Follow-up to 13a2b2a9d, from a multi-agent review of that commit. Three of its
claims did not hold.
1. fn_delegate_task bypassed the gate entirely (P0). It reaches the same
createAgentTask primitive, was registered unconditionally in both session
lanes, and validated only that the TARGET agent is non-ephemeral — never the
caller. Under Deny an ephemeral worker could enumerate agents and delegate
unlimited tasks. It is now withheld under Deny, and also under
upon_validation: delegation has no proposal channel, so leaving it available
would launder a create past the operator review that policy requires.
2. The widened dedupe window was capped at 5 minutes. The store query in
branch-and-pr-entities.ts carried its own independent `?? 60_000` /
`min(300_000, …)` pair, so widening only duplicate-guard.ts under-delivered
and made the new ceiling unreachable. Both sites now share
FINGERPRINT_WINDOW_DEFAULT_MS / FINGERPRINT_WINDOW_MAX_MS.
3. The pi-extension gate does not fire at all. pi's ExtensionContext carries no
agentId — the read is a speculative cast and only tests supply one, so every
real call short-circuits as a human caller. The fail-closed direction is kept
for the day an identity signal exists, but the limitation is now documented
instead of implied to be enforcement.
Also: the session prompt now states when creation is disabled and names
fn_task_log as the fallback (the base prompt still taught fn_task_create, which
is the same instruction/capability mismatch that fed the retry storm);
suppression emits an `agent:task-create-withheld` run-audit event; and the two
source-text ratchet tests are replaced with behavioral assertions on the tool
list the executor actually hands the model — verified to fail when the guard is
broken, which the string assertions did not.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Operator report: with project policy "Ephemeral agent follow-up tasks = Deny",
an executing agent filed ten follow-up tasks — five parallel fn_task_create
calls it reported as timed out, then five sequential retries.
Two defects:
1. Deny was advisory. fn_task_create was registered for every session and only
refused inside execute(), so the model still saw the tool, planned around it,
and retried it. The pi extension's isEphemeralCallerAgent also failed OPEN
whenever the caller id did not resolve to an agent row — which is the normal
shape of an ephemeral task-worker — so on that lane Deny was a no-op.
2. The deterministic content-fingerprint duplicate window was 60s, which only
covered concurrent in-flight creates. A retry two minutes later saw nothing
and filed a second task.
Fixes: isAgentTaskCreateToolAvailable() withholds the tool from ephemeral
sessions under Deny in both engine lanes (outer execution session, per-step
workflow session); isEphemeralCallerAgent fails closed on an unresolvable
caller id; the fingerprint window goes 60s -> 10m (clamp ceiling 5m -> 1h).
upon_validation keeps the tool, and permanent-agent and human/chat callers are
unaffected.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Add the `autoUpdateAndRestart` global setting (default off, Settings ->
General next to Release channel). When enabled, the dashboard host installs
available updates on the selected channel by itself and requests the
supervised in-place restart. Supervised hosts only: without a parent to
respawn, installing would leave a running process whose code no longer
matches its own install.
Fix two ways the restart affordance could silently do nothing:
- The supervisor now stamps FUSION_SUPERVISOR_PID and supervision is only
counted when that pid is the real parent. FUSION_RESTART_SUPERVISED is
inherited by every process Fusion spawns, so `fn dashboard` launched from
an agent terminal skipped its own supervisor while still advertising
restart support -- a restart request then killed it for good.
- Settings and the update banner probe /system/info on mount and treat
capability as advisory: the button always issues the request and shows the
server's actual refusal instead of sitting disabled after a failed probe.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
CLI JSON/create success lines write via result() (raw stdout) so quiet mode
cannot drop machine-readable output; capture that seam in research/update/task
tests instead of console.log. Allowlist nested voiceInput settings for the
FN-7505 default-description guard and ship locale keys for Voice Input UI.
A paused engine with nothing running derived executorState "idle", so the
footer badge was indistinguishable from a healthy engine waiting for work.
That is exactly the state a pause settles into once in-flight tasks drain:
triage and planning stall while the board keeps moving, with the pause only
visible by opening the Engine Control menu. Pause state now dominates run
state; the adjacent counters still report throughput.
In the CLI TUI, the global `t` (Git view) branch returned before the Utilities
dispatch in the same key handler, so the advertised "[t] Toggle Engine Pause"
was unreachable dead UI. It now yields when Utilities owns input, and because
the shortcut can stop the board, pausing takes a second `t` within 5s while
resuming stays single-press.
Tests assert the invariant across the whole state matrix (both pause flags x
0/1/5 running), not just the zero-running repro, plus the TUI routing, the
two-press pause, single-press resume, and re-arm behavior.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Full Suite shard 4 timed out runTaskShow lock-exhaustion at the default 5s
budget (run 30096660913): each case re-imported the heavy task.js graph under
fake timers, so cold CI workers spent the whole budget on transform and left
unhandled store.getTask / process.exit races after the timeout.
Import task.js once per cached-store describe under real timers via a mutable
store holder, and enable fake timers only around the backoff body. Exhaustion
cases now finish in ~1ms locally while keeping the shipped entry path.
Stale-test drift, no product changes: FN-8424 inbox reply help text and
fn message inbox --user routing, beta-track prerelease suffixes in version
regexes (0.73.0-beta.N), FN-8399 onMigrationProgress in createTaskStoreForBackend,
#2400 workflow-docs heading, and the durable planning-session store mocks for
fn task plan (0412113de/fdd120232).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Reported bug (screenshot): deleting the task created from a plan left the
session permanently stuck on PLANNING_CREATED_TASK_MISSING — Retry create
replayed the same 409 forever. A linked task absent from the
include-archived scan (task-row authority; a successful scan proves
deletion, not a flaky read) now clears the stale linkage and creates a
fresh task, in both the create-task route and createTaskFromPlanSession;
a still-listed-but-unreadable task keeps failing closed.
Multi-agent review of fdd120232 (correctness/adversarial/reliability):
- P1: CLI planning sessions were memory-only — setAiSessionStore only ran
in the dashboard server, so --resume could never find a session across
invocations. New ensureDurablePlanningSessionStore wires the durable
AiSessionStore over the board store's public asyncLayer in runTaskPlan.
- P1: resume failures now THROW instead of process.exit (fn_task_plan
runs inside the pi host — an exit killed the whole agent session), and
a no-question resume requires an explicit refine focus (the provided
description) so merely resuming never rotates the epoch.
- P2: claim and finalize CAS gained the same expected-epoch WHERE guard
as reconcile, so a stale-epoch creator can no longer finalize an
old-epoch task onto a rotated session.
- Side-effect failures (documents, logEntry, validate, reconcile) are now
logged instead of swallowed; post-insert failures no longer mislabel
the just-created task alreadyCreated:true; the keep-refining readline
closes on thrown prompts and a failed refine after creation returns the
created task id with a resume hint; cross-process generating guard
added to createTaskFromPlanSession.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Closes the P1 agent-native gap from the multi-task review: the CLI and
fn_task_plan pi tool created tasks via a raw store.createTask with no
proposalClaimId — no idempotency, no session linkage, and tasks outside
the epoch sequence, so a later dashboard Proceed would duplicate them.
- New shared createTaskFromPlanSession in @fusion/dashboard/planning:
the agent-surface twin of POST /planning/create-task (epoch-derived
claim key, claim/finalize/reconcile/release CAS lifecycle with the 30s
stale-lease takeover, formatPlanningPlanMd task shape, plan/original-
description documents, validate-on-create, generating guard).
- runTaskPlan creates through it (making the FN-7734 retry wrapper
genuinely safe), prints the session id, and offers an interactive
keep-refining loop that creates further tasks from the evolved plan.
- fn task plan --resume <sessionId> / fn_task_plan resumeSessionId reopen
an existing session — even a validated one whose task exists — and the
no-question resume regenerates the interview via a refine turn, which
rotates the creation epoch server-side.
Tests: CLI suite pins claim-aware creation, the continue prompt, and the
resume flow; dashboard suite pins createTaskFromPlanSession idempotent
replay and epoch-aware second creation.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
## Summary
Every PR's blocking checks were dominated by redundant full rebuilds,
not by tests. Measured on recent runs: the Gate job spent ~6 of its ~7.5
min on a cold `pnpm build` (the exact-key dist cache missed on virtually
every PR) for ~45s of actual boot smoke + gate tests; Build ran ~8 min
and Typecheck ~4 min, both fully cold every time. Expected end state
once the warm job has run on main: all four blocking checks in roughly
2–4 min wall-clock.
### Gate job
- New `gate-dist-*` cache namespace with `restore-keys`, additionally
caching `.fusion/cache/plugin-build-cache.json` (build-workspace's
per-package content-hash skip cache) and `packages/cli/dist`. The
always-run `pnpm build` reconciles a near-match restore by content hash
and rebuilds only the packages the PR touched. This is safe *because*
the gate builds after restoring — the shard jobs' "no restore-keys" rule
(FN-4232/FN-4605 stale-dist incidents) still stands there, since they
consume dist without building.
- `FUSION_CLI_FULL_PACKAGE=0` on the gate build: skips the multi-minute
CLI desktop/plugins/DTS packaging tail nothing in the gate consumes
(same shape `pnpm verify:fast` proves locally). Full CLI packaging
coverage stays blocking in the Build job.
### Build job
- Restore-only tap (`actions/cache/restore`) of the same warmed cache.
Restore-only because this job runs FULL CLI packaging (`CI=true`) and
saving that shape would swap the cache's canonical fast-CLI contents out
from under the Gate job. Its distinctive coverage is preserved:
`ensureFullPackageCliPlanned` force-plans the CLI in full mode
regardless of cache state.
### Typecheck job
- Caches per-package tsc incremental buildinfo — self-validating (tsc
hashes every input against it and re-checks whatever changed), so
`restore-keys` is correctness-neutral by construction.
- **Fixes a real incrementality bug:** `tsconfig.json` and
`tsconfig.app.json` in the dashboard both inherited
`${configDir}/dist/.tsbuildinfo` from `tsconfig.base.json`, so the two
typecheck programs clobbered each other's buildinfo and re-checked the
full program every run — incremental typechecking never worked for the
dashboard, in CI or locally. `tsconfig.app.json` now writes
`dist/.tsbuildinfo-app`. Measured: dashboard typecheck 44s cold → 5.6s
warm.
### Warm job (full-suite.yml, push to main)
- New `warm-gate-build-cache` job saves both caches from main on every
push. Caches saved on a PR merge ref are invisible to other PRs, so
without this every PR's *first* run would still build/check cold.
## Guardrails
`ci-workflow.test.ts` pins the coupled invariants so they can't drift
apart silently:
- restore-keys requires the reconciling `pnpm build` after restore,
before boot smoke
- the mtime-defeating seed step stays exact-hit-only
- byte-identical cache path lists between the Gate/Build/warm blocks
(actions/cache versions caches by path list — a drifted list makes
caches mutually invisible)
- Build stays restore-only and must NOT opt out of full CLI packaging
- Typecheck cache shape + the distinct dashboard app buildinfo path
## Notes
- First PR runs after this lands still build cold until the warm job has
run once on main.
- No changeset: CI config + test-only per AGENTS.md.
## Verification
- `ci-workflow.test.ts` + `package-config.test.ts`: 106 tests pass
- Dashboard typecheck run twice locally: 44s cold → 5.6s warm, both
`.tsbuildinfo` and `.tsbuildinfo-app` written, exit 0
- Cache block path/key parity verified programmatically across both
workflow files
🤖 Generated with [Claude Code](https://claude.com/claude-code)
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Performance**
* Improved CI build and type-check performance through incremental
caching.
* Added cache warming from the main branch to speed up pull request
checks.
* Enabled faster CLI packaging during gate validation while retaining
full packaging coverage elsewhere.
* **Bug Fixes**
* Prevented dashboard TypeScript build information from being
overwritten, preserving incremental type-checking reliability.
* **Tests**
* Added coverage to verify CI cache behavior, build ordering, cache
paths, and packaging modes.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
## Summary
Fusion npm installs now expose a working `agent-browser` command on
Windows, Linux, and macOS. Fusion publishes a top-level shim backed by
the exact pinned native package, preserving the existing Browser
Verification probe contract on every supported desktop OS.
## Validation
- Added a packed consumer-install matrix for `ubuntu-latest`,
`macos-latest`, and `windows-latest`.
- Each platform executes npm's generated shim, verifies `agent-browser
0.26.0`, and asserts the matching native binary is installed.
- Confirmed the packed Fusion manifest, installed dependency, and
command output all retain the exact declared version pin.
- Passed 101 focused CLI workflow/package tests, full workspace lint and
typecheck, strict changeset validation, and a real macOS ARM64
packed-install smoke.
## Post-Deploy Monitoring & Validation
- Search task logs for `agent-browser not found on PATH`, `Missing
native executable`, and Browser Verification availability warnings.
- Healthy signal: npm installs on Windows, Linux, and macOS resolve
`agent-browser --version` without missing-shim or native-payload errors.
- Failure signal: command resolution errors, version-pin mismatches,
missing native payloads, or increased browser-verification fast-bails.
- Validation window: first release cycle after publish; owner: Fusion
maintainers.
- Mitigation: revert the dependency and top-level shim if install
compatibility regresses.
---
[](https://github.com/EveryInc/compound-engineering-plugin)
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Added cross-platform `agent-browser` installation through Fusion CLI.
* Exposed an `agent-browser` command for Windows, macOS, and Linux.
* Pinned the included `agent-browser` version to ensure consistent
installations.
* **Bug Fixes**
* Improved reliability of platform-specific executable selection and
command setup across supported operating systems.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
Post-extension modelRegistry.refresh() had no timeout, so a hung remote
catalog fetch left the TUI on "Loading extensions…" forever. Use a shared
15s-bounded refresh across dashboard/serve/daemon and related registration paths.
## Summary
Fusion can now ship on two release tracks. Betas are cut from `main` as
`vX.Y.Z-beta.N` (npm dist-tag `beta`, GitHub prerelease), stable
releases are promoted to a long-lived `release` branch and published to
`latest`, and users pick their track with the new `updateChannel` global
setting — via **Settings → General → Release channel** or `fn update
--channel <stable|beta>`. Previously everything was single-track: every
publish landed on `latest` and every update surface could only see it.
| | beta | stable |
|---|---|---|
| Cut from | `main` | `release` branch |
| Version | `X.Y.Z-beta.N` (changesets pre-mode) | `X.Y.Z` |
| npm dist-tag | `beta` | `latest` |
| GitHub Release | prerelease | latest |
| Homebrew tap / X draft | skipped | bumped / printed |
## How releasing works now
`pnpm release` prompts for the channel and **defaults to beta**, so
day-to-day releases are betas; stable is always an explicit choice.
Choosing stable from `main` triggers assisted promotion: the script
proposes the newest beta tag reachable from HEAD, verifies `release`
fast-forwards to it, then runs the whole stable release inside a
temporary git worktree on `release` — the primary checkout never leaves
`main`. Changesets pre-mode preserves changeset files across betas, so
the promoted stable release aggregates every changeset since the last
stable into one clean changelog entry.
## Design decisions
- **Every publish path names an explicit `--tag`.** A beta accidentally
landing on `latest` is the one unrecoverable failure of a dual-track
scheme, so nothing relies on npm's implicit default (`release.mjs`,
`version.yml`).
- **Beta channel resolves to semver-max of `latest` and `beta`**, so
beta users are offered each promoted stable once it overtakes their
prerelease. Switching beta → stable never downgrades; `fn update
--channel stable --force` is the explicit escape hatch.
- **One comparator instead of three.** CLI, dashboard, and desktop each
had their own `isRemoteNewer` that ignored prerelease identifiers —
`0.73.0-beta.2`, `-beta.3`, and `0.73.0` all compared equal, which
breaks the moment any beta exists. They now share full SemVer-precedence
helpers (`compareVersions`, `resolveUpdateTargetVersion`) from
`@fusion/core`.
- **Installs pin exact versions** (`@runfusion/fusion@0.73.0-beta.2`),
never a dist-tag, so an install can't silently land on the wrong track.
- **Desktop channels via electron-updater manifests.** Beta tags build
desktop artifacts with `publish.channel=beta` (emitting `beta*.yml`);
the app sets `channel`/`allowPrerelease` from the shared setting,
re-read on every manual check.
- **Update caches are channel-stamped** — a cache written for one
channel is never served to the other, so switching tracks takes effect
on the next check instead of after TTL.
## Test plan
- New unit coverage: SemVer precedence + channel resolution in
`@fusion/core` (30), channel behavior of the dashboard update check (28,
incl. 9 new) and `fn update` (16, incl. 8 new: persist `--channel`,
no-downgrade, `--force`, cache channel mismatch).
- `pnpm verify:fast` green (scoped typecheck, builds, CLI build, boot
smoke); desktop + settings-section suites green.
- `release.mjs` dry-run matrix exercised by hand: channel prompt
(default/override/invalid), branch preflights per channel,
assisted-promotion target selection, fast-forward guard against a
diverged `release` branch, and bootstrap when no `release` branch
exists.
- Not exercised live: an end-to-end publish (needs TTY authorization +
real npm publish). First real run is the first `pnpm release --channel
beta`.
---
[](https://github.com/EveryInc/compound-engineering-plugin)
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Added beta and stable release channels across CLI, dashboard, and
desktop updates.
* Users can select a channel via Settings or `fn update --channel
<stable|beta>` (stored as a global default).
* Desktop beta releases now generate beta update manifests and publish
as prereleases.
* **Documentation**
* Expanded release-track, settings, and CLI references to explain
channel semantics and workflows.
* **Bug Fixes**
* Updates now pin the resolved version per channel, improve version
comparison, and prevent unintended cross-channel downgrades unless
`--force` is used.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
## Summary
Windows installs with slow native dependencies now get five minutes to
finish, and a real timeout is reported as an actionable terminal retry
instead of a wall of preceding npm deprecation warnings. Registry
`ETIMEDOUT` errors keep their network diagnosis, including after the
legacy-bin `--force` retry.
Compound Engineering personas are now included in the published CLI
bundle, with complete source-to-staged coverage for all persona
definitions and a clear startup error if the bundled assets are missing
or empty.
The PostgreSQL statement visible in the report was validated by the
existing real-Postgres schema reapply test. Its actual `caused by`
detail was truncated, so this PR deliberately makes no speculative
database change.
## Validation
- Dashboard updater tests: 22 passed
- CLI updater tests: 16 passed
- CE persona installer tests: 7 passed
- Published bundle persona assertion: passed against every source
persona
- CLI and CE plugin typechecks: passed
- Changed production/config lint and strict changeset validation: passed
- Real PostgreSQL schema reapply integration test: passed
---
[](https://github.com/EveryInc/compound-engineering-plugin)
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Bug Fixes**
* Windows CLI and dashboard updates now allow up to five minutes for
installation and restore Compound Engineering agent personas during npm
installs.
* Update failures now surface clearer, terminal timeout guidance (while
preserving specific network connection diagnostics) and avoid misleading
“deprecated”/generic timeout text.
* Persona assets are reliably included in plugin builds and bunded
persona installation now errors clearly when definitions are missing or
empty.
* **Tests**
* Expanded update and bundling coverage for the new 5-minute timeout and
error-handling scenarios.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
## Summary
Repeated review and executor sessions could replay a follow-up creation
step and produce another live task whenever the wording changed. In the
incident behind this fix, 21 creation calls for three intended
follow-ups left 18 duplicate tasks.
Agent-created tasks now retain their parent and agent provenance across
step sessions, heartbeats, and the published CLI surface. Same-parent
paraphrases converge on the existing task through a serialized pre-check
and a database-backed intent claim, while distinct sibling actions
remain separate. Candidate lookup is parent-indexed, uniqueness failures
abort creation, and canonical reuse no longer emits misleading creation
audit events or workflow claims.
Related: FN-8277
## Validation
- Core duplicate guard and intake: 24 tests passed
- Engine task creation and heartbeat: 154 tests passed
- CLI extension: 66 tests passed, 95 skipped
- Core, engine, and CLI typechecks passed
- Core, engine, and CLI builds passed before rebase; the rebase was
conflict-free
- Scoped lint, strict changeset validation, and diff checks passed
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Bug Fixes**
* Prevented retried agent steps from creating duplicate follow-up tasks.
* Improved parent-scoped deduplication for paraphrased follow-ups while
preserving distinct sibling actions.
* Preserved parent-task and agent context for created follow-ups.
* Concurrent follow-up requests are now serialized/deduplicated so
duplicates link to the existing task instead of showing as newly
created.
* Updated agent follow-up/heartbeat activity so reused follow-ups no
longer appear in run results as fresh creations.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
The one-time SQLite→PostgreSQL migration runs inside createTaskStoreForBackend
before any HTTP server listens, so browsers saw "connection refused" and open
tabs failed silently for minutes. Now:
- CLI: a temporary holding server binds the dashboard port for the boot window,
serving an auto-reloading "Database migration in progress" page and an
/api/health payload with status "migrating" + structured progress; the port
is handed off (awaited) to the real app.listen().
- Dashboard SPA: already-open tabs render the new MigrationInProgressBanner
from the 15s health poll when status is "migrating".
- Desktop: LocalRuntimeManager publishes migration progress on
DesktopRuntimeStatus via the new core onMigrationProgress option;
DesktopLaunchGate shows the live label and extends its 30s startup timeout
while progress advances (2min stall cap), in both boot and first-run flows.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
## Summary
In a centrally-installed, multi-project Fusion server (one process
serving several repos, `process.cwd()` = the install dir, not any repo),
every task under `mergeStrategy: "pull-request"` fails at the auto-merge
stage with:
```
Could not determine repository. Specify owner/repo in params or run from a git repository with a GitHub remote.
```
PR creation from the dashboard and status polling work; only the
engine's automatic PR path fails. This is the **non-workspace sibling of
#1924** (FN-7610 routed workspace-mode tasks to direct merge but does
not cover regular multi-project tasks) and the completion of
#1797/FN-7133 (which fixed only the `getPrMergeStatus` arguments).
## Root cause
`GitHubClient.resolveRepo()` (`packages/dashboard/src/github.ts`) falls
back to a cwd-less `getCurrentRepo()` — i.e. `git remote get-url origin`
in `process.cwd()` — whenever a PR method is called without explicit
`owner`/`repo`. The engine merge path already resolves the correct repo
from the per-project cwd (`prRepo = getCurrentRepo(cwd)`, FN-7133) but
only threaded it into `getPrMergeStatus`. Every other GitHub call
omitted it:
- `processPullRequestMergeTask`: `findPrForBranch` / `createPr` /
`mergePr` on both the per-task and shared-branch-group paths
- `createGroupPrCallback` (group-PR promotion): `findPrForBranch` /
`createPr`
- `createPrNodeGithubOps` (`pr-create`/`pr-merge` workflow nodes):
cwd-less `getCurrentRepo()` persisted `entity.repo` as `""` (poisoning
the downstream `splitRepoSlug` consumers), and the git
push/`createPr`/`mergePr` ran against `process.cwd()`
- the engine's review-response run (`buildRespondCallback`):
`respondOps.getCwd` collapses to `process.cwd()` because no CLI
composition site wires `getTaskWorktree`, so its git ops and response
agent ran outside the project repo
In a central install the fallback throws; worse, if `process.cwd()`
happens to be inside some *other* git repo, it silently targets the
**wrong repository**.
## What changed
- `fix(pr-merge): thread repo identity into PR auto-merge GitHub calls`
— widens the CLI-local `GitHubOperations` interface (optional
`owner`/`repo`, already accepted by `GitHubClient`'s
`FindPrParams`/`CreatePrParams`/`MergePrParams`) and passes `prRepo` at
all six call sites in `processPullRequestMergeTask`.
- `fix(pr-merge): resolve group-PR repo from project cwd in
createGroupPrCallback` — resolves via `getCurrentRepo(cwd)` from the
callback input (same T4 pattern as `syncGroupPrCallback`) with a loud
failure instead of a silent wrong-repo fallback.
- `fix(pr-merge): resolve PR-node repo from task worktree instead of
process cwd` — `resolvePrSource` resolves from `task.worktree`, git ops
run in `getTaskWorktree(...) ?? task.worktree ?? process.cwd()`, and
`createPr`/`mergePr` pass `owner`/`repo` parsed from `entity.repo`.
- `fix(pr-merge): resolve review-response run cwd from the task
worktree` — the engine owns the store, so `buildRespondCallback` prefers
the task's recorded `worktree` for the response run's git ops + agent,
keeping `ops.getCwd` as the single-project fallback (defensive against
structural `PrNodeStore`s without `getTask`).
- Changeset (`@runfusion/fusion` patch, structured body) included.
Deliberately **not** done: a constructor-scoped default repo on
`GitHubClient` — one client instance is shared across all projects in a
central install (`serve.ts`/`daemon.ts`/`dashboard.ts`), so per-call
`owner`/`repo` is the only correct scope.
## Testing
- New regression tests simulate the central-install topology
(`getCurrentRepo` mocked as `(cwd?) => cwd ? repo : null`, exactly the
failing environment) and drive the merge flow end-to-end on the per-task
path, the shared-branch-group path, `createGroupPrCallback`, and all
three `createPrNodeGithubOps` ops, asserting every GitHub call carries
explicit `owner`/`repo` (45 tests in
`packages/cli/src/commands/__tests__/task-lifecycle.test.ts`, all
green).
- `packages/engine/src/__tests__/pr-respond-cwd-resolution.test.ts`
covers the respond-run cwd: worktree preferred, `ops.getCwd` fallback
when the task has no worktree, when the lookup fails, and when a
structural store has no `getTask`.
- Existing exact-argument assertions were extended to the new call
contract (no assertions weakened or removed).
- `pnpm lint`, `pnpm typecheck`, and `pnpm build` green locally; `pnpm
test:gate`'s engine-core suite green (294/294) — its PostgreSQL-backend
lane needs local PG credentials this environment lacks, so that lane
defers to CI. `pnpm verify:fast` (scoped typecheck/build + CLI build +
boot smoke) also passes.
## Repro
1. Install the CLI centrally; run the server from a dir that is not a
git repo, serving ≥1 project with a GitHub `origin` and `mergeStrategy:
"pull-request"`.
2. Run a task to completion and let it reach the merge stage.
3. Before this fix: the auto-merger throws `Could not determine
repository …` (tasks with a persisted PR poll fine but never merge).
Merging the same task from the Pull Requests tab succeeds, because the
dashboard route resolves the repo explicitly (`parseBadgeUrl(...) ??
getCurrentRepo(rootDir)`).
Full analysis: https://github.com/Tchori-Labs/Fusion/issues/4
---
Developed with Claude (co-authored on all commits).
https://claude.ai/code/session_01ChEa8SHFYNAzjCdFbwFMfh
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
- **Bug Fixes**
- Resolved pull request auto-merge failures in centrally installed,
multi-project deployments.
- Ensured explicit repository context (`owner/repo`) is used for pull
request lookup, creation, and merging throughout the merge workflow.
- Improved pull request response handling to prefer the task worktree
for working-directory resolution, with safe error behavior when task
details are unavailable.
- **Tests**
- Expanded coverage for multi-repository merge workflows and
worktree-based repository/cwd resolution in PR response handling.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: Claude <noreply@anthropic.com>
Repair CLI test mocks and retry-reset expectations for the current store contracts.
- Add the backend store factory to the experiment-finalize mock.
- Provide global settings directory access in backup test stores.
- Assert all manual retry reset fields in task command tests.
Files changed:
.../extension-experiment-finalize.test.ts | 13 ++++++++++
.../commands/__tests__/backup-lock-retry.test.ts | 2 ++
packages/cli/src/commands/__tests__/backup.test.ts | 17 ++++++++++++-
packages/cli/src/commands/__tests__/task.test.ts | 28 +++++++++++++++++-----
4 files changed, 53 insertions(+), 7 deletions(-)
Fusion-Task-Id: FN-8222
Fusion-Task-Lineage: 033ce6a6-c699-409c-a59e-2d1f5e041cfc
Co-authored-by: Fusion (runfusion.ai) <noreply@runfusion.ai>
Align daemon startup mocks with the model runtime initialization path.
- Add the auth-storage setModelRuntime mock.
- Stub the Fusion model registry factory to use the shared test registry.
Files changed:
packages/cli/src/commands/__tests__/daemon.test.ts | 9 +++++++++
1 file changed, 9 insertions(+)
Fusion-Task-Id: FN-8220
Fusion-Task-Lineage: 6e00980f-2e24-4502-a4d8-e91849df33b9
Co-authored-by: Fusion (runfusion.ai) <noreply@runfusion.ai>
## Summary
An audit of the SQLite→PostgreSQL store migration found data-store paths
still reaching the removed SQLite stub in backend (PG) mode. In backend
mode `store.db`/`getDatabase()` throw the removed-SQLite error, so each
of these either threw on every run or — worse — had the throw swallowed
into a silent wrong result. This PR routes all of them through the
`AsyncDataLayer` (and removes one dead primitive).
## The 6 live bugs fixed
| Fix | Was |
|-----|-----|
| `executor.ts` authoritative assigned-agent fallback now inherits the
TaskStore `asyncLayer` | silently returned `null` → model drift to the
pi built-in (the exact thing its comment guards) |
| `pruneAgentLogFilesAsync` replaces the sync self-healing prune call |
threw `SQLite Database is not available` every maintenance sweep →
agent-log pruning never ran |
| `cleanupOrphanedMaterializedSteps` deletes PG `workflow_steps` rows on
a failed create | swallowed the throw → leaked rows |
| `deleteTaskBackendImpl` now runs the async mission feature/task-link
unlink | PG hard delete left orphaned mission links |
| `getWorkflowSettingsProjectId` returns `rootDir` in backend mode
without touching the stub | swallowed throw for unscoped backend stores
|
| `fn plugin` unregistered-project fallback bootstraps a `CentralCore`
`AsyncDataLayer` | layerless `PluginStore` threw in PG |
## The 4 latent traps, fixed properly
- **`cleanupArchivedTasks`** — real async port (enumerate archived
soft-deleted rows, guarantee cold snapshot, hard-delete project row +
purge selection rows + rm dir).
- **`deleteWorkflowStep`** — real async port (delete `workflow_steps`
via the layer with `.returning()` to preserve the not-found contract).
- **`applyTaskPatch`** — **removed** (zero-caller SQLite column-patch
primitive with no backend analogue; impl + facade + import deleted).
- **`AgentStore.importLegacyFileRuns`** — clean backend no-op (no legacy
SQLite run-files exist in a PG deployment; its only `init()` caller
early-returns in backend mode).
## Symptom Verification
New PG regression suite
`packages/core/src/__tests__/postgres/store-sqlite-residue-fixes.pg.test.ts`
reproduces the original failures against real embedded Postgres and
asserts they're gone:
- orphaned `workflow_steps` are actually deleted (no swallowed throw)
- `pruneAgentLogFilesAsync` resolves and prunes inactive-task log files
- hard delete unlinks the mission feature from the task
- `deleteWorkflowStep` removes the row / reports not-found
- `cleanupArchivedTasks` hard-deletes the project row while retaining
the cold snapshot
## Verification
- `@fusion/core`, `@fusion/engine`, `@runfusion/fusion` typecheck clean
- ~50 existing + 5 new PG tests pass; lint clean; changeset validates
🤖 Generated with [Claude Code](https://claude.com/claude-code)
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Bug Fixes**
* Prevented PostgreSQL backend maintenance from hitting removed legacy
SQLite code paths, avoiding datastore failures and residue cleanup
issues.
* Fixed workflow-step deletion and “not found” behavior in backend mode.
* Ensured backend hard-deletes correctly unlink related mission
feature/task links and clean orphaned materialized steps.
* Prevented legacy file-run imports from incorrectly reporting success
in backend mode.
* **New Features**
* Added async agent-log pruning for inactive tasks and updated
maintenance to use it.
* **Tests**
* Added PostgreSQL regression coverage for residue fixes and
archive/workflow cleanup.
* **Refactor**
* Removed an unused task patch operation and updated task-store cleanup
methods to be async where needed.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Restore the CLI package configuration test to the active test lane with current build expectations.
- Allowlist WhatsApp plugin-only tsup externals as non-runtime CLI dependencies.
- Assert the full workspace build command in the verification contract.
- Remove the stale package-config test quarantine.
Files changed:
packages/cli/src/__tests__/package-config.test.ts | 12 +++++++++++-
packages/cli/vitest.config.ts | 9 ++++-----
2 files changed, 15 insertions(+), 6 deletions(-)
Fusion-Task-Id: FN-8210
Fusion-Task-Lineage: aa29d866-430f-4097-affa-a89d107474b2
Co-authored-by: Fusion (runfusion.ai) <noreply@runfusion.ai>
