previously split between packages/views/locales/glossary.md
and scattered comments — now lives at apps/docs/content/docs/developers/conventions.{mdx,zh.mdx}
with both English and Chinese versions kept in sync.

Three sections per page:
1. Code naming — routes, packages, files, DB, Go, TS, commits
2. i18n translation glossary — entity vs concept rule, what to translate,
   word combination, plurals, interpolation, key naming
3. Chinese voice + style — punctuation, principles, where to look in doubt

Side effects:
- packages/views/locales/glossary.md collapses to a stub redirecting to
  the docs page; do not edit it
- CLAUDE.md gets a new top-level "Conventions reference" section so any
  Claude session sees the pointer before any other rule
- apps/docs/content/docs/developers/ gets a stub English meta.json so the
  conventions page is reachable on the EN side (contributing.zh.mdx /
  architecture.zh.mdx remain ZH-only — separate work)
- Both root sidebars get a new "Developers" group

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(i18n): apply zh voice rules + translate project/autopilot

Two-part cleanup driven by the conventions doc landed last commit:

Voice violations (mechanical sweep across 10 zh-Hans namespaces):
- 「」 (Japanese-style brackets) → \" to match the EN source's straight
  double quotes (~13 sites)
- … (single-char ellipsis) → ... three dots (~43 sites)
- Drop translation-ese pronoun "我们" where it's a pure narrator
  ("我们已发送" → "已发送", "我们替你托管" → "由 Multica 托管"); keep
  "告诉我们" where "we" is the legitimate brand recipient
- "作为父级 / 作为子级" → "设为父级 / 设为子级"
- "任务" mistranslated as the task entity → `task` (lowercase EN entity)
- Dialog title "Autopilot" → "autopilot"

Translate project / autopilot per industry consensus:
- `project` → 「项目」 (~42 value sites). Feishu / Tower / Teambition /
  PingCode / GitHub Projects all translate; no Chinese product keeps
  `project`.
- `autopilot` → 「自动化」 (~34 value sites). Avoids the Tesla-style
  「自动驾驶」 association; matches Notion / Feishu's industry term.
- Issue / skill / task remain lowercase EN per dev-team familiarity.
- Sidebar nav-label entities get Title Case ("Issue" / "Skill" / "我的
  Issue") so the entry-point label reads as a proper UI signal; body
  prose stays lowercase.

Conventions doc (EN + ZH) reflects the decision and adds a "why these
translate but issue/skill/task don't" rationale block.

Verification: parity test 45/45, full monorepo typecheck green.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat(i18n): translate chat session delete + project resources section

Two features main shipped while this branch was idle never went through
the i18n pass:

- Chat session delete confirmation dialog (#2115) and history toggle
  tooltip (#2117): adds session_history.delete_dialog.* and
  session_history.row_delete_*, plus window.history_show_tooltip /
  history_back_tooltip.
- Project resources sidebar (#1926/#2080/#2111): entire component
  including toasts, popover form, attach/remove tooltips. New
  projects.resources subtree.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-06 16:16:12 +08:00
Bohan Jiang
ae985ae2a3 fix(daemon): tighten 404 task-not-found semantics — server + final guard (#2127) 
* fix(server): return 500 for transient DB errors in daemon task lookup

requireDaemonTaskAccess used to turn any GetAgentTask error into
404 "task not found", including transient DB connection / pool errors.
Combined with PR #2107 — which added 404+"task not found" as a daemon
cancellation trigger — that means a single DB hiccup could kill an
in-flight agent run.

Distinguish pgx.ErrNoRows (real "task gone", 404) from other errors
(transient, 500 + warn log) using the existing isNotFound helper.

Tests cover both paths via the mockDB pattern already used by
TestFindOrCreateUserGating.

Co-authored-by: multica-agent <github@multica.ai>

* fix(daemon): honor task-deleted signal in post-runTask completion guard

The final pre-completion check in handleTask only looked for
status == "cancelled" and ignored errors. After PR #2107 added a 404
task-deleted cancellation path to the in-flight watcher, this trailing
guard fell out of sync — if the task was deleted between the watcher's
last poll and runTask returning, handleTask would still try to call
CompleteTask and only learn about the deletion via the 404 from that
callback.

Reuse shouldInterruptAgent so the same truth table (cancelled OR
404 task-not-found, but NOT transient errors) drives both polling and
the final guard.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-05-06 16:02:59 +08:00
DimaS
b1be9ed27f fix(daemon): cancel running agent when task is deleted server-side (#2107) 
When the server deletes a task while the daemon's agent is still running
(issue removed, agent reassigned, workspace cleanup), GetTaskStatus
starts returning 404 "task not found". The previous polling loop only
checked for status == "cancelled" and silently swallowed the error, so
the local agent kept emitting tool calls against a dead task until its
own timeout fired — minutes of wasted model spend and patch_apply
operations against a workdir nobody would consume.

Changes:

- Add isTaskNotFoundError next to isWorkspaceNotFoundError so the daemon
  can distinguish "task gone" 404 from "workspace gone" 404 (already
  handled separately) and from generic network errors.
- Extract the cancellation polling goroutine in handleTask into
  watchTaskCancellation, plus a pure shouldInterruptAgent decision
  helper. The pure helper makes both signals (cancelled status and 404
  task) easy to unit-test without spinning up a real backend.
- Trigger interruption on the new 404 path. Transient errors (5xx,
  network) intentionally still don't cancel — the next poll will retry
  and a flaky link should not kill an in-flight agent.

Tests cover the helper truth table, the existing "status cancelled"
path, the new "task deleted (404)" path, and a negative case ensuring a
running task is not interrupted.

Co-authored-by: “646826” <“646826@gmail.com”>
 2026-05-06 15:45:03 +08:00
Bohan Jiang
144661e68f fix(daemon/execenv): refresh stale Codex auth.json across env reuse (#2126) 
`ensureSymlink` previously short-circuited whenever `dst` already existed
as a regular file ("Regular file exists — don't overwrite"). On Windows
that branch is reachable via the createFileLink copy fallback that fires
when `os.Symlink` is unavailable, so once a per-task `codex-home/auth.json`
was written as a copy it would never be refreshed by subsequent
Prepare/Reuse calls. If the shared `~/.codex/auth.json` rotated (e.g.
Codex Desktop refreshed the token in the background), the daemon kept
handing Codex a now-revoked refresh_token, which the OAuth server
rejected with `refresh_token_reused` / `token_expired`. Renaming the
workspace directory was the only recovery path.

Treat any non-matching dst — wrong-target symlink, broken symlink, or
stale regular file — as something to delete and re-create via
createFileLink, so each Prepare/Reuse mirrors the current shared source.
Add a `logCodexAuthState` info log (file kind, link target, size, mtime —
never contents) so operators chasing the same symptom can see at a glance
whether the per-task home is tracking the shared auth or has drifted.

Tests cover: stale regular-file dst is replaced, copy-fallback dst is
refreshed when the shared source rotates, and a high-level
prepareCodexHome regression simulating the Windows + token-rotation
scenario from issue #2081.

Co-authored-by: multica-agent <github@multica.ai>
 2026-05-06 15:18:04 +08:00
Matt Van Horn
0dbfbfed2e fix(daemon/execenv): refuse to write .gc_meta.json when issue_id is empty (#2077) 
A non-trivial fraction of completed task workdirs (~28% in field reports)
end up with .gc_meta.json files containing issue_id: "". Empty issue_id
defeats the daemon's own GC loop (gc.go:139 calls
GetIssueGCCheck(meta.IssueID)) and external retention scripts that
cross-reference issue status before deleting orphaned workdirs.

Refuse to write the file when issueID is empty, logging a Warn so
operators have a starting point for debugging the upstream race
condition. Skip is preferred over a sentinel-marker file: it keeps the
data invariant clean (a .gc_meta.json file always carries a valid
issue_id) and matches the repo CLAUDE.md preference for not preserving
dual-state behavior.

WriteGCMeta now takes a *slog.Logger so it can emit the warning. The
package already uses log/slog (Prepare/reuseEnv), and daemon.go:884 has
taskLog in scope at the only call site.

Closes #1913

Co-authored-by: Matt Van Horn <455140+mvanhorn@users.noreply.github.com>
 2026-05-06 15:02:16 +08:00
Bohan Jiang
09f04847d3 feat(server): redis-backed runtime liveness with DB fallback (#2121) 2026-05-06 14:31:33 +08:00
Bohan Jiang
b08594f2f6 fix(daemon): isolate runtime poll & heartbeat schedules per runtime (#2116) 
* fix(daemon): isolate runtime poll & heartbeat schedules per runtime

A daemon serving multiple workspaces ran a single round-robin poll loop
and a single HTTP heartbeat loop across every registered runtime. A 30s
HTTP timeout for any one runtime serialized that delay across all the
others — observed in production as one workspace's runtimes wedging
every other workspace's runtimes on the same daemon.

This change:

- Replaces the shared runtime-set channel with a multi-subscriber
  watcher so taskWakeupLoop, heartbeatLoop, and pollLoop can each
  react to runtime-set changes independently.
- Splits heartbeatLoop and pollLoop into supervisor + per-runtime
  worker goroutines. Each runtime owns its claim cadence and its
  heartbeat ticker, so a slow request on one runtime no longer blocks
  any other.
- Stagers the per-runtime heartbeat first tick by a jittered delay up
  to one full interval to avoid a thundering herd at startup.
- Sizes the WS writer channel to scale with the runtime count
  (max(16, 2*N)) so a full per-runtime heartbeat batch always fits;
  the previous fixed 8-slot buffer dropped heartbeats whenever a
  daemon watched more than ~8 runtimes.

Co-authored-by: multica-agent <github@multica.ai>

* fix(daemon): acquire execution slot only after ClaimTask, drain pollers before taskWG

Two issues from review on the previous commit:

1. Acquiring the shared task slot before ClaimTask reintroduced the very
   head-of-line blocking the refactor was meant to remove. With
   MaxConcurrentTasks=1, a slow claim on one runtime parked the only slot
   for the duration of the HTTP timeout (up to 30s), starving every other
   runtime's claim attempts. Slots are now acquired after the claim
   returns a task; other runtimes' pollers stay free to claim. The
   already-dispatched task waits for a slot under MaxConcurrentTasks
   bounds, which is the same backpressure shape we had before.

2. pollLoop's shutdown path called taskWG.Wait immediately after
   cancelling pollers, but a poller could still be between ClaimTask
   returning a task and taskWG.Add(1). When taskWG's counter is zero
   that races with Wait — undefined sync.WaitGroup misuse, sometimes
   panic. Added a pollerWG so the supervisor blocks until every poller
   goroutine has actually returned before reaching taskWG.Wait.

Tests:
- TestRunRuntimePollerIsolatesSlowRuntime now uses MaxConcurrentTasks=1
  (was 4) so it would have failed under the old slot-before-claim path.
- New TestPollLoopShutdownWaitsForPollersBeforeTaskWG drives the exact
  race window — claim returns a task at the same moment shutdown fires —
  under -race.

Co-authored-by: multica-agent <github@multica.ai>

* fix(daemon): acquire slot before ClaimTask so capacity-waiters never enter dispatched

The previous commit moved slot acquisition AFTER ClaimTask to address a
review concern about head-of-line blocking with MaxConcurrentTasks=1.
That introduced a strictly worse failure mode: server-side ClaimTask
flips the task to `dispatched` immediately (agent.sql:174-176), and the
runtime sweeper fails any task in `dispatched` for >300s with
`failed/timeout` (runtime_sweeper.go:25-28). When local execution
capacity is full and the next claimed task can't acquire a slot within
5 minutes, the user sees the exact failure this issue is fixing —
`dispatched_at` set, `started_at` NULL, `failure_reason=timeout`.

Reverted to slot-before-claim. The trade-off is the original review
concern: with MaxConcurrentTasks=1 and a slow ClaimTask, other
runtimes' claims are delayed by up to client.Timeout=30s. That's a
30s polling delay, not a failure — server-side those tasks remain
`queued` (no timeout in that state) until a slot frees. 30s ≪ 300s,
so other runtimes' tasks cannot get sweeper-failed because of this.

The pollerWG fix from the previous commit (avoiding sync.WaitGroup
misuse on shutdown) is preserved.

Tests:
- TestRunRuntimePollerIsolatesSlowRuntime: MaxConcurrentTasks back to
  4 (the pre-issue baseline) — the headroom case where slot-before-
  claim still gives full per-runtime isolation.
- New TestRunRuntimePollerSkipsClaimWhenAtCapacity: holds the only
  slot and verifies the poller never calls ClaimTask while sem is
  empty. The previous "claim first" path would have failed this.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-05-06 14:13:27 +08:00
Bohan Jiang
a4fac51cf5 fix(projects): add resource_count breadcrumb instead of inlining resources (#2118) 
* fix(projects): add resource_count breadcrumb to project responses

Closes #2087

`multica project get` previously returned project metadata with no signal
that resources existed. Agents that fetched a project this way had no way
to discover its attached resources without already knowing about
`/api/projects/{id}/resources` or the on-disk `.multica/project/resources.json`.

Rather than inline the full resource list into the parent payload (which
conflates parent metadata with a child sub-collection and locks the
resource_ref shape into the project endpoint's contract), this adds a
scalar `resource_count` breadcrumb to ProjectResponse. The actual list
stays at the dedicated sub-collection endpoint.

Changes:
- GetProjectResourceCounts :many — new batched sqlc query
- ProjectResponse.ResourceCount populated in GetProject, ListProjects,
  SearchProjects, and the with-resources CreateProject echo
- multica project get prints a stderr hint pointing at
  multica project resource list <id> when count > 0; the JSON on stdout
  stays parseable
- Meta-skill (runtime_config.go) lists multica project get and
  multica project resource list in Available Commands so agents that
  read CLAUDE.md / AGENTS.md know about both paths

Co-authored-by: multica-agent <github@multica.ai>

* fix(projects): wire ResourceCount through Update + Create event payload

Review feedback on #2118.

- UpdateProject now reloads ResourceCount before responding/publishing.
  Previously a title- or status-only PUT served (and broadcast over WS)
  resource_count: 0 even when resources existed.
- The with-resources CreateProject path sets resp.ResourceCount before
  the project:created publish, so the WS event payload matches the HTTP
  echo. The hand-rolled response map collapses to an embedded
  ProjectResponse + resources array — one source of truth for the
  serialized shape.
- packages/core/types/project.ts: Project gains resource_count: number
  to keep the TS contract aligned with the server response.

Tests:
- TestProjectResourceCountBreadcrumb extends to assert UpdateProject
  preserves the breadcrumb.
- TestCreateProjectWithResourcesEchoesCount asserts the create echo
  carries resource_count matching the attached resources.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-05-06 14:09:35 +08:00
Bohan Jiang
60b215f44f feat(chat): support deleting chat sessions (#2115) 
* feat(chat): support deleting chat sessions

Replaces the unreachable archive endpoint with a real hard delete and
exposes it from the chat history panel.

- DELETE /api/chat/sessions/{id} now hard-deletes the session and its
  messages (CASCADE), cancels any in-flight tasks before removal so the
  daemon doesn't keep running work whose result has nowhere to land,
  and broadcasts chat:session_deleted.
- Frontend adds a per-row delete button with a confirmation dialog,
  optimistically drops the session from both list caches, and clears the
  active session pointer locally + on other tabs via the WS handler.

Co-authored-by: multica-agent <github@multica.ai>

* fix(chat): make session delete atomic and keep archived sessions read-only

Address review feedback on #2115.

- DeleteChatSession now runs lock + cancel + delete in a single tx and
  only broadcasts events post-commit. The new LockChatSessionForDelete
  query takes FOR UPDATE on chat_session, which blocks the FK validation
  of any concurrent SendChatMessage trying to enqueue a task for this
  session — that insert fails after we commit, so it can no longer
  produce an orphaned task whose chat_session_id is nulled by
  ON DELETE SET NULL. Cancel failure now aborts the delete instead of
  warn-and-continue.
- SendChatMessage refuses non-active sessions again. The archive code
  path is gone, but legacy rows with status='archived' may still exist
  in the DB; keep the guard until we explicitly migrate them.
- Frontend re-reads allChatSessionsOptions to disable ChatInput on
  legacy archived sessions so the UX matches the server-side guard.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-05-06 13:22:53 +08:00
Bohan Jiang
f1082b10a4 feat(cli): add --assignee-id / --to-id / --user-id for unambiguous targeting (#2114) 
* feat(cli): add --assignee-id / --to-id / --user-id for unambiguous targeting

`multica issue {create,update,list}`, `issue assign`, and `issue subscriber
{add,remove}` accepted only fuzzy name matching, which fails in workspaces
where one user's name is a substring of another (e.g. agent "J" vs
"Cursor - J" / member "Jiayuan"). #1642 added UUID acceptance through the
existing flags, but there was still no explicit path that signals "this is a
UUID, not a name" — important for scripts that read IDs from
`multica workspace members --output json`.

Adds an `-id`-suffixed counterpart for every assignee-taking flag:

- `issue list`     : --assignee-id
- `issue create`   : --assignee-id
- `issue update`   : --assignee-id
- `issue assign`   : --to-id
- `issue subscriber {add,remove}` : --user-id

The new flags route through `resolveAssigneeByID`, a strict resolver that
requires a canonical UUID and fails with a clear error when the entity is
not in the workspace (no name fallback). A shared `pickAssigneeFromFlags`
helper enforces mutual exclusion between the name and id flags so a script
that accidentally sets both never silently applies one over the other.

Refs MUL-1254.

Co-authored-by: multica-agent <github@multica.ai>

* fix(cli): detect assignee flag presence via Changed, not value-emptiness

`pickAssigneeFromFlags` previously branched on `flag value != ""`, so
explicitly passing an empty UUID silently routed through the "no flag set"
path:

  multica issue list --assignee-id ""        # listed every issue
  multica issue create --assignee-id ""      # created an unassigned issue
  multica issue subscriber add --user-id ""  # subscribed the caller

This is exactly the failure mode the strict-UUID flag was added to prevent —
a script interpolating `--assignee-id "$MAYBE_UUID"` against a missing env
var should fail loudly, not silently degrade to a different operation.

Switch the picker (and the assign-command top-level guard) to use
`Flags().Changed`, so an explicit empty value reaches `resolveAssigneeByID`
/ `resolveAssignee` and surfaces a clear "expected a canonical UUID" /
"no member or agent found matching" error.

Co-authored-by: multica-agent <github@multica.ai>

* docs(cli): cover --assignee-id / --to-id in user docs and quick-create prompt

Follow-up to the --*-id flag rollout: surface the new flags everywhere the
old ones are documented so users (and agents) can discover them.

- assigning-issues.{mdx,zh.mdx}: the page explicitly calls out the
  duplicate-name footgun ("first one listed wins, so rename before
  assigning") — replace that workaround with a --to-id <uuid> example
- cloud-quickstart.{mdx,zh.mdx}: add a --to-id hint after the substring-
  match callout so first-time users learn about the strict path
- internal/daemon/prompt.go (quick-create injected prompt):
  - default-to-self: pass --assignee-id <task.Agent.ID> instead of
    --assignee <name>; the picker agent's UUID is already in scope and
    UUID matching is unambiguous in workspaces with overlapping agent
    names (J / Cursor - J / Pi - J etc.)
  - user-named: tell the agent to prefer --assignee-id <uuid> using the
    user_id/id from the JSON it already fetched; --assignee <name> stays
    a fallback for unambiguous workspaces

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-05-06 13:13:36 +08:00
LinYushen
44a0ced558 fix(runtime): persist CLI update requests in Redis (#2113) 
Co-authored-by: multica-agent <github@multica.ai>
 2026-05-06 13:00:11 +08:00
Bohan Jiang
89b939b07c fix(storage): build region-qualified S3 public URLs (#2051) (#2065) 
* fix(storage): build region-qualified S3 public URLs (#2051)

The uploadedURL fallback (no CloudFront, no custom endpoint) wrote
"https://<bucket>/<key>" — missing the ".s3.<region>.amazonaws.com"
suffix — so any deployment that pointed S3_BUCKET at a real AWS bucket
without a CDN got broken image URLs back to the client. Avatar URLs
were persisted in this broken form on the user/agent rows, so profile
pictures uploaded via the SDK never rendered.

- Track S3_REGION on S3Storage and emit
  https://<bucket>.s3.<region>.amazonaws.com/<key> by default;
  fall back to path-style https://s3.<region>.amazonaws.com/<bucket>/<key>
  when the bucket name contains dots, since the AWS wildcard cert
  can't validate dotted virtual-hosted hosts.
- Teach KeyFromURL to recognise the new region-qualified hosts (both
  styles) and keep recognising the legacy bucket-only host so historical
  records can still be deleted/migrated.
- Document that S3_BUCKET is the bucket name only, not a hostname,
  in env-vars docs (en+zh), self-hosting guides, and .env.example.

Co-authored-by: multica-agent <github@multica.ai>

* feat(storage): warn at startup when S3_BUCKET looks like a hostname

Catches the most common misconfiguration shape (S3_BUCKET set to
"<bucket>.s3.<region>.amazonaws.com") with a startup log line so
operators don't silently end up with a config that signs uploads
against an invalid bucket name.

A real bucket name can never legitimately contain "amazonaws.com",
so the check is a single substring match — no false positives
worth carving out.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-05-06 12:45:55 +08:00
Ákos Seres
64c605e227 fix(execenv): write OpenCode skills to .opencode/skills/ for native discovery (#2016) 
* fix(execenv): write OpenCode skills to .opencode/skills/ for native discovery

* fix(repocache): exclude OpenCode skill directory
 2026-05-06 11:48:06 +08:00
Jiayuan Zhang
1476c268dd refactor(quick-create): exempt git-describe daemons from CLI gate (#2108) 
* refactor(quick-create): remove daemon CLI version gate

Local-source daemons report dev-suffixed versions (e.g.
v0.2.15-235-gdaf0e935) that the picker pre-check and server gate both
treat as too old, blocking quick-create during local testing.

Drops the gate end-to-end: removes MinQuickCreateCLIVersion +
CheckMinCLIVersion in pkg/agent, the checkQuickCreateDaemonVersion
handler and readRuntimeCLIVersion helper in handler/issue.go, and the
mirrored cli-version.ts plus the modal's pre-check, blocked-state UI,
and daemon_version_unsupported error branch.

Co-authored-by: multica-agent <github@multica.ai>

* refactor(quick-create): skip daemon CLI version gate in dev

Restores the gate (reverts the full-removal commit) and bypasses it in
non-production environments instead. The motivation for the original
removal — local source-built daemons report a `git describe` version
like v0.2.15-N-gHASH that parses below 0.2.20 and blocks dev testing —
is now handled by checking APP_ENV on the server and NODE_ENV on the
client. Production keeps the original "needs upgrade" UX.

Co-authored-by: multica-agent <github@multica.ai>

* refactor(quick-create): exempt git-describe daemons instead of env bypass

Replaces the per-environment bypass added in the previous commit with a
shared daemon-version signal. CheckMinCLIVersion / checkQuickCreateCliVersion
now treat any daemon whose CLI version matches the
`vX.Y.Z-N-gHASH[-dirty]` git-describe shape as OK; tagged releases keep
going through the normal min-version comparison.

Why: Emacs flagged that (a) NODE_ENV !== "production" also disables the
gate on staging and other non-prod deployments, undoing the protection
for the case the gate was originally written for, and (b) NODE_ENV (web
client) and APP_ENV (server) are not equivalent, so the modal pre-check
and server gate could disagree on the same request. Both go away when
the signal is intrinsic to the daemon's version string.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-05-06 09:00:11 +08:00
Bohan Jiang
5c42ed1649 fix(server): allow re-inviting after invitation expires (#2059) 
The uniqueness check on workspace invitations only filtered by
status='pending', not by expires_at. Combined with the partial unique
index idx_invitation_unique_pending (also keyed only on status), a
past-due pending row permanently blocked re-inviting the same email.

Now, before creating a new invitation, the handler flips any past-due
pending row for the same (workspace_id, invitee_email) to 'expired',
freeing the unique slot. Also tightens GetPendingInvitationByEmail to
require expires_at > now(), matching the existing list queries.

Closes multica-ai/multica#2055.
 2026-05-04 21:24:56 +08:00
Jiayuan Zhang
d492b9d7a6 Revert "feat(quick-create): add preset issue fields (#2002)" (#2042) 
This reverts commit a039c4d803.
 2026-05-03 20:02:40 +02:00
Bohan Jiang
3dc3e49a47 fix(daemon): remove Co-authored-by hook when workspace setting is off (#2035) 
* fix(daemon): remove Co-authored-by hook when workspace setting is off

The prepare-commit-msg hook is installed in the bare repo's shared
hooks dir, so once installed it persists across worktrees. CreateWorktree
only installed the hook when the setting was enabled, but never removed
it — so disabling the workspace toggle had no effect on subsequent
commits.

Add removeCoAuthoredByHook and call it in both CreateWorktree branches
when the setting is disabled. Use a marker comment in the hook script so
removal only deletes hooks the daemon owns; user-installed hooks at the
same path are left alone.

Co-authored-by: multica-agent <github@multica.ai>

* fix(daemon): recognize legacy Multica prepare-commit-msg hook on removal

The first cut of removeCoAuthoredByHook only recognized hooks installed
by the new code (containing the multicaHookMarker sentinel). Bare clones
already on disk from previous daemon releases carry the older script
without that line, so toggling the workspace setting off would have
treated them as user hooks and left the trailer in place — exactly the
state reported in MUL-1704.

Match against a list of known daemon signatures (current marker + the
legacy "Installed by the Multica daemon." comment), and add a test that
seeds the verbatim legacy hook before CreateWorktree(... disabled) to
keep recognition aligned with what production hosts actually have on
disk.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-05-03 21:09:16 +08:00
Kagura
cc94fbd305 fix: handle square brackets in agent names for mention parsing (#1992) 
* fix: handle square brackets in agent names for mention parsing (#1991)

The mention regex used [^\]]* to match labels, which broke when agent
names contained square brackets (e.g. David[TF]). The ] inside the name
caused the regex to stop matching prematurely, silently dropping the
mention.

Changes:
- Backend (mention.go): Switch to .+? (non-greedy) anchored on
  ](mention:// to correctly match labels with brackets
- Frontend (mention-extension.ts): Same regex fix in tokenizer, plus
  escape [ and ] in renderMarkdown to prevent creating ambiguous
  markdown syntax
- Add comprehensive tests for ParseMentions covering bracket names

Fixes #1991

* fix: add optional chaining for match group access

Fixes TS2532: Object is possibly 'undefined' on match[1] when calling
.replace() in the mention tokenizer.

* fix: tighten mention tokenizer to reject ordinary Markdown links

- Replace .+? with (?:\\.|[^\]])+  in start() and tokenize() regexes
  so the label cannot cross a ]( Markdown link boundary
- Escaped brackets (\[ \]) from renderMarkdown() are still accepted
- Add frontend tokenizer/serializer round-trip tests:
  - Plain mention
  - Escaped brackets (David[TF]) round-trip
  - Normal Markdown link + mention on same line (regression)
  - Multiple links before mention
  - Nested brackets (Bot[v2][beta])
  - Issue mentions without @ prefix

Addresses review feedback on #1992.

* fix: add type assertions for tiptap MarkdownTokenizer interface in tests

The tiptap MarkdownTokenizer type allows start to be string | function
and tokenize to accept 3 arguments. Our extension always provides
single-arg functions, so cast them for TypeScript satisfaction.

Fixes CI typecheck failure in @multica/views package.

* fix: cast renderMarkdown to single-arg shape and reset file modes to 0644
 2026-05-03 19:39:26 +08:00
ayakabot
a039c4d803 feat(quick-create): add preset issue fields (#2002) 
Fixed: #2001
 2026-05-03 19:37:12 +08:00
Bohan Jiang
c4352da126 fix(daemon): drain background repo syncs before test teardown (#2026) 
TestRegisterTaskReposSurvivesWorkspaceRefresh started flaking on CI
after #1988 (`feat: support repo checkout ref selection`) extended the
bare-clone path to run an extra `git fetch` to backfill
refs/remotes/origin/* under the new refspec layout. The race was
already latent: registerTaskRepos kicks off `go syncWorkspaceRepos(...)`
to clone a repo into the cache root, which in tests is `t.TempDir()`.
Once the test waited on `repoCache.Lookup` to return a path it would
proceed and return — but the bg goroutine was still inside
`ensureRemoteTrackingLayout` running git operations on the clone dir.
`t.TempDir`'s cleanup then races with those git commands and surfaces
either as "directory not empty" or "fatal: cannot change to ... No such
file or directory", with no hint that the failure is unrelated to the
test's actual assertion.

Track the background goroutine on the Daemon via a sync.WaitGroup and
expose `waitBackgroundSyncs()` for tests. `newRepoReadyTestDaemon`
registers a t.Cleanup that calls it, so every test that uses the
helper now drains in-flight syncs before t.TempDir cleanup runs. No
production-behavior change — registerTaskRepos still fires-and-forgets
from the caller's perspective.

Verified with `go test ./internal/daemon -run
TestRegisterTaskReposSurvivesWorkspaceRefresh -count=30` (was failing
within ~10 iterations before, 30 green after) and the full
`go test ./internal/daemon/...` suite.

Co-authored-by: multica-agent <github@multica.ai>
 2026-05-03 12:24:56 +08:00
Bohan Jiang
a414a00b4a refactor(repocache): clarify resolveBaseRef comment and cover tag refs (#2023) 
Follow-up nits from PR #1988 review:

- Move the comment that documents getRemoteDefaultBranch's resolution
  walk into the resolveBaseRef call site description, and rephrase the
  "" branch so it's clear that path only fires for the default-branch
  case (the requested-ref path returns an explicit error before
  reaching it).
- Add TestCreateWorktreeWithRequestedTagRef to lock in the
  refs/tags/<ref> candidate. The test tags the initial commit, advances
  the default branch past it, then asserts the worktree HEAD matches
  the tagged commit (so the tag must have been resolved, not the
  default branch).

Co-authored-by: multica-agent <github@multica.ai>
 2026-05-03 11:30:25 +08:00
Prince Pal
862b0509df feat: support repo checkout ref selection (#1988) 2026-05-03 11:27:16 +08:00
Bohan Jiang
ba5b7db78e fix(server): persist ModelListStore across replicas via Redis (#2022) 
* fix(server): persist ModelListStore across replicas via Redis

The model picker uses a pending-request pattern: the frontend POSTs to
create a request, the daemon pops it on its next heartbeat, runs
agent.ListModels locally, and reports back. Until now the store was a
plain in-memory map per Handler instance.

That works for self-hosted single-instance deploys but fails in any
multi-replica environment (Multica Cloud). Each replica has its own
map, so:

  POST /runtimes/:id/models               → request stored in replica A
  GET  /runtimes/:id/models/<requestId>   → polls land on B/C → 404
  daemon heartbeat                        → only A sees PendingModelList
  POST .../<requestId>/result             → daemon's report has to land on A

Success probability ~1/N². The visible symptom is "No models available"
in the picker for every provider, even those (Claude/Codex) whose
catalog is statically populated end-to-end.

Same shape of bug, same Redis-backed fix as multica-ai/multica#1557 did
for LocalSkillListStore / LocalSkillImportStore. Reuse the operational
playbook (namespaced keys, ZSET-backed pending queue, atomic
ZREM+SET-running via the shared Lua script) so we don't introduce a
second concurrency model for the same primitive.

Changes:
- Convert ModelListStore from struct to interface with context-aware
  methods. Add HasPending for cheap heartbeat-side probing.
- InMemoryModelListStore — single-node fallback, used when REDIS_URL
  is unset (self-hosted dev / tests).
- RedisModelListStore — multi-node implementation using the same key
  layout and Lua atomic claim as RedisLocalSkillListStore.
- Use RunStartedAt (not UpdatedAt) as the running-timeout reference
  point, matching the local-skill stores so subsequent UpdatedAt
  bumps don't reset the running clock.
- Heartbeat now uses the probe-then-pop pattern for the model queue
  (matching local-skills) so a slow Redis can't stall every connected
  daemon. Extends heartbeatMetrics + slow-log with probe_model_ms /
  pop_model_ms / probe_model_timed_out for parity.
- Wire the Redis backend in NewRouterWithOptions when rdb != nil.
- Tests for both backends. Redis tests gate on REDIS_TEST_URL so
  laptop runs without Redis still pass; CI provides it.

Co-authored-by: multica-agent <github@multica.ai>

* fix(server): persist RunStartedAt + retry model report on transient failures

Two follow-ups from PR #2022 review:

1. RedisModelListStore was dropping ModelListRequest.RunStartedAt on
   persistence — the field is tagged json:"-" so it doesn't leak into
   the HTTP response, which made plain json.Marshal(req) silently
   discard it. Across-node readers saw RunStartedAt=nil and
   applyModelListTimeout's running branch became a no-op, so the 60s
   running-timeout escape hatch never fired. CI's
   TestRedisModelListStore_RunningTimeout was failing on this exact
   case. Fix mirrors RedisLocalSkillImportStore's envelope pattern —
   wrap in an internal struct that re-promotes the field. HTTP shape
   stays clean. Adds a no-Redis unit test that pins the round trip.

2. Daemon's handleModelList called d.client.ReportModelListResult
   directly and swallowed any 5xx, leaving the pending request
   stranded in "running" until its 60s server-side timeout — exactly
   the failure mode the multi-node store fix was meant to eliminate.
   Generalize the existing local-skill retry helper into
   reportRuntimeResultWithRetry (kind: model_list / local_skill_list /
   local_skill_import) and wire handleModelList through a new
   reportModelListResult helper. Renames the test-overridable
   var localSkillReportBackoffs → runtimeReportBackoffs to match.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-05-03 11:13:34 +08:00
Bohan Jiang
e665b597b3 refactor(server): polish runtime-guard nits from PR #1905 review (#2021) 
- Expand chat-resume comment in ClaimTaskByRuntime to spell out *why* the
  task-row fallback exists (single failed turn must not drop chat memory)
  and that it covers more than just legacy NULL rows.
- Replace the sessionRuntimeID := t.RuntimeID; sessionRuntimeID.Valid = ...
  pattern in CompleteTask/FailTask with a clearer var-then-assign that makes
  the "no session_id, leave runtime_id alone" coupling obvious.
- Add TestClaimTask_ChatLegacyNullRuntimeFallsBackToTaskRow covering the
  case the prior PR's tests didn't reach: chat_session.runtime_id IS NULL
  (legacy / unbackfilled) plus a matching-runtime task row, fallback
  should resume. This is the dominant post-migration shape and was
  previously only covered transitively.

No behavior change beyond the new test; runtime-guard semantics stay
identical to PR #1905.

Co-authored-by: multica-agent <github@multica.ai>
 2026-05-03 11:00:32 +08:00
Multica Eve
f85b7cce91 fix: make CLI update completion status reliable (#2018) 
Co-authored-by: Devv <devv@Devvs-Mac-mini.local>
Co-authored-by: multica-agent <github@multica.ai>
 2026-05-03 10:52:14 +08:00
Bright Zheng
cf47d9b702 fix: guard session resume by runtime (#1905) 2026-05-03 10:51:31 +08:00
Bright Zheng
c2f199650a feat(cli): add agent avatar upload command (#1760) 
* feat(cli): add UploadFileWithURL and AttachmentResponse to APIClient

* feat(cli): add agent avatar command and show avatar_url in agent get output

* fix(server): include id and url in no-workspace file upload response

* fix(cli): remove dead HTTPClient timeout swap, extend ctx to 60s for avatar upload

The 30s context deadline was tighter than the 60s HTTPClient timeout
swap, so the swap was dead code and did nothing for slow connections.
Both Neo and Omni Mentor flagged this in review.

Fix: extend the command context to 60s and remove the HTTPClient
mutation. This is simpler, thread-safe, and actually works for slow
uploads.

* fix: align fallback upload response shape and honor context deadline

- file.go: fallback returns {id, url, filename} instead of {filename, link},
  matching the no-workspace path response shape.
- client.go UploadFileWithURL: tolerate empty attachment ID (S3 succeeded
  but DB record failed — the file is still usable via its URL).
- client.go UploadFileWithURL: use a context-deadline-aware HTTP client so
  that the 60s upload timeout set by the avatar command actually takes
  effect instead of being shadowed by the default 15s client timeout.
- client_test.go: update 'missing id' test to verify empty-id success
  (fallback tolerance).

* fix(cli): shallow-copy HTTP client to preserve Transport on upload timeout

When the context deadline exceeds the default 15s HTTP client timeout,
UploadFileWithURL was creating a bare &http.Client{Timeout: remaining},
silently dropping any custom Transport, Jar, or CheckRedirect configured
on the original client. This causes obscure connection failures when the
CLI uses an authenticated proxy, custom TLS, or mock transport in tests.

Fix: perform a shallow copy of the original client struct and only
mutate the Timeout field on the copy.
 2026-05-03 10:49:02 +08:00
Jiayuan Zhang
3df95c84b8 fix(daemon): add safe.directory=* to gitEnv to fix CI dubious ownership errors (#1980) 
* fix(daemon): add safe.directory=* to gitEnv to fix CI dubious ownership errors

TestRegisterTaskReposAllowsProjectOnlyURL and
TestRegisterTaskReposSurvivesWorkspaceRefresh fail on GitHub Actions CI
because git clone --bare from local temp directories triggers git's
safe.directory ownership check when the runner UID differs from the
directory owner.

Set safe.directory=* via GIT_CONFIG env vars in gitEnv() so all daemon
git subprocesses trust any directory. The daemon manages its own bare
caches and worktrees, so the ownership check provides no security value.

Co-authored-by: multica-agent <github@multica.ai>

* fix(daemon): preserve existing GIT_CONFIG_* entries in gitEnv

Instead of resetting GIT_CONFIG_COUNT to 1, read the existing count
from the environment and append safe.directory at the next available
index. This preserves any env-scoped git config (auth, URL rewrites,
extra headers) injected into the daemon process.

Adds TestGitEnvPreservesExistingConfig to verify the append behavior.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-05-01 16:18:58 +02:00
Jiayuan Zhang
13fe614903 fix(daemon): optimize quick-create prompt for high-fidelity descriptions (#1969) 
The previous description rule ("stay faithful + keep it concise") caused
agents to over-compress user input into vague single-sentence summaries,
losing context that the executing agent needs.

Key changes:
- Replace "keep it concise" with structured two-section format:
  User request (faithful restate) + Context (verifiable external facts)
- Add hard rules against information compression and semantic downgrading
- Remove "one-line description" phrasing (UI supports richer input)
- Strip redundant behavioral rules from issue_context.md (already
  covered by AGENTS.md guardrails and per-turn prompt)

Co-authored-by: multica-agent <github@multica.ai>
 2026-05-01 08:14:55 +02:00
wucm667
2305f7d180 fix(skill): sanitize null bytes in all skill update/upsert paths to prevent PostgreSQL UTF8 error (#1959) 2026-04-30 22:34:24 +02:00
LinYushen
51fdc5aec3 Increase empty claim cache TTL (#1938) 2026-04-30 17:13:56 +08:00
Naiyuan Qing
51bc5a818f fix(onboarding): decouple from workspace state and route invitees correctly (#1936) 
PR #1868 conflated "has workspace" with "completed onboarding" —
restore `onboarded_at` as the single signal, and route invited users
through a dedicated /invitations page before they ever see onboarding.

- Backend: CreateWorkspace + AcceptInvitation atomically set
  onboarded_at alongside the member insert, establishing the
  invariant "member row exists ↔ onboarded_at != null" at the DB
  layer.
- Migration 065: one-shot backfill closes the dirty rows produced
  by PR #1868 (users with a workspace but onboarded_at == null).
- Entry points (web callback, login, desktop App): if onboarded_at
  is null, look up pending invitations by email and route to the
  new batch /invitations page; otherwise the resolver picks
  workspace / new-workspace as before.
- OnboardingPage: stops bouncing on hasWorkspaces; only
  hasOnboarded bounces. Unblocks the user from completing
  Step 3 (workspace creation) → Steps 4 / 5.
- StarterContentPrompt: only shows when the user is the solo
  member of the workspace, so invited users never get prompted to
  import starter content into someone else's workspace.

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-04-30 16:05:53 +08:00
Bohan Jiang
2dddfaa196 feat(daemon): Redis empty-claim fast path for /tasks/claim polling (#1860) 
* feat(daemon): Redis empty-claim fast path for /tasks/claim polling

Daemons poll /tasks/claim every 30s per runtime; the steady-state
warm-empty case currently runs ListPendingTasksByRuntime against
Postgres on every poll. This collapses that path:

- New ListQueuedClaimCandidatesByRuntime query restricts to status =
  'queued' (the old query also returned 'dispatched' rows that can
  never be reclaimed) and is backed by a partial index keyed on
  (runtime_id, priority DESC, created_at ASC).
- New EmptyClaimCache caches the negative verdict in Redis with a
  30s TTL. ClaimTaskForRuntime checks the cache before SELECT and
  populates it on confirmed-empty results.
- notifyTaskAvailable now invalidates the runtime's empty key before
  kicking the daemon WS, so newly enqueued tasks become claimable
  immediately rather than waiting out the TTL.
- AutopilotService.dispatchRunOnly now goes through
  TaskService.NotifyTaskEnqueued so run_only tasks get the same
  invalidate-then-wakeup contract as every other enqueue path.

Co-authored-by: multica-agent <github@multica.ai>

* fix(daemon): close MarkEmpty/Bump race in empty-claim fast path

GPT-Boy's review on PR #1860 caught a real concurrency bug. Under the
prior implementation it was possible for a slow claim to write an
empty verdict AFTER a concurrent enqueue had already invalidated it:

  T1 claim:   SELECT -> empty
  T2 enqueue: INSERT row, DEL empty key (no-op, key not set yet),
              wakeup
  T1 claim:   SET empty (writes a stale "empty" verdict)
  T3 wakeup:  IsEmpty -> hit -> returns null

The just-queued task would then sit idle until the empty key's TTL
expired (up to 30s).

Replace the DEL-based invalidation with a per-runtime version
counter:

- CurrentVersion(rt) is a Redis INCR counter at
  mul:claim:runtime:version:<rt> with a 24h sliding TTL.
- Claim samples version BEFORE the SELECT and passes it to MarkEmpty,
  which stores the verdict's value as the observed-version string.
- IsEmpty MGETs both keys and trusts the verdict only when the
  empty-key value equals the current version.
- Enqueue Bumps the version (INCR + EXPIRE) before the wakeup,
  causing any verdict written under a prior version to be rejected
  on the next read.

Also bound every Redis call from this cache with a 250ms timeout —
notifyTaskAvailable uses a background context so a wedged Redis
must not block enqueue.

Tests against a real Redis (REDIS_TEST_URL) cover:
- MarkEmpty + IsEmpty under matching version returns hit
- Bump invalidates a prior empty verdict (race-fix pin)
- A MarkEmpty written under a stale pre-Bump version is rejected
- TTL clamping, per-runtime isolation, nil-cache safety
- notifyTaskAvailable Bumps before the wakeup fires

Co-authored-by: multica-agent <github@multica.ai>

* chore(daemon): renumber claim-candidate index migration to 067

Slot 064 was taken on main by 064_notification_preference. The
migration runner tracks per-version in schema_migrations and would
silently skip the second 064_*, leaving the index uncreated.
Rename to 067 (next free slot).

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-04-30 15:50:05 +08:00
Ayman Alkurdi
cbe7f2c886 fix(api): batch-update no-op responses report updated=0 (#1660) (#1759) 
The `POST /api/issues/batch-update` handler walked every issue ID and
incremented `updated` regardless of whether the iteration carried any
mutation. When the caller's payload had no recognized field in
`updates` — e.g. status placed at the top level instead of nested,
"update" misspelled as singular, or "updates" missing entirely —
the loop ran N no-op UPDATEs (each if-guard skipped, each COALESCE
preserved the existing value) and the response cheerfully reported
`{"updated": N}` while nothing changed. Reporters mistook the
positive count for success and chased a phantom persistence bug.

Detect at the top of the handler whether any known mutation field is
present in the parsed `updates` payload; if none is, short-circuit
with `{"updated": 0}`. The wire shape stays 200 + `{updated}`
so existing callers don't break — only the count becomes truthful.

Tests cover the three caller shapes that hit this path (status at top
level, empty `updates: {}`, misspelled "update") plus a positive
case that locks in happy-path persistence and counting.

Closes #1660.
 2026-04-30 15:35:12 +08:00
Bohan Jiang
1d1dedbf6e fix(daemon): reclaim disk on long-open issues + correct cancelled-status check (#1931) 
* fix(daemon): reclaim disk on long-open issues + correct cancelled-status check

Two related fixes for GitHub #1890 (self-hosted disk space growth):

- The GC's done/cancelled branch compared `status.Status` against `"canceled"`
  (single l), but the issue schema and the rest of the daemon use `"cancelled"`
  (double l). Cancelled issues therefore never matched and only fell out via the
  72h orphan TTL, which itself doesn't fire because cancelled issues are still
  reachable. Aligning the spelling lets cancelled-issue task dirs be reclaimed
  on the normal TTL path.

- Add a third GC mode, artifact-only cleanup, for the common case the report
  flagged: an issue stays open for days while many tasks complete on it, so
  per-task `node_modules`, `.next` and `.turbo` directories accumulate without
  ever becoming GC-eligible. The new branch fires when `.gc_meta.completed_at`
  is older than `MULTICA_GC_ARTIFACT_TTL` (default 12h), the env root is not
  currently in use by an active task, and the issue is still alive. It removes
  only directories whose basename matches `MULTICA_GC_ARTIFACT_PATTERNS`
  (default narrow: `node_modules,.next,.turbo`); source, `.git`, `output/`,
  `logs/` and the meta file are preserved so subsequent tasks can still resume
  the workdir. Patterns containing path separators are dropped, `.git` subtrees
  are never descended into, symlinked matches are not followed, and every
  removal target is verified to live inside the task dir.

Bookkeeping: `Daemon` now tracks active env roots with a refcounted set so the
GC loop never reclaims a directory that is mid-execution; `runTask` claims the
predicted root early plus the prior workdir on reuse paths. The cycle log is
extended with bytes reclaimed and per-pattern counts so self-hosted operators
can see what was freed.

Docs: extend the daemon configuration table in CLI_AND_DAEMON.md with the new
GC env vars and add a Workspace garbage collection section explaining the
three modes and the artifact-pattern contract.

Co-authored-by: multica-agent <github@multica.ai>

* fix(daemon): protect active env root from full GC removal too

Address GPT-Boy's PR #1931 review: the active-root guard only fired in the
artifact-cleanup branch, leaving a real race on the full-removal paths. A
follow-up comment on a long-done issue dispatches a task that reuses the prior
workdir, but `CreateComment` does not bump issue.updated_at — so the issue
still satisfies the done+stale GCTTL window and `gcActionClean` would
`RemoveAll` the directory mid-execution. The orphan-404 path is similarly
exposed when a token's workspace access is in flux.

Move the `isActiveEnvRoot` check to the top of `shouldCleanTaskDir` so all
three delete actions (clean, orphan, artifact) skip an in-use env root in one
place, and drop the now-redundant guard from the artifact branch.

Add tests covering the three at-risk paths: active root + done/stale issue,
active root + 404 issue past orphan TTL, active root + no-meta orphan past
TTL.

Also align two stale comments noted in the same review: cleanTaskArtifacts now
documents that symlinks are skipped entirely (the previous note implied the
link itself was removed), and GCOrphanTTL no longer claims that 404s are
cleaned immediately — the implementation gates them on the same TTL.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-04-30 15:34:16 +08:00
Bohan Jiang
da5dbc6224 refactor(repos): drop unused description + tighten create-project layout (#1930) 
* refactor(repos): drop unused description + tighten create-project layout

Two related changes that touch the workspace-repos surface together.

1. Remove the per-repo `description` field everywhere it was threaded.
   The only place it ever surfaced was a markdown table column the daemon
   wrote into the agent runtime config, where most rows just read "—"
   anyway. Agents already discover project structure by running
   `multica project` / `multica issue` against the CLI, so the human-
   readable description string carried no real value while taking up an
   extra Settings input row and propagating through six layers (settings
   UI → workspace.repos jsonb → handler RepoData → daemon RepoData →
   repocache.RepoInfo → execenv.RepoContextForEnv).

   - Settings → Repositories drops the description input; the URL field
     now spans the whole row.
   - WorkspaceRepo TS type loses `description`; backend RepoData /
     RepoInfo / RepoContextForEnv all collapse to URL only.
   - Daemon's runtime_config Repositories block changes from a
     `| URL | Description |` markdown table to a simple bullet list.
   - Tests updated; jsonb residue in existing workspaces is dropped at
     normalize time, so no migration needed.

2. Tighten the Create Project modal footer: pull the Status / Priority /
   Lead / Repos pills onto the same row as the Create Project button
   (Linear-style single-row footer) instead of stacking them above it,
   and swap the Repos pill icon from `FolderGit` to a real GitHub mark
   (lucide-react v1 dropped brand icons, so the mark lives inline as a
   small SVG component in this file).

   I tried promoting Repos to its own "Resources" strip above the footer
   to separate the resources abstraction from project metadata, but with
   a single pill it looked too sparse — leaving a TODO comment in the
   footer to revisit once we add Linear / Notion / Figma / Slack
   resource types.

* fix(daemon test): drop residual Description field on RepoData literals

* fix(repos): drop Description residue surfaced after rebase on #1929

Project-resource github_repo lift path (#1929) and registerTaskRepos
both still constructed RepoData{...Description: ...} after the rebase.
Two test sites in daemon_test.go and execenv_test.go also reintroduced
the field. Strip them so the Description-removal change builds and
tests pass with the latest main.
 2026-04-30 14:55:03 +08:00
Bohan Jiang
2129aa3dee feat(projects): project github_repo resources override workspace repos (#1929) 
* feat(projects): project github_repo resources override workspace repos

When an issue's project has at least one github_repo resource, the daemon
claim handler now sends only those as resp.Repos — workspace-level repos
are hidden to avoid mixing two repo lists in the agent prompt. With no
project github_repos (or no project), behavior is unchanged: workspace
repos are surfaced as before.

Lifts each project github_repo's url (and label, when present) into a
RepoData entry so `multica repo checkout` and the meta-skill render the
same URLs. The full structured list still ships at
.multica/project/resources.json for skills that want everything.

Adds TestProjectReposReplaceWorkspaceReposInMetaSkill covering the
rendering side. Docs updated to spell out the new precedence.

Co-authored-by: multica-agent <github@multica.ai>

* fix(daemon): allow project repo URLs through the checkout allowlist

When ClaimTaskByRuntime narrows resp.Repos to project github_repo URLs,
the daemon receives URLs that may not exist in the workspace's
GetWorkspaceRepos response. The existing checkout flow rejected those
with ErrRepoNotConfigured because the allowlist (and cache) was built
only from workspace-bound repos.

Adds registerTaskRepos in daemon.runTask: before agent spawn, merge
task.Repos into a new task-scoped allowlist (separate from the
workspace-scoped one so a workspace refresh doesn't wipe project URLs)
and kick off a background cache sync. ensureRepoReady now treats either
allowlist as valid.

Tests:
- TestRegisterTaskReposAllowsProjectOnlyURL — project-only URL is
  checkout-able and does not trigger a workspace-repos refresh
- TestRegisterTaskReposSurvivesWorkspaceRefresh — task URLs persist
  across refreshWorkspaceRepos
- TestClaimTask_ProjectGithubReposOverrideWorkspaceRepos — claim
  handler returns only project repos when present, no workspace leakage
- TestClaimTask_ProjectWithoutRepos_FallsBackToWorkspaceRepos — fall
  back to workspace repos when project has no github_repo resources

Docs updated to spell out the daemon-side allowlist behavior.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-04-30 14:37:51 +08:00
Bohan Jiang
b1345685a3 fix(task): rerun starts a fresh session, skip poisoned resume (#1928) 
* fix(task): rerun starts a fresh session, skip poisoned resume

When a task ended in a known agent fallback ("I reached the iteration
limit and couldn't generate a summary.", "Put your final update inside
the content string. Keep it concise.") the (agent_id, issue_id) resume
lookup would still pick that session, so a manual rerun inherited the
poisoned state and reproduced the same bad output.

Two complementary guards:

1. Daemon classifies poisoned terminal output and routes it through the
   blocked path with failure_reason set ('iteration_limit' /
   'agent_fallback_message'). GetLastTaskSession excludes failed tasks
   with those reasons, so even comment-triggered tasks no longer resume
   them. Tasks that failed mid-flight (timeout, runtime_recovery, etc.)
   are still resumable, preserving MUL-1128's auto-retry contract.

2. Manual rerun marks the new task force_fresh_session=true. The daemon
   claim handler skips the resume lookup entirely when the flag is set,
   capturing the user-intent signal that "the prior output was bad" even
   when poisoned classification misses a future fallback wording.

Auto-retry of orphaned mid-flight failures (MaybeRetryFailedTask →
CreateRetryTask) does not take this path, so it keeps resuming.

Tests: classifyPoisonedOutput unit test; integration tests assert the
SQL filter excludes poisoned classifiers, RerunIssue flips the flag,
and the normal enqueue path leaves it false.

Co-authored-by: multica-agent <github@multica.ai>

* fix(daemon): cap poisoned-output matcher to short trimmed text

GPT-Boy review on MUL-1630: the previous strings.Contains match would
classify any output that quoted the marker substring — including a
review/analysis that simply discussed the marker itself. Real fallback
messages are short single-sentence affairs, so cap the candidate at
~one paragraph and trim whitespace before matching. Adds regression
tests covering a long quoting review and a marker buried in a long
real conclusion; both must stay classified as completed.

Co-authored-by: multica-agent <github@multica.ai>

* fix(migrations): rename 065 force_fresh_session → 066 to clear collision

main introduced 065_project_resources after this branch was cut, so
both files shared the 065_ prefix. The readiness check
(server/cmd/server/health.go → migrations.LatestVersion) takes the
last entry by lexical order, which is 065_project_resources, leaving
this branch's 065_force_fresh_session unguarded — a deploy that
applied project_resources but not force_fresh_session would still
report ready, and the next enqueue / rerun / claim would crash on
"column force_fresh_session does not exist".

Renaming to 066_force_fresh_session puts it strictly after
project_resources so readiness blocks until it's applied.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-04-30 14:17:53 +08:00
Bohan Jiang
44608713bb feat(projects): typed project resources + agent runtime injection (#1926) 
* feat(projects): typed project resources + agent runtime injection

Adds a `project_resource` table that lets a project carry typed pointers
(github_repo today, more later) and surfaces them at agent runtime.

Server
- migration 065: project_resource (resource_type TEXT + resource_ref JSONB)
- sqlc CRUD + handler at /api/projects/{id}/resources
- claim handler attaches project_id/title + resources to issue tasks

Daemon
- TaskContextForEnv carries project context
- writes .multica/project/resources.json into workdir
- adds "## Project Context" block to CLAUDE.md / AGENTS.md / GEMINI.md
  via type-dispatched formatter so new resource types just add a case

CLI
- multica project create --repo <url> attaches repos in one step
- multica project resource add/list/remove

Frontend
- Project create modal: Repos pill (workspace repos + ad-hoc URL)
- Project detail sidebar: collapsible Resources section with attach/remove

Docs
- New "Project Resources" chapter explaining the abstraction and
  exactly what code to touch when adding a new resource type

Co-authored-by: multica-agent <github@multica.ai>

* fix(projects): transactional resources[] on create + generic CLI ref + test fix

Addresses review feedback on PR #1926:

1. CI red: TestProjectResourceLifecycle delete step called withURLParam
   twice, which replaced the chi route context and dropped the project id.
   Switched to the existing withURLParams helper from daemon_test.go.

2. POST /api/projects now accepts resources[] and attaches them in the
   same transaction as the project. Invalid refs roll back the whole
   create — no more half-attached projects on failure. Web modal + CLI
   `project create --repo` both use the new bundled payload.

3. CLI `project resource add` now accepts a generic --ref '<json>' flag
   so a new resource_type works without a CLI change. Per-type
   shortcuts (--url for github_repo) remain as a convenience but are no
   longer the only way in. Docs updated to drop the CLI from the
   "files you must touch" list.

Adds two new server handler tests:
- TestCreateProjectAttachesResources (resources[] happy path)
- TestCreateProjectRollsBackOnInvalidResource (transactional rollback)

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-04-30 14:00:43 +08:00
Bohan Jiang
72d5135bf0 fix(quick-create): subscribe requester to issues created via quick-create (#1924) 
The agent runs the daemon CLI, so issue.creator_type is `agent` and the
issue:created event listener only auto-subscribes the agent — not the
human requester. Result: the requester gets a single completion inbox
item but never sees follow-up comments or updates on their own issue.

Subscribe the requester (reason=`creator`, the only matching value
allowed by issue_subscriber's CHECK constraint without a migration)
inside notifyQuickCreateCompleted, after the issue lookup succeeds and
before the inbox write. Best-effort: log on failure, don't block the
inbox. On success, publish subscriber:added so the UI stays in sync
with manual subscribe and the listener-driven path.

Adds two integration tests in cmd/server: success path subscribes the
requester; failure path (agent finished without creating an issue)
leaves no subscriber rows.

Co-authored-by: multica-agent <github@multica.ai>
 2026-04-30 13:19:34 +08:00
Prince Pal
924c69114d feat(daemon): expose concurrent task slot env (#1889) 
* feat(daemon): expose concurrent task slot env

* fix(daemon): address task slot review nits
 2026-04-30 12:56:40 +08:00
Jiayuan Zhang
b6a3f8ed58 feat(daemon): add Co-authored-by trailer for Multica Agent to git commits (#1907) 
* feat(daemon): add Co-authored-by trailer for Multica Agent to git commits

Install a prepare-commit-msg hook in worktree bare repos that appends
"Co-authored-by: multica-agent <github@multica.ai>" to every commit
made by agents. Uses git interpret-trailers for proper formatting and
skips duplicates.

* feat(settings): add Co-authored-by toggle in workspace Labs settings

Add a workspace-level toggle to enable/disable the Co-authored-by
trailer for agent commits. Default is enabled (on).

Backend:
- Include workspace settings in daemon register response
- Store settings in daemon workspaceState
- Thread CoAuthoredByEnabled through WorktreeParams to conditionally
  install the prepare-commit-msg hook
- Parse co_authored_by_enabled from workspace settings JSONB

Frontend:
- Replace empty Labs tab placeholder with a Git section containing
  a Switch toggle for the Co-authored-by trailer setting
- Optimistically update the workspace query cache on toggle

* chore(daemon): skip squash commits in Co-authored-by hook

Test commit to verify the prepare-commit-msg hook appends the
Co-authored-by trailer automatically.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>
 2026-04-29 23:02:50 +02:00
Jiayuan Zhang
8c9c52b023 feat(inbox): add notification preferences to control inbox noise by event type (#1906) 
Users can now mute specific notification categories (assignments, status
changes, comments & mentions, priority/due-date updates, agent activity)
from Settings > Notifications. Muted event types are silently filtered at
notification creation time — no inbox items are created for muted groups.

- Add notification_preference table (migration 064)
- Add GET/PUT /api/notification-preferences endpoints
- Filter notifications in notifyIssueSubscribers, notifyDirect, and
  notifyMentionedMembers based on user preferences
- Add Notifications tab in Settings with per-group toggle switches
 2026-04-29 22:51:29 +02:00
Jiayuan Zhang
562949e1cb fix(daemon): prevent Quick Create from inventing requirements beyond user input (#1903) 
The description rule in buildQuickCreatePrompt() instructed the agent to
"always provide a rich, self-contained description" and "spell out what
needs to be done", which caused the agent to fabricate detailed product
specs, implementation phases, and design decisions from a one-line input.

Replace with a faithfulness-first rule: enrich with factual context
(fetched PR details, linked resources) but never invent requirements,
design decisions, or constraints the user did not express.

Fixes MUL-1605
 2026-04-29 21:12:17 +02:00
Jiayuan Zhang
aeccd4f26e feat(quick-create): enrich issue title and description with URL context (#1892) 
* feat(quick-create): enrich issue title and description with URL context

Update the quick-create agent prompt to fetch context from URLs in user
input (GitHub PRs, issues, web pages) before creating the issue. The
agent now produces semantically rich titles (e.g. "Review PR #123:
Refactor auth to OAuth2" instead of "review PR #123") and includes
summarized link content in the description so issues are self-contained.

* refactor(quick-create): let agent decide when to fetch URL context

Replace prescriptive URL enrichment instructions (hardcoded gh/WebFetch
commands) with goal-oriented guidance. The agent now uses its own
judgment to decide whether fetching referenced URLs would produce a
meaningfully better title/description, rather than being told exactly
which tools to use.

* fix(quick-create): always generate rich description for agent execution

The description was previously optional ("omit if simple request"). Since
quick-create issues are executed by agents, richer context leads to
better execution — update the prompt to always produce a substantive
description with actionable context.

* fix(quick-create): remove Chinese text from prompt, use English only

Replace Chinese examples in priority mapping and assignee matching with
language-agnostic English equivalents, per project coding rules.

* fix(quick-create): remove language-related hints from prompt

Agent doesn't need to be told about language handling — remove
"(in any language)" and "or equivalent in any language" qualifiers.
Keep prompt purely in English with no language-related content.
 2026-04-29 18:19:11 +02:00
Multica Eve
472e78022e fix: improve quick create inbox previews (#1883) 
Co-authored-by: Devv <devv@Devvs-Mac-mini.local>
Co-authored-by: Jiayuan Zhang <forrestchang7@gmail.com>
 2026-04-29 20:56:27 +08:00
Jiayuan Zhang
6a665c68a3 fix(inbox): improve quick-create notification to show issue title prominently (#1873) 
The inbox notification for quick-create showed "Created MUL-1577: <title>"
which truncated the actual issue title. Now the title field shows just the
issue title (the most useful info), and the detail label shows "Created
MUL-XXXX" as context.
 2026-04-29 13:54:08 +02:00
Bohan Jiang
cfa38df97b feat(quick-create): gate on daemon CLI version with pre-check + server enforcement (#1857) 
* fix(quick-create): bound dialog height + scroll editor when content overflows

Pasting a screenshot into the agent-create prompt expanded the editor
unbounded, which dragged DialogContent past the viewport since the agent
mode className had no max-height. Manual mode was unaffected because
manualDialogContentClass pins `!h-96`.

- Cap agent-mode DialogContent at `!max-h-[80vh]` (width stays
  `!max-w-xl`); short prompts still render compact, tall content stops
  at 80% of the viewport.
- Switch the editor wrapper to `flex-1 min-h-[140px] overflow-y-auto`
  so it absorbs the remaining vertical space inside the now-bounded
  DialogContent and scrolls internally instead of pushing the dialog.

* feat(quick-create): gate on daemon CLI version with pre-check + server enforcement

The agent-create flow depends on multica CLI behavior introduced in
v0.2.20 (URL attachment handling, no-retry semantics on
`multica issue create` failure — see PR #1851 / MUL-1496). Older
daemons either double-create issues on partial CLI failures or
mishandle pasted screenshot URLs. Per J's review on MUL-1496, gate
the flow at two layers — frontend pre-check for fast feedback,
server re-check as the trust boundary, both fail-closed on
missing/unparsable versions.

Server:
- New MinQuickCreateCLIVersion + CheckMinCLIVersion helper in
  pkg/agent (with sentinel errors for missing vs too-old).
- QuickCreateIssue handler reads runtime metadata.cli_version and
  returns a stable 422 { code: "daemon_version_unsupported",
  current_version, min_version, runtime_id } before enqueuing.
- The check runs after the existing online + ownership validation,
  so all rejections surface uniformly through the modal's existing
  error path.

Frontend:
- New @multica/core/runtimes/cli-version with the min version
  constant, parser, and runtime-metadata reader (tiny semver, no
  new lib dep).
- AgentCreatePanel resolves the selected agent's runtime, runs the
  same check, shows an inline amber notice below the agent picker
  when missing/too old, and disables the Create button.
- Submit handler also catches the server's 422 (defensive race —
  runtime can re-register between pre-check and submit) and
  surfaces the same wording in the error row.

Switching to manual create remains a clean escape hatch — manual
mode doesn't talk to a daemon at all, so an outdated CLI doesn't
block the user from filing the issue.
 2026-04-29 18:44:19 +08:00