A rate-limited reviewer filled a task's Chat tab with 14 identical
"Reviewer using model: ..." markers and no review text, hammering an
already-limited provider.
Root cause: the reviewer was the only AI lane that never classified
provider errors, so a 429 became an UNAVAILABLE verdict. With no
validator fallback configured the fallback ladder re-ran the SAME model
instantly, and fn_review_step answered with "code review remains
blocking; retry once" — bounding the loop with prompt text rather than
code. The tool's catch-all also swallowed the error into tool output, so
withRateLimitRetry, UsageLimitPauser and RetryStormError never fired.
- reviewer: throw ReviewerProviderError for usage-limit/transient errors
instead of laundering them into UNAVAILABLE, and never spend the
fallback budget (which bounds bad reviews) on an outage.
- reviewer: absorb flaky-network blips in-lane via withRetry with
jittered backoff; rate limits still escalate immediately.
- executor: re-raise the fatal after the prompt via
throwDeferredReviewerFatal — pi-agent-core converts tool throws into
tool_error results, so a tool cannot throw out of session.prompt().
- executor: give code review a real MAX_CODE_REVIEW_UNAVAILABLE_RETRIES
counter, mirroring the plan/spec limiter.
- reviewer: dedupe the model marker on text, so same-model retries stay
silent while a genuine model switch still emits.
Also fixes the run-on rendering: AgentLogType gains `status` for complete
engine messages. `text` means "streamed delta" and is re-glued with
join(""), which is why N standalone markers rendered as one string. The
split is at the type, not a separator — a separator would reintroduce the
FN-5787/5789/5803 streamed-spacing regression.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Route workflow selections, model lanes, goals, skills, and reliability reads through project-scoped async stores. Recover heartbeat agents parked against an unrelated project model and preserve workflow JSONB patches atomically.
## Summary
- add asynchronous PostgreSQL parity to research commands and engine
execution paths
- persist Roadmap, Compound Engineering sessions, and WhatsApp state in
PostgreSQL
- harden cancellation, concurrency, reconnect, replay-claim, and
detached-promise behavior
- bundle the PostgreSQL-backed integration implementations in the
published CLI
This is PR 2 of 2 and is intentionally stacked on #2088. It contains 44
changed files; merge #2088 first, then retarget this PR to `main` if
GitHub does not do so automatically.
## Verification
- `pnpm check:changesets --strict`
- `pnpm lint`
- `pnpm test:gate`: 463 tests passed
- Compound Engineering plugin: 299 tests passed
- Roadmap plugin: 144 tests passed
- WhatsApp plugin: 27 tests passed
- research CLI: 18 tests passed
- `pnpm verify:fast`: all scoped typechecks, builds, CLI build, and boot
smoke passed
## Post-Deploy Monitoring & Validation
- deploy only after #2088 and verify schema migration `0002` is present
- monitor research cancellation, automation claims, agent execution,
plugin schema initialization, and unhandled rejections
- validate Roadmap ownership, Compound Engineering session recovery, and
WhatsApp reconnect/replay deduplication
- compare per-project plugin and workflow counts after cutover
- restore the pre-deploy backup for data rollback; avoid an in-place
schema downgrade
# Migrate storage from SQLite to PostgreSQL — full dashboard cutover
Migrates Fusion's storage layer to the embedded PostgreSQL
`AsyncDataLayer` (the default backend) and **completes the
satellite-store + feature cutover** so every dashboard and Command
Center surface works in PG mode.
## Status — every surface works in embedded-PG mode
Verified live against a running embedded-Postgres dashboard (all
**200**, zero 5xx) and gate-tested (**23 files / 99 tests** on embedded
PG, plus engine-core 294 and ci-shape 63 in the blocking merge gate;
core/engine/cli/dashboard typecheck clean).
| Area | Surfaces | State |
|---|---|---|
| Satellite stores | workflows, todos, insights, research, missions,
goals, mailbox | ✅ |
| Views | artifacts, documents, evals | ✅ |
| Command Center | activity, productivity, team, tokens, tools,
**workflows**, **github**, **signals**, **plugin-activations**, **live**
(all 10) | ✅ |
| Run execution | insight generation, research run execution | ✅
(store-path; AI step needs a provider) |
| Live updates | SSE push for mission/research/insight events | ✅ |
| Workflow editing | create / update / delete / select (+ id counter) |
✅ |
| Engine | mission autopilot, incident-signal ingestion, regression
storm-guard, agent wake-on-message | ✅ |
| Core | tasks, agents, secrets, automations, memory, chat, usage, PRs,
git | ✅ |
## Approach
Each satellite store gets an `Async<Store>` wrapper exposing the sync
store's method names over the existing `async-*-store.ts` helpers;
`get<Store>Store()` returns a `Sync | Async` union; consumers `await`
(harmless on sync), and engine/CLI paths that can't convert use
`instanceof Sync` graceful fallback. Analytics aggregators branch on
`"ping" in dbOrLayer` to run schema-qualified raw SQL over `project.*`
(snake_case) in PG. Executors/orchestrators/autopilot are
await-converted to drive the union store; the async store wrappers
extend `EventEmitter` so SSE live-push fires in both backends.
Not-yet-ported capabilities degrade gracefully (never 500) and are
individually called out in commits.
## Sync with main
The branch is kept continuously merged with `main` (currently through
FN-7845, 2026-07-12); the earlier "final rebase deferred" note no longer
applies. Use **Create a merge commit** (or squash) to land it — GitHub's
rebase-merge cannot replay a merge-maintained branch.
## Residual Review Findings
Multi-agent code review of the PostgreSQL satellite-store ports (U1–U5)
applied 3 safe fixes (see `fix(review): apply autofix feedback`). The
following are **real but gated** — recorded here as follow-up work
rather than auto-applied. All are SQLite→PostgreSQL
**concurrency/atomicity regressions**: the sync stores were immune only
by SQLite's single-writer, single-threaded-handler execution; the async
ports open multi-await read-modify-write windows. **Reachability is low
today** because the execution engines that generate concurrent same-run
mutations (insight run executor, research orchestrator/dispatcher) are
`instanceof`-gated to sync mode in PG. No process-crash class survived
(all engine fallbacks correctly guard the sync store).
- **[P1] Research `appendResearchEvent` dual-write is non-atomic**
(`packages/core/src/async-research-store.ts`, corroborated: adversarial
+ reliability). The `research_run_events` insert (own transaction) and
the `run.events` jsonb update are separate writes — a crash between
them, or two concurrent appends, splits the table count from the jsonb
array. **Fix:** perform the seq-insert and the jsonb update in one
`layer.transactionImmediate`.
- **[P1] Research run terminal-reversion via stale full-row persist**
(`async-research-store.ts` `persistResearchRun`/`updateResearchStatus`).
Concurrent `PATCH /runs/:id/status` + `POST /runs/:id/events` can revert
a terminal run to `running` by overwriting the whole row, bypassing the
transition guard. **Fix:** scoped column `UPDATE`s with a `WHERE status
…` guard, or optimistic version column.
- **[P2] `updateResearchRun`/`updateInsightRun` read-then-write TOCTOU**
— concurrent PATCHes last-writer-wins on the lifecycle merge. **Fix:**
`SELECT … FOR UPDATE` / enclosing transaction.
- **[P2] `upsertRun`/`createRunOrThrowConflict` check-then-create race**
(`async-insight-store.ts`) — two callers can each create an "active"
run. **Fix:** partial unique index on `(projectId, trigger) WHERE status
IN ('pending','running')`.
- **[P3] `createResearchRetryRun` return-value divergence** — sync
returns the pre-update `queued` snapshot; async returns the reloaded
`retry_waiting` run (persisted state is identical). Pick one side for
cross-backend parity.
- **[P2/perf] Mission `getMissionWithHierarchy`/`getMissionHealth` N+1
fan-out** — O(milestones×slices) sequential round-trips hold one pool
slot per request; can starve the pool for large hierarchies. **Fix:**
batched/joined reads.
- **Testing gaps:** no PG-mode concurrency tests (interleaved
status/event mutations), no sync↔async parity assertion for the
lifecycle-error codes, and no mission status/health rollup parity test
vs the sync `MissionStore`.
~~Out of scope (deferred): AI run *execution* (insight/research) +
mission autopilot + live SSE mission events remain sync-gated/degraded
in PG mode.~~ **Since ported** — insight/research run execution, mission
autopilot, and SSE live push all run on the async layer now, which also
makes the concurrency findings above genuinely reachable; they remain
open follow-ups.
---
## Update — 2026-07-12: production-readiness hardening & live acceptance
Everything below landed on this branch since the description above was
written:
**Production blockers from review — fixed**
- `recoverStaleTransitionPending` ported to the async layer (backend
moves write + clear the crash-safe marker; startup/maintenance sweeps no
longer throw).
- Lost-update class fixed: `atomicWriteTaskJson`/`WithAudit` write
changed columns only (full-row upserts silently resurrected stale fields
across concurrent store instances — the "task stuck unplanned forever"
bug).
- First-boot **auto-migration**: booting the PG backend over a project
with a legacy `fusion.db` migrates it automatically (loud failure,
SQLite kept as backup), and the dashboard shows a one-time **"your data
was migrated" banner** with the backup paths and a Need-help Discord
link.
- `pg_dump`/`pg_restore` discovered from common install locations for
embedded-mode backups.
- The PG suite is part of the blocking merge gate (`test:pg-gate`).
**Multi-project isolation (PR #2007, merged into this branch)**
- `project_id` partition key on tasks / archived tasks / config,
`taskProjectScope` threaded through every scan/claim/count, per-project
config rows, layer bound to the project at startup.
- Review P1 follow-up: the shared cold-storage `archive.archived_tasks`
table is also partitioned and all archived-board reads/counts/searches
are scoped.
- Schema drift self-heal generalized to schema-qualified columns so
existing databases upgrade in place.
**Other changes**
- Node settings sync **removed** in PG mode (409
`settings-sync-disabled-postgres`) — nodes share state by connecting to
the same database; auth sync kept (per-machine file).
- Perf (review findings): `listTasks` pushes column filter + ORDER BY +
LIMIT/OFFSET into SQL; `getConversation` capped to the most recent 200
messages.
- Fixed a false "operator action required" pause-abort log fired on
every successfully auto-merged task.
**Live acceptance — PASSED (2026-07-12)**
A sandboxed instance (isolated HOME, embedded PG, real Opus executor)
ran a task through the complete cycle: create → triage (AI spec) →
execute → in-review → AI squash-merge landed on the project's `main` →
done. A write+read sweep of every data surface (settings, comments,
documents, attachments + artifact bridge + artifact edit, chat with real
generation, goals, missions, agent mail, secrets, workflows, memory, CC
analytics) was green on embedded PG.
**Known remaining work**
- The per-project `config` PK re-key has no upgrade path for
pre-isolation embedded-PG databases (needs a real `DROP
CONSTRAINT`/re-key migration; fresh databases are fine).
- `pg_dump`/`pg_restore` binaries are not yet bundled in release
artifacts (PATH/common-location discovery only).
- The satellite-store concurrency findings listed above.
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Phil Larson <hello@phillarson.xyz>
Co-authored-by: fusion-merge <fusion-merge@local>
Artifact-registration mailbox notifications now render a shared inline preview and open-artifact link instead of plain text metadata.
- Add MailboxArtifactAttachment component rendering an inline image/document preview plus an "open artifact" link from message.metadata (artifactId/artifactType/mimeType) via artifactMediaUrl
- Wire MailboxModal and MailboxView to render the new attachment for artifact-registered messages, with supporting CSS
- Emit metadata.mimeType from notifyArtifactRegistered in agent-tools.ts so mailbox surfaces can pick the right preview affordance without an extra artifact fetch
- Add/extend tests for the new component and for MailboxView/agent-artifact-tools coverage
- Update dashboard guide docs and add a changeset for the feature
Files changed:
.changeset/fn-7864-artifact-mail-link.md | 7 ++
docs/dashboard-guide.md | 2 +-
.../app/components/MailboxArtifactAttachment.tsx | 103 +++++++++++++++++++++
packages/dashboard/app/components/MailboxModal.css | 74 +++++++++++++++
packages/dashboard/app/components/MailboxModal.tsx | 15 +++
packages/dashboard/app/components/MailboxView.tsx | 15 +++
.../__tests__/MailboxArtifactAttachment.test.tsx | 65 +++++++++++++
.../app/components/__tests__/MailboxView.test.tsx | 93 +++++++++++++++++++
.../src/__tests__/agent-artifact-tools.test.ts | 32 ++++++-
packages/engine/src/agent-tools.ts | 5 +
10 files changed, 409 insertions(+), 2 deletions(-)
Fusion-Task-Id: FN-7864
Fusion-Task-Lineage: a6502e18-5f7f-4c67-80fb-a709e4a52c50
Co-authored-by: Fusion (runfusion.ai) <noreply@runfusion.ai>
Issue #2015: product-code executor tasks were repeatedly routed to a
liaison-only agent because every routing path gated only on the coarse
role field, and several binding primitives had no guard at all.
- Add runtimeConfig.assignmentPolicy ("auto" | "explicit-only" | "none");
"none" can never be bound to implementation tasks by ANY path — no
override bypasses it (the liaison guarantee)
- Route every binding surface through one shared evaluator
(evaluateImplementationTaskBind): claimTaskForAgent, the previously
unguarded checkoutTask/assignTask primitives, selectNextTaskForAgent
(including the in-progress re-selection loop), scheduler auto-assign
pool, heartbeat inbox/auto-claim, fn_delegate_task, CLI agent-id
validation, and dashboard assign/checkout/inbox routes
- Lock project isolation with a regression test: a foreign-project
agent id is rejected by every binding primitive
- Expose Assignment Policy in Agent Detail settings; document in
docs/agents.md; add changeset
Fusion-Task-Id: FN-7851
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- contain fn_artifact_register path payloads: realpath-canonicalized
containment before stat/read — relative paths require and must stay
inside baseDir, absolute paths allowed only under baseDir or the OS
temp dir (deliberate allowance for browser/screenshot tooling);
the process.cwd() fallback is gone, symlink escapes rejected
- bind task-scoped heartbeat artifact registration to the acquired
worktree (baseDir: sessionCwd rebind after acquisition); no-task
heartbeat prompt now says to pass absolute temp-dir paths
- enforce exactly-one payload source (content/uri/dataBase64/path);
content+uri combos are now rejected to match the documented contract
- add FNXC rationale comments at both visual-artifact instruction sites
in the planning prompts (sync contract with the executor prompt)
- media route: statSync -> await stat from node:fs/promises
- range tests ride the in-memory MockSocket harness (TestResponse gains
binary-safe bodyBuffer; real-TCP helper deleted) and assert the full
206 Content-Range/Content-Length contract for every range form
- add PdfViewer coverage (iframe src/title) in DocumentsView tests
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Video was registrable but effectively unusable, and HTML/PDF deliverables
had no first-class path from agents to the gallery.
- media route now serves HTTP byte ranges (Accept-Ranges, 206 +
Content-Range, 416 on unsatisfiable) so <video>/<audio> seeking works
and Safari plays media at all
- video attachments (mp4/webm/mov, 100MB cap vs 5MB for other types)
bridge into the artifact registry like images; multer transport ceiling
raised to 100MB with per-type caps enforced in the store
- fn_artifact_register path payloads are signature-validated for video
(ftyp box / EBML header) and PDF (%PDF- prefix), mirroring images
- HTML doc artifacts (mimeType text/html) render as live sandboxed
iframe previews by default in the doc viewer, with a Preview/Source
toggle and the same FileEditor edit mode
- executor/heartbeat/planning prompts and tool descriptions now cover
the full type matrix: images, videos, audio, HTML mockups, PDFs, and
markdown docs, each with the registration recipe
Verified live: range requests (200/206/416) via curl, an ffmpeg-generated
mp4 playing to completion in the gallery lightbox, and an interactive
HTML mockup rendering in the sandboxed preview.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Agents could never get screenshots/wireframes/mocks into the Artifacts view:
fn_artifact_register was gated on assignedAgentId (never set in default
ephemeral mode), the only image payload source was inline base64, and no
prompt ever told agents to register visual deliverables.
- always expose fn_artifact_register to executor sessions ("executor" author
fallback), resolve relative paths against the task worktree, and default
taskId to the executing task (heartbeat task lane too)
- add a `path` payload source: file read with 50MB cap, extension MIME
inference, PNG/JPEG/GIF/WebP signature + SVG sniff validation, persisted
through managed artifact storage
- executor/heartbeat/planning prompts + engine-tools reference now instruct
agents to register screenshots, wireframes, mockups, and recordings
- new ArtifactsGallery: Images/Docs/PDFs/Videos/Audio/Other category sections
and filter chips, visual tile grid + lightbox, embedded PDF viewer, audio
player rows, download rows; mobile-responsive down to the 768px breakpoint
- doc artifacts open a full viewer rendered as markdown by default with an
in-place edit mode using the shared CodeMirror FileEditor; persisted via new
GET/PATCH /api/artifacts/:id + TaskStore.updateArtifact and live-refreshed
through the new artifact:updated SSE event
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Agents that must edit files beyond a task's declared ## File Scope had no
way to keep the scope in sync, so those edits were stranded at merge (the
squash merge is scoped to ## File Scope, and cross-task overlap blocking +
the merge file-scope invariant both read it).
New executor tool fn_task_file_scope_add validates repo-relative
paths/globs with isValidFileScopeEntry, de-dupes against existing scope,
appends them to the ## File Scope section of PROMPT.md, and persists via
store.updateTask({ prompt }) (same validation + task.json/PROMPT.md sync as
fn_task_prompt_write). Registered in the main coding-agent tool list; the
base executor prompt now instructs the agent to call it when editing beyond
the declared scope. Merge-time peer-claim refusal is unchanged and remains
the cross-task backstop.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Task creation surfaces stopped hardcoding column:"triage", so new tasks now land in the selected-or-default workflow's resolved intake column instead of always jumping to Planning/triage.
- Removed hardcoded column:"triage" override in engine's createTaskCreateTool (fn_task_create), letting TaskStore.createTask resolve the landing column from the workflow's intake-trait column.
- Removed the equivalent hardcoded override in the pi extension's fn_task_create, and updated its response text to echo the actual landing column instead of a fixed "Column: triage" string.
- Fixed signal-route, GitHub-import, and planning-subtask-route task creation to stop forcing column when no workflowId is given (or, for planning subtask routes, even when one is provided).
- Custom workflows with a non-triage intake column (e.g. Inbox) now correctly capture new cards inert until released, while the default builtin:coding workflow still resolves to "triage" byte-identically.
- Added regression coverage (agent-tools-intake-column.test.ts, extension-workflow-tools.test.ts) and a patch changeset documenting the fix.
Files changed:
.changeset/fn-7611-intake-column.md | 7 ++
.../src/__tests__/extension-workflow-tools.test.ts | 70 +++++++++++
packages/cli/src/extension.ts | 10 +-
.../src/__tests__/register-signal-routes.test.ts | 8 +-
.../dashboard/src/__tests__/routes-github.test.ts | 2 -
.../dashboard/src/routes/register-git-github.ts | 16 ++-
.../src/routes/register-planning-subtask-routes.ts | 24 +++-
.../dashboard/src/routes/register-signal-routes.ts | 8 +-
.../__tests__/agent-tools-intake-column.test.ts | 138 +++++++++++++++++++++
packages/engine/src/__tests__/agent-tools.test.ts | 1 -
packages/engine/src/agent-tools.ts | 21 +++-
11 files changed, 288 insertions(+), 17 deletions(-)
Fusion-Task-Id: FN-7611
Fusion-Task-Lineage: daf7f755-b1c7-4859-b74f-f15593d5e79e
Co-authored-by: Fusion (runfusion.ai) <noreply@runfusion.ai>
Add TaskExecutor.blockOuterDispatchWhenEphemeralDisabled, gating all three
workflow dispatch paths (graph / authoritative / work-engine) on
ephemeralAgentsEnabled at the top of execute(). Previously the toggle was
enforced only on the legacy scheduler/EphemeralWorkerManager path — whose
onTaskStart spawn refusal is a fire-and-forget callback that runs after
execution begins — so unassigned tasks reaching execute() off a non-scheduler
path still ran. Unassigned tasks are now re-queued for permanent-agent
assignment; permanent-agent-bound tasks still run. Adds regression coverage
across all three entry points.
Also includes the ephemeralAgentsCanCreateTasks project setting (default on)
gating fn_task_create for ephemeral callers in both the pi extension and the
executor task-worker tool.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- classify image artifact support as a minor feature changeset
- reject non-image, empty, and signature-mismatched dataBase64 payloads
- trap keyboard focus in the task image lightbox
- assert media streaming with raw image bytes
Fusion-Task-Id: FN-7143
Co-authored-by: Fusion (runfusion.ai) <noreply@runfusion.ai>
- base-commit-capture.ts: shell-quote integration branch with a real
single-quoted POSIX literal instead of JSON.stringify (not shell-safe).
- TaskCard.tsx: memo compares full workspaceWorktrees values, not just key
sets, so a same-key worktreePath/branch change re-renders.
- TaskDetailModal.tsx: gate/render workspace summary off hydrated workingTask.
- worktree-acquisition.ts: null the singular worktree/branch columns in the
workspaceWorktrees write so isWorkspaceTask stays true; wrap non-fatal
post-acquire observability so logEntry/audit can't re-escalate to fatal.
- agent-tools.ts: register sub-repo worktree via onAcquired unconditionally
(idempotent) so a resumed/already-acquired path is tracked after restart.
- executor.ts: DB liveness fallback also checks task.workspaceWorktrees paths.
- executor-workspace.test.ts: root non-git assertion runs in fx.rootDir (".").
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Resolves the actionable CodeRabbit threads on the workspace-mode foundation:
- project-resolver: defer saveWorkspaceConfig until after the user confirms init
and store.init() succeeds (no partial .fusion/ on a declined/non-interactive run).
- git-repository: validate each candidate with a real `git rev-parse` work-tree
probe before counting it (no false-positive repos from stray .git markers);
loadWorkspaceConfig now rejects absolute paths, `..` escapes, and non-string
entries so a corrupt/malicious config can't resolve outside the workspace root.
- executor: gate workspace mode on repos.length > 0 at all three sites so an
empty { repos: [] } can't bypass the git-repo guard or enable an empty workspace.
- worktree-acquisition: thread the configured-command runner through the workspace
acquire path (sub-repos run their init setup); validate repoRelPath as an in-root
relative path before joining; liveness-check a remembered worktree before
reporting it ready (pruned paths fall through to re-acquire); clear the singular
task.worktree/branch after persisting per-repo state (per-repo state lives only
in workspaceWorktrees).
- agent-tools: forward runContext into acquireWorkspaceRepoWorktree for log attribution.
The executor-workspace test's mock-the-subject pattern is left for the
session-scoping follow-up that rewrites it with a real two-repo fixture (FN-5048).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
ce-code-review (5 personas) on Phase A. No P0; the workspace-root-removal path
was ruled out and the contract changes verified additive. Applied:
P1: fn_acquire_repo_worktree now catches WorkspaceRepoAcquireBusyError (and
generic failures) and returns a sanitized retryable isError instead of an
uncaught throw into the agent loop; runContext is forwarded so audit/log keep
attribution. Per-repo acquired worktree paths are now registered into the
executor's activeWorktrees Set (via an onAcquired callback) — previously the Set
only held the browse-only root, making the U1 per-repo liveness invariant hollow.
Post-acquire identity-guard install and base-SHA capture are now non-fatal
(log-and-continue): a hook/branch failure no longer strands the on-disk worktree
(the worktree is usable without the guard; an undefined baseCommitSha is already
an accepted state).
P2: the KTD3 settings-strip also strips settings.baseBranch (resolveFromSettings
falls back integrationBranch → baseBranch, so a shared baseBranch leaked); the
workspaceWorktrees write re-reads the task fresh before merging to avoid a
sibling-repo clobber on sequential acquires (store-level atomic merge deferred to
Phase B); the busy-path logging is wrapped so it can't mask the busy error; the
TaskCard memo compares key-sets not counts; the stuck-kill no-op for workspace
tasks is now logged; the exclusivity check-then-act synchrony is documented.
Residuals (Phase B): per-repo worktree teardown, orphan-scan coverage, reaper
dedup, store-level atomic merge.
Gate green: typecheck, lint, build, test:gate (649+58), affected (25 + TaskCard 251).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Allow registering a non-git parent directory that contains multiple git
repositories as a single Fusion project. The agent acquires per-repo
worktrees on demand via a new `fn_acquire_repo_worktree` tool as it
discovers it needs to work in each sub-repo.
This commit lays the foundation:
- detectWorkspaceRepos / loadWorkspaceConfig / saveWorkspaceConfig in
@fusion/core (config persisted to .fusion/workspace.json)
- Task.workspaceWorktrees data model + store plumbing (per-repo
worktree/branch map, distinct from the singular task.worktree)
- acquireWorkspaceRepoWorktree wraps acquireTaskWorktree per sub-repo,
clearing the singular worktree/branch fields so each sub-repo gets a
fresh worktree instead of resuming a sibling repo's worktree
- fn_acquire_repo_worktree agent tool + workspace prompt injection
- executor git-repository validation bypassed when a workspace config
is present
- CLI `fn init` detects a non-git dir containing sub-repos and writes a
workspace config
Known gap (intentionally left for design discussion, see PR): the
executor's main worktree-acquisition path still assumes a single git
root and is not yet workspace-aware. End-to-end execution (skipping the
root acquisition, per-repo merge, per-repo session scoping) is a
follow-on once the execution model is agreed.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Ensure agent-created untitled tasks still request AI-generated titles even when project auto-summarization is disabled.
- force summarize=true for agent-created tasks that do not provide an explicit title
- add engine tests covering untitled, titled, and title-generation behavior with autoSummarizeTitles disabled
- document the agent-task summarization behavior in settings docs
- add a patch changeset for the published CLI package
Files changed:
.changeset/agent-task-title-summaries.md | 5 ++
docs/settings-reference.md | 2 +-
packages/engine/src/__tests__/agent-tools.test.ts | 63 +++++++++++++++++++++++
packages/engine/src/agent-tools.ts | 1 +
4 files changed, 70 insertions(+), 1 deletion(-)
Fusion-Task-Id: FN-6107
Fusion-Task-Lineage: d0c25981-a5cb-48ca-b835-9267817b4edd
