Three orthogonal fixes bundled together so they re-land as a unit after
earlier worktree-based reverts kept wiping them individually.
1. `--no-auth` flag now actually disables auth. Previously a stale
FUSION_DAEMON_TOKEN in .env silently re-armed bearer-token auth despite
the CLI flag. Added a `noAuth` option to ServerOptions; auth-middleware's
isDaemonAuthActive/getDaemonToken short-circuit to false/undefined when
set; CLI plumbs opts.noAuth through both createServer call sites.
2. Workflow review failures no longer reset every completed step. Previously
a single CSS nit from a workflow reviewer could drag 5+ already-approved
steps back through plan review, code review, and re-execution because
determineRevisionResetStart fuzzy-matched feedback tokens against step
names. handleWorkflowRevisionRequest, handleWorkflowStepFailure, and
sendTaskBackForFix now call a new reopenLastStepForRevision helper that
flips only the last non-pending step back to pending (with currentStep
rewind via a newly-accepted updateTask field) — all earlier done steps
stay done, and the agent applies the feedback as an in-place patch per
the updated PROMPT.md instructions. determineRevisionResetStart stays
exported as @deprecated so existing unit tests still link.
3. Heartbeat scheduling is now state-driven. Previously a non-ephemeral
agent with a stale runtimeConfig.enabled=false on disk would never tick
and the Pause/Resume button couldn't arm the timer without also flipping
that hidden flag. HeartbeatTriggerScheduler's watchAgentLifecycle now
registers on transitions into active/running and clears on transitions
out; the tick and assignment-trigger guards key off state + ephemeral
classification. InProcessRuntime's created/updated listeners and startup
scan mirror the same semantics. runtimeConfig.enabled is only retained
for ephemeral (task-worker) opt-out.
Tests updated: agent-heartbeat.test.ts — one test renamed from "skips
registration when enabled is false" (obsolete behavior) to
"registers regardless of the legacy enabled flag"; 4 assignment-watching
tests now pass a realistic `state: "active"` on mock agents. 207 heartbeat
tests + 330 executor tests pass.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Gate executor pre-merge workflow-step execution on executionMode and bypass it for fast runs
- Omit the review_step tool from agent tool injection when tasks run in fast mode
- Add explicit executor logs and task log entries when fast mode skips validation gates
- Extend executor messaging-tool tests to cover fast vs standard tool availability
Atomic worktree creation: on `git worktree add` failure, remove the
partial directory so retries see a clean slate (executor.ts,
step-session-executor.ts). Add `reapOrphanWorktrees()` sweep on engine
startup to delete `.worktrees/*` dirs that have no `.git` file and are
not registered with git. Force readonly workflow steps to run pre-merge
so review personas reuse the coding worktree instead of allocating a
fresh post-merge one. Inject a Frontend UX Criteria checklist into
TRIAGE_SYSTEM_PROMPT when a task touches dashboard UI, so the first
coding pass can meet UX reviewer expectations.
Motivated by FN-2185 post-mortem: an incomplete `.worktrees/pale-raven`
dir blocked retries, and three Step 6 revision cycles over cosmetic UX
issues consumed ~13 hours.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
When the engine restarts mid-step, an in-progress step may have already
passed plan + code review but not yet been flipped to done by the agent's
next task_update call. Previously, the next executor pass re-entered the
step and replayed both reviews — measured at 5-20 min of pure waste per
restart (observed in FN-2215 Step 1 and FN-2207 Step 6).
recoverApprovedStepsOnResume scans the task log for any in-progress step
whose most recent "code review Step N: APPROVE" entry is newer than its
most recent "Step N → pending" transition, and marks those steps done
before execute() runs. Safely skips steps that were reset after approval
(e.g. by a workflow revision) or only received REVISE verdicts.
Called from both the engine-restart path (resumeOrphaned) and the
unpause path, matching the two places the task log shows as vulnerable
to this race.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Move the assigned agent badge out of the card header into a dedicated metadata row below task details
- Add a new .card-agent-row container to control spacing and alignment for the badge block
- Update .card-agent-badge styling to use token-aligned pill radius and color-mix backgrounds while removing monospace/fixed-width conventions
- Expand TaskCard agent badge tests to verify new DOM placement and enforced badge style rules
Resume paths (unpause, drift recovery, engine restart) bypassed the
scheduler's todo->in-progress clear, leaving actively executing tasks
labeled status="queued" with a lingering blockedBy. Broadened
clearResumeFailureState to null both fields alongside the existing
failure cleanup, and added a defensive UI backstop so the "Queued"
badge no longer renders for tasks in the in-progress column.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Fixes two classes of task failures found while investigating stuck in-review
tasks FN-2165 (worktree base ref missing) and FN-2152 (stray .tmp-fn-2152
gitlink accidentally committed via merger amend).
FN-2165 — stale baseBranch:
- resolveWorktreeStartPoint now returns null instead of throwing
NonRetryableWorktreeError when the stored baseBranch is gone. Caller clears
task.baseBranch and falls back to branching from the default base (HEAD) so
the task self-heals instead of failing permanently.
- New TaskStore.clearStaleBaseBranchReferences() nulls baseBranch on any
dependent task when its upstream branch is deleted. Wired into
cleanupBranchForTask (archive/delete), merger branch cleanup, self-healing
orphan-branch sweep, executor dep-abort and conflict-cleanup paths, and
stale-branch recovery.
Nested worktrees:
- assertWorktreePathNotNested guard in tryCreateWorktree refuses to create a
worktree inside another registered worktree (previously produced pathological
paths like .worktrees/green-finch/.worktrees/amber-panda when rootDir pointed
at a worktree instead of the main repo).
Context-overflow recovery (FN-2182 class):
- Reduced-prompt retry budget raised from 1 → 3 within the same session.
- Adds a fresh-session requeue path when same-session retries still overflow:
task moves back to todo with worktree retained, bounded by
computeRecoveryDecision / MAX_RECOVERY_RETRIES. Prevents late-step context
exhaustion from becoming terminal.
Gitlink prevention (FN-2152 class):
- .gitignore now excludes .tmp-fn-* and .tmp-kb-* so stray worktrees at the
repo root cannot be captured by git add -A.
- Merger amend flow now scans staged entries for 160000 gitlinks and unstages
them with a loud warning; the project uses no submodules, so any such entry
is a bug (this is how f8f90f26 landed in HEAD as .tmp-fn-2152).
Tests: new coverage for baseBranch fallback, nested-worktree guard, and
clearStaleBaseBranchReferences. Full engine + core + dashboard + cli suites
pass (15349 tests).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Root cause: during a triage split the AI could set a child task's
`dependencies` to the parent id. The parent is hard-deleted after the split,
and the scheduler's dep check treats a missing id as unmet — permanently
blocking the dependent. This stranded FN-2164 behind the deleted FN-2163.
- core/store.deleteTask: refuse to delete when any live task still has the id
in its `dependencies` array. Throws TaskHasDependentsError listing dependents
so callers can rewrite or recover. Covers the triage-split path and any
future caller.
- engine/triage task_create: validate each proposed dependency before creating
a child — reject the parent id, reject unknown task ids, allow siblings
created earlier in the same split or pre-existing tasks.
- engine/triage split cleanup: wrap the parent deleteTask in try/catch that
keeps the parent alive (safer than stranding dependents) and logs the reason.
- engine/triage prompts: both the mandatory-split and proactive-split prompts
now explicitly state that subtask deps must never reference the parent.
- dashboard/routes /subtasks/create-tasks: reject parent-id deps, drop unknown
deps with an audit log entry, surface parentTaskCloseError + droppedDependencies
in the response instead of silently swallowing them.
- engine/executor: on execute entry, detect the drift state (in-progress task
with no worktree) and emit a loud log + task log entry; the existing
fresh-worktree path then recovers. Prevents silent "operating without a
worktree" behavior that we saw on FN-2152.
Tests:
core: 2907/2907 pass (+5 new, incl. deleteTask guard regression)
engine: 2554/2554 pass (+17 new, incl. task_create dep validation)
dashboard: 9064/9064 pass (+2 new for /subtasks/create-tasks).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Rename engine export and call sites to use createFnAgent consistently across runtime flows
- Update core lazy engine loader and dashboard agent-generation/planning/chat paths to reference createFnAgent
- Refresh affected unit and integration tests, including renaming pi-create-kb-agent.test.ts to pi-create-fn-agent.test.ts
- Update AGENTS.md documentation references to match the new createFnAgent name
- Update executor, reviewer, core prompt templates, and path-boundary messaging to reference the .fusion/memory/ directory instead of a single MEMORY.md file
- Broaden worktree boundary checks in pi path validation to allow .fusion/memory/ directory access from task worktrees
- Align memory backend metadata and dashboard backend labels to display file backend storage as .fusion/memory/
- Refresh core/engine tests to assert the new directory-based memory wording and boundary behavior
- Remove legacy .fusion/memory.md fallback references and normalize prompts/docs to .fusion/memory/MEMORY.md
- Stop legacy mirror writes and fallback reads in core memory backend and project memory flows
- Update engine worktree boundary checks and tests for canonical memory file handling
- Align dashboard memory/settings surfaces and route tests with canonical memory behavior
- Add model-favorites persistence test coverage for mission interview and new agent dialogs
- Update core and engine prompt text to reference .fusion/memory/ with MEMORY.md and daily-note guidance
- Allow worktree tool boundary access to .fusion/memory/ files while preserving legacy .fusion/memory.md compatibility
- Revise memory compaction/insight extraction messaging and related type/docs comments to use canonical memory paths
- Refresh dashboard and test expectations across core/engine/dashboard/docs for the new memory path wording
- Add defensive cleanup for in-memory task tracking when tasks move state or agents pause
- Improve stuck-detection and self-healing flow to reduce leaked state and missed recovery paths
- Add logging around previously swallowed errors and tighten executor cleanup behavior
- Expand restart and self-healing reliability tests to cover regression scenarios
Adds a SelfHealingManager scan that finds tasks parked in in-review with
a failed pre-merge workflow step and no active session, and sends them
back through the existing sendTaskBackForFix flow (PROMPT.md injection,
step reset, todo → in-progress). Bounded by a new maxPostReviewFixes
setting (default 1) and a per-task postReviewFixCount so a persistently-
failing verifier cannot ping-pong a task indefinitely.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Add list_agents tool for discovering agents by role, state, or includeEphemeral filter
- Add delegate_task tool for assigning work to a specific agent by ID
- Add comprehensive test coverage for both delegation tools in agent-tools-delegation.test.ts
- Update executor.ts to wire up delegation tools when agentStore is configured
- Document both tools in AGENTS.md and docs/agents.md
- Auto-delete spawned child agents when their parent task terminates (reportsTo cleanup)
- Auto-delete task-worker agents when their owned task completes
- Add includeSystem filter to AgentStore.list() and REST API
- Hide system agents by default on the agents page (show only user-facing agents)
- Wire includeSystem toggle through the API layer and AgentsView component
- Add changeset for @gsxdsm/fusion patch release
- Add comprehensive tests for agent cleanup and includeSystem filtering
- Detect context-window-limit errors from AI sessions and automatically compact
the session conversation history before retrying (one attempt per session)
- Centralize auto-compaction in promptWithFallback (pi.ts) so executor, merger,
and step-session-executor all benefit from the same mechanism
- Remove scattered context-limit error handling from executor.ts, merger.ts, and
step-session-executor.ts in favor of the centralized approach
- Add comprehensive tests for auto-compaction retry behavior in pi.test.ts
- Remove unused compactSessionContext import from step-session-executor.ts
- Add memory note documenting the centralized auto-compaction design
- Add sendTaskBackForFix() method that encapsulates the 'verification failed
— send back to in-progress' pattern
- Replace all 4 hard failure locations with sendTaskBackForFix() calls
- Update failure feedback template to mention 'sent back to in-progress'
- Update test assertions for workflow step failure cases
- Rename test that verifies passing workflow step behavior
- Add new test for verification failure send-back flow
- Add addTaskComment mock to store factories for testing
The task executor now sends tasks back to in-progress (instead of 'in-review'
with 'failed' status) when workflow step verification fails and retries are
exhausted. This allows the executor to attempt to fix the issues on the
next pass, mirroring the existing deterministic verification failure behavior
in project-engine.ts.
- Add MailboxView as a full-page navigation view replacing the modal-based approach
- Migrate MessageStore from filesystem to SQLite backend for message persistence
- Implement conversation grouping for inbox display with unread badge state
- Remove modal plumbing (useModalManager, AppModals exports) and simplify App.tsx
- Add evictStaleProcessing() to TriageProcessor for self-healing hung triage sessions
- Add comprehensive MailboxView tests and Header mailbox tests
- Update README documentation with MailboxView features
- Add CSS styles for MailboxView component
- Fix MobileNavBar tests for mailbox tab visibility
- executor.test.ts: remove unused imports (Column, StuckTaskDetector),
replace Function type with EventListener, add MockTaskStore interface
- restart.integration.test.ts: replace require() with ESM import,
replace Function types with proper function signatures
- All tests pass
- Remove unused imports across 25 files in engine package
- Remove unused variable declarations in ipc-worker.ts, child-process-runtime.ts, and mission-autopilot.ts
- Clean up unnecessary imports in agent-instructions.ts, agent-tools.ts, cron-runner.ts, executor.ts, and other modules
- Minor cleanup in notifier.ts, peer-exchange-service.ts, pi.ts, plugin-runner.ts, and other files
- Improves code quality and reduces potential confusion from unused code
- Add SkillSelectionResolver and SkillSelectionContext for deterministic skill sets
- Implement skill-resolver.ts with request/filter/resolve workflow
- Add session-skill-context.ts helper for unified skill selection across all session types
- Wire skill override into createKbAgent via AgentOptions
- Apply skill selection to triage, executor, reviewer, merger, and heartbeat sessions
- Add comprehensive tests for skill-resolver and session-skill-context
- Export skill resolver types and functions from engine index
- Add skill selection resolver pattern to memory documentation
An AI review agent (FN-1506) killed the running dashboard by finding
the process on port 4040 via lsof and running kill -9, causing exit
code 137 (SIGKILL) with no logs. This adds multi-layer guardrails:
- AGENTS.md: project-level rule reserving port 4040
- Executor/reviewer system prompts: explicit prohibition on killing
port 4040 processes, with instruction to use --port 0 instead
- Core agent-prompts.ts: same guardrails in all prompt variants
- Reviewer told to issue REVISE if executor violates the rule
- SIGHUP handlers in dashboard.ts and serve.ts for resilience
- Background engine reconciliation in dashboard/serve startup
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Background stale spec checks now skip tasks that are already in an
active work state (in-progress, in-review, done, merging, merging-pr)
to avoid interrupting work that is underway or complete.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Add spec-staleness evaluator to check task specification age before execution
- Guard executor startup and resume to prevent running tasks with stale specs
- Guard scheduler dispatch to skip stale tasks and move them back to triage
- Add comprehensive tests for spec staleness detection (7d/14d thresholds)
- Add getFusionDir mock for executor staleness check tests
