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>
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>
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.
Closesmultica-ai/multica#2055.
* fix(codex): handle MCP elicitation server requests correctly
Fixes#1942.
handleServerRequest responded with {} to unrecognized Codex server
requests including mcpServer/elicitation/request. Codex 0.125+ expects
{action, content, _meta} for elicitation — the empty object causes a
deserialization error and the MCP tool call is reported as user-rejected.
Changes:
- Add mcpServer/elicitation/request case with correct response schema
- Add respondError helper for JSON-RPC error responses
- Return proper JSON-RPC method-not-found error for unknown server
requests instead of silent empty object
- Add tests for MCP elicitation and unknown method handling
* fix: use cfg.Logger instead of global slog in codex handleServerRequest
Switch the unhandled-server-request warning from global slog.Warn to
c.cfg.Logger.Warn for consistency with all other log calls in codex.go.
This ensures the warning appears in daemon run-logs and per-task
pipelines where operators look during triage.
* 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>
* 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
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>
Hermes ACP can flush queued session updates from the previous turn
before the current turn actually starts — both as session/resume
history replay and as chunks queued before our session/prompt response
streams. Without a gate those updates were appended to output and
re-emitted to the UI, so the previous answer appeared duplicated next
to the new one. Closes#1997.
PR #1789 added the acceptNotification hook field to hermesClient and
the call site in handleNotification, but never assigned it for Hermes,
so the guard short-circuited and every notification was processed.
This change mirrors the working Kiro pattern (kiro.go:87/97/240):
- declare a streamingCurrentTurn atomic.Bool in the backend.
- assign acceptNotification, onMessage, onPromptDone gates that all
return early when the flag is false.
- flip the flag to true immediately before c.request("session/prompt").
Adds TestHermesClientAcceptNotificationGate as a regression test that
exercises the gate directly on hermesClient.
Verified with `go test ./pkg/agent`.
Co-authored-by: multica-agent <github@multica.ai>
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>
* 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>
Latest Codex CLI ships with GPT-5.5 / GPT-5.5 mini, but the static
catalog still topped out at GPT-5.4 so users couldn't pick the new
model from the agent picker.
Add gpt-5.5 + gpt-5.5-mini to codexStaticModels and promote 5.5 as
the default badge. Keep the older 5.4 / 5.3-codex / gpt-5 / o3
entries for users on older Codex CLI builds. Add a regression test
mirroring TestGeminiStaticModelsExposesAliasesAndGemini3 so the
next OpenAI release isn't a silent miss.
Co-authored-by: multica-agent <github@multica.ai>
- 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>
* fix(cli): make `multica login --token` accept the PAT as a value
The flag was registered as a Bool, so `multica login --token <PAT>` parsed
`--token` as `true` and dropped the supplied value as an unused positional
argument, then unconditionally prompted "Enter your personal access token:".
This contradicted the user-facing docs (`cli.mdx`, `CLI_AND_DAEMON.md`,
the in-app `connect-remote-dialog`) which show `--token <mul_...>`.
Switch `--token` to a String flag. Both `--token mul_...` and
`--token=mul_...` now bind the value and skip the prompt. Passing
`--token=` with an empty value (or `multica login --token=""`) still
falls through to the interactive prompt for users who don't want the
token in shell history. Updates the few internal docs that showed the
no-value form.
Fixes#1994
Co-authored-by: multica-agent <github@multica.ai>
* fix(cli): preserve `multica login --token` (no value) prompt path and tighten regression test
Addresses review feedback on #2017:
1. Restore the legacy no-value form. After the prior commit, `multica
login --token` (no value) errored with `flag needs an argument:
--token`, which broke the CLI_INSTALL.md / CLI_AND_DAEMON.md flow for
headless users. Set `NoOptDefVal` on the `--token` flag to a sentinel
that runAuthLoginToken treats as "prompt me," so:
- `--token mul_xxx` and `--token=mul_xxx` consume the value (the
#1994 fix is preserved),
- `--token` alone falls through to the interactive prompt,
- `--token=""` (explicit empty) also prompts.
pflag with `NoOptDefVal` won't bind the next positional as the flag's
value, so runAuthLogin recovers `--token mul_xxx` (the form from
#1994) by promoting a single positional arg into the token. loginCmd
gains `Args: cobra.MaximumNArgs(1)` so multi-positional typos still
error fast.
2. Tighten regression coverage. Split into TestLoginTokenFlagWiring
(asserts the production loginCmd.Flags().Lookup("token") is a String
flag with the prompt-mode NoOptDefVal — would fail if anyone reverts
the flag to Bool) and TestLoginTokenFlagParsing (drives all five
documented invocation forms through the same flag wiring + the
runAuthLogin space-form recovery). The synthetic-only test that the
reviewer flagged is gone.
Co-authored-by: multica-agent <github@multica.ai>
---------
Co-authored-by: multica-agent <github@multica.ai>
* 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.
* 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>
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>
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>
* 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>
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.
* 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>
* 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.
* 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>
* 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>
* 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>
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>
* 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>
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
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
* 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.
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.
* 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.
A complete UX upgrade for chat sending → receiving → recovering.
* StatusPill replaces the orphan spinner — stage-aware copy
("Reading files · 12s", "Searching the web · 14s", "Typing · 24s"),
shimmer text, monotonic timer, derived effective status, > 60s
warning tone, > 5min cancel button.
* WS writethrough on task:queued / task:dispatch / task:cancelled so
pendingTask cache stays in sync with the daemon state machine without
invalidate-refetch latency. broadcastTaskDispatch now includes
chat_session_id when the task is for a chat session — the existing
payload only carried it on the generic task: events, leaving the pill
stuck at "Queued" until completion.
* Failure fallback — FailTask writes a chat_message tagged with
failure_reason (mirrors the issue path's system comment, gated on
retried==nil). Front-end renders an inline note ("Connection failed",
with a Show details collapsible) instead of the previous black hole.
* Elapsed timing — chat_message.elapsed_ms persists task.completed_at -
task.created_at on success/failure rows. UI shows "Replied in 38s" /
"Failed after 12s" beneath assistant bubbles. Format helper shared
between StatusPill and the persisted caption so the live timer and
final reading never disagree.
* Optimistic burst rebalanced — pendingTask seed + created_at moved
before the HTTP roundtrip so the pill appears the instant the user
hits send; handleStop is fire-and-forget so cancel feels immediate
(server confirmation arrives via task:cancelled WS).
* Presence integration — chat avatars use ActorAvatar (status dot +
hover card); OfflineBanner above the input on offline/unstable;
SessionDropdown shows per-row in-flight/unread pip plus a
cross-session aggregate pip on the closed trigger.
* Editor blur on send so the caret stops competing with the StatusPill
/ streaming reply for the user's attention.
* Chat panel isOpen now persists globally; defaults to OPEN for new
users (storage key absence) so the feature is discoverable. Existing
users' prior choice is respected.
* DB: migrations 062 (failure_reason) + 063 (elapsed_ms), both
ADD COLUMN NULL — fast, non-blocking, backwards compatible.
* WS: task:failed chat path now invalidates chatKeys.messages — fixes
a pre-existing bug where the failure bubble required a page refresh
to appear.
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(execenv): default-disable Codex native multi-agent in per-task config
Recent Codex app-server releases enable features.multi_agent by default,
exposing spawn_agent / wait / close_agent tools that let a parent thread
spawn nested subagents. The daemon currently models only the parent thread,
so the parent's turn/completed is treated as task completion even when
spawned children are still running — leading to premature task completion
and dropped child output.
Disable features.multi_agent by default in the per-task CODEX_HOME/config.toml
so Multica's task lifecycle is the only orchestration layer in play. Strip
both the dotted-key form (features.multi_agent) at TOML root and the
multi_agent key inside a [features] table; siblings and unrelated tables
are preserved. Honor MULTICA_CODEX_MULTI_AGENT=1 as an opt-out for users
who explicitly want Codex native subagents inside a Multica task.
The user's global ~/.codex/config.toml is never modified — only the daemon's
isolated per-task copy.
Also widen managedBlockRe to consume `\n*` rather than `\n?` so reruns
don't accumulate blank lines when both the sandbox and multi-agent managed
blocks coexist.
* fix(execenv): inject managed multi_agent inside existing [features] table
Per PR review (codex_multi_agent.go:77-83 vs :112-115): when the user's
config.toml already has a top-level `[features]` table, writing
`features.multi_agent = false` at the TOML root implicitly redefines the
same `features` table. The strict TOML parser used by Codex (`toml-rs`)
rejects that with `table 'features' already exists`, so Codex would fail
to load the per-task config and refuse to start the thread. Verified the
strict-parser failure with pelletier/go-toml/v2; the previous
BurntSushi/toml-based regression test was permissive enough to miss it.
Detect a root-level `[features]` header and place the managed block
inside that table (`multi_agent = false` with marker comments). When no
such header exists, keep the existing root-level dotted-key form. The
managed-block regex matches both layouts so reruns and layout
transitions stay idempotent. A `[features.experimental]` sub-table
without a bare `[features]` header still uses the root dotted-key form,
which is spec-valid (no explicit redefinition).
Tests now use pelletier/go-toml/v2 to actually parse the output and
assert features.multi_agent decodes to false; the regression case from
the PR review is covered explicitly.
* fix(execenv): recognize feature table header variants
---------
Co-authored-by: Devv <devv@Devvs-Mac-mini.local>
* feat(server/auth): cache PAT lookups in Redis with 60s TTL
Personal access tokens used to hit Postgres on every request: a SELECT
to resolve token_hash → user_id, plus a fire-and-forget UPDATE of
last_used_at. For a CLI / daemon making many requests per second this
is wasted DB load — the token is the same and the answer hasn't changed.
Add a Redis-backed cache (auth.PATCache) keyed by token hash, TTL 60s:
- On cache hit, the auth middleware skips both the SELECT and the
last_used_at UPDATE. last_used_at is now refreshed at most once per
TTL window per token, not per request.
- On cache miss the middleware falls back to today's behavior: query
Postgres, populate the cache, async-update last_used_at.
- On revoke, the handler invalidates the cache entry so revocation
takes effect immediately rather than waiting for the TTL to expire.
This required changing RevokePersonalAccessToken from :exec to :one
RETURNING token_hash.
The cache is nil-safe: when REDIS_URL isn't configured, NewPATCache
returns nil and the middleware degrades to today's always-hit-DB
behavior. JWT validation is untouched (already DB-free).
Tested with REDIS_TEST_URL — same gating pattern the rest of the
suite uses for Redis-backed tests. New tests cover nil-safety, set/
get/invalidate, TTL, and the middleware short-circuit on cache hit.
* fix(server/auth): clamp PAT cache TTL to token's remaining lifetime
GPT-Boy review caught: a PAT expiring in <60s would still be cached
for the full PATCacheTTL window, so the token could continue passing
auth on cache hit for up to ~60s after its expires_at. The DB query
filters expired tokens (revoked = FALSE AND expires_at > now()), but
that filter never ran on a cache hit.
Make Set take an explicit ttl, and add TTLForExpiry to compute it:
- no expires_at → full PATCacheTTL
- expires_at far → full PATCacheTTL
- expires_at <60s → time until expiry
- already expired → 0, Set skips caching (TOCTOU defense between
the SELECT and the Set, since the SELECT
already filters expired rows)
Regression test pins the clamp behavior end-to-end against Redis.
* feat(server/auth): cache daemon-token + PAT lookups in DaemonAuth, bump TTL to 10m
Daemon /api/daemon/* requests (heartbeat, claim task) hit DaemonAuth
which previously did its own GetDaemonTokenByHash on every request and
*also* duplicated the PAT lookup on the mul_ fallback — bypassing the
cache added in 1cdd674c. Today's daemons authenticate via mul_ PATs
(mdt_ minting isn't wired up yet), so the duplicate PAT path is the one
that actually matters for hot-path DB load.
Three changes:
1. New auth.DaemonTokenCache mirrors PATCache for the mdt_ path
(key = mul:auth:daemon:<sha256>, JSON value = {workspace_id, daemon_id}).
Forward-looking infrastructure for when daemon tokens get minted; the
middleware short-circuits the DB SELECT on cache hit. TTL clamped to
the token's expires_at via the shared TTLForExpiry helper.
2. DaemonAuth now also consults PATCache on its mul_ fallback, sharing
the same cache as the regular Auth middleware. A daemon making 4 hb/min
collapses from 4 GetPersonalAccessTokenByHash + 4 last_used_at writes
per minute to ~1 of each per AuthCacheTTL window (~10 minutes).
3. Rename PATCacheTTL → AuthCacheTTL and bump from 60s to 10 minutes.
The constant is now shared between PAT and daemon caches; 10m matches
the user-requested longer TTL for further DB write reduction. Revoke
latency on the happy path is still instant via active invalidation;
the worst-case (Redis Del miss / direct-DB revoke) grows from ~60s to
~10m.
Tests cover nil-safety, set/get/invalidate, TTL, clamped TTL on near-
expiry tokens, and the middleware short-circuit for both cache paths
(mdt_ via DaemonTokenCache, mul_ fallback via PATCache).
* feat(server/auth): cache PAT lookups on the WebSocket auth path
The third place a PAT is resolved — patResolver.ResolveToken used by
realtime.HandleWebSocket — was still hitting Postgres on every /ws
auth and firing an unconditional last_used_at UPDATE, bypassing the
cache added in 1cdd674c. Wire it through the same shared PATCache so
revoking a token through any path (Auth middleware, DaemonAuth PAT
fallback, or WS auth) hits all three caches with one Invalidate.
Also leaves a comment on DeleteDaemonTokensByWorkspaceAndDaemon —
the query has no caller today, but a future deregister/rotate flow
must remember to call DaemonTokenCache.Invalidate(hash) for each
deleted row, otherwise deleted daemon tokens stay valid until TTL.
PR #1744 fixed literal `\n\n` rendering for the CLI surfaces (`issue
create / update --description`, `issue comment add --content`) but the
agent-completion path bypasses the CLI entirely: the daemon POSTs the
agent's stdout to `/api/daemon/tasks/:id/complete`, and `TaskService.
CompleteTask` writes `payload.Output` straight into `createAgentComment`
and `CreateChatMessage` without decoding. Models (e.g. Codex) routinely
emit Python/JSON-style `\n` literals in their final output, which then
land in the DB as the 4-char escape sequence and render as one wall of
text in the issue/chat panel — exactly the bug report in #1820.
- Move `unescapeFlagText` from `server/cmd/multica/cmd_issue.go` to
`server/internal/util/text.go` as `UnescapeBackslashEscapes` so the
CLI and the service layer share one implementation. The full
contract-boundary test suite moves with it.
- Apply `UnescapeBackslashEscapes` to `payload.Output` before it
reaches `createAgentComment` and `CreateChatMessage` in
`TaskService.CompleteTask`. Same `\n / \r / \t / \\` decoding as the
CLI; other escape sequences (`\d`, `\w`, `\u`, etc.) pass through
verbatim so regex/format strings in agent output survive.
Closes#1820
Two coordinated fixes for a quick-create case where the agent ended up
creating duplicate issues. Repro: user pasted an image into the
quick-create prompt; the front-end uploaded it and embedded the URL as
markdown in the user input; the agent saw the URL, assumed it was an
attachment, and ran `multica issue create … --attachment "https://…"`.
The CLI POSTed the issue first, then failed to read the URL as a file
(`os.ReadFile("https://…")`) and exited 1. The agent treated exit 1 as
"create failed" and retried — but the first issue already existed, so
the workspace ended up with two of them.
CLI (`server/cmd/multica/cmd_issue.go`):
- `runIssueCreate` pre-validates `--attachment` BEFORE POSTing. URLs are
warned about and skipped (they are never local files); local-path
read errors fail before the issue is created so no half-baked issue
lands. Once the POST succeeds, post-create upload failures only
print a stderr warning and the issue metadata is still emitted —
never a non-zero exit, so callers cannot mistake "attachment upload
hiccup" for "create failed" and retry.
- `runIssueCommentAdd` already uploads attachments BEFORE the comment
is created, so its failure mode is fine; it just gets the same
URL-skip behaviour for consistency.
Quick-create prompt (`buildQuickCreatePrompt`):
- Tells the agent NOT to pass `--attachment` for prompt-embedded image
URLs (they are already part of the description as markdown).
- Hardens the "no retry" rule: even on a non-zero exit, do not retry
`issue create` — the issue may already exist.
GitHub #1839: when an issue is reassigned from agent A to agent B, B
often only reads the issue body and misses context A added in comments
(e.g. which repo to clone). The assignment-triggered workflow injected
into CLAUDE.md / AGENTS.md said "Read comments for additional context
or human instructions" — vague enough that agents routinely skipped
it. The comment-triggered branch already gives an explicit
`multica issue comment list` invocation, so behavior diverged.
Promote step 3 to a concrete CLI call, mark it mandatory, and surface
the most common failure mode (stale instructions on reassignment) so
the agent recognizes when it matters. Reorder so comments are read
*before* flipping status to `in_progress`, matching how a human would
catch up on a thread before claiming work.
* feat(quick-create): default assignee to picker agent when user didn't name one
The quick-create prompt previously told the agent to OMIT --assignee
when the user's input didn't name a person. That left almost every
quick-created issue unassigned, which doesn't match user intent — the
user opened quick-create with a specific agent picked, so that agent
is the obvious owner.
Both prompt surfaces (BuildPrompt for the dispatched message, plus
the workflow block in injected CLAUDE.md / AGENTS.md) now instruct
the agent: if the input doesn't name an assignee, pass
`--assignee "<your name>"`. The picker agent's name is interpolated
into the prompt at task-build time so the agent has a literal value
to use rather than guessing its own name. The "explicitly named
assignee → resolve via members" branch is unchanged.
* refactor(execenv): drop duplicated quick-create field rules from CLAUDE.md/AGENTS.md
The quick-create field rules (title / description / priority / assignee
fallback / project / status) lived in two places — the per-turn user
message built by BuildPrompt, and the workflow block injected into
CLAUDE.md / AGENTS.md by buildMetaSkillContent. Same content, two
sources, easy to update one and forget the other (the assignee-default
change in this PR had to touch both).
Quick-create is one-shot, so the per-turn user message is always
present and is the natural single source of truth. The injected
file's quick-create section now keeps only the hard guardrails:
"do exactly one issue create, no issue get / status / comment add,
exit on CLI error". Those guardrails stay in BOTH surfaces because
they're the safety net for providers that don't propagate the user
message into resumed-session context.
renderQuickCreateContext (issue_context.md) was already
guardrails-only — no change needed there.
