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>
- Add Frontend UX Design template for verifying UI/UX design implementation
- Include WCAG 2.1 compliance checks, design system adherence, and user flow validation
- Add template to WorkflowStepManager with category-based organization
- Update documentation with new template description
- Add tests for template API endpoints and workflow step manager
Root cause: if a merge crashed or the process restarted mid-merge, the
"merging" status was never cleared. On next startup the stale task kept
its "merging" status while the queue moved on to the next task, resulting
in two tasks appearing to merge at once.
Two fixes:
1. Add "merging"/"merging-pr" to BLOCKING_TASK_STATUSES so tasks with
active merge status are not re-enqueued by the retry sweep.
2. Clear stale "merging" statuses during startup merge sweep — no merge
is actually running at engine start, so any such status is a leftover.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Add process lifecycle diagnostics for dashboard and serve commands
- Add SQLite database health check to diagnostics endpoint
- Add store listener count diagnostics for debugging subscription leaks
- Audit and fix SSE connection management to prevent connection leaks
- Audit and fix timer/interval cleanup in engine and CLI shutdown handlers
- Fix res.on() call guard for test mocks compatibility
- Fix variable declaration ordering in serve.ts
- Update memory with diagnostic findings for future debugging
This change aligns agent editing surfaces across dashboard UI, API payloads,
and import/template flows so every first-class editable agent field is
round-trippable.
Backend changes:
- Add memory and bundleConfig fields to POST/PATCH /api/agents routes
- Fix agent-companies-parser to use first-class fields (title, icon, role,
reportsTo, instructionsText) instead of metadata fallbacks
- Update import dry-run preview to show more manifest fields
UI changes:
- Enable identity field editing (name, title, icon, role, reportsTo) in
AgentDetailView ConfigTab
- Add instruction bundle configuration (mode, entry file, files, external
path) to ConfigTab
- Add memory field to NewAgentDialog
- Fix AI generation mapping to preserve systemPrompt as instructionsText
- Update AgentImportModal preview to show icon, reportsTo, and instructions
Test fixes:
- Update parser tests for new first-class field behavior
- Add MissionExecutionLoop mock for serve tests
Documentation:
- Add agent field parity matrix to docs/agents.md
- Create changeset for @gsxdsm/fusion
- Add recordValidatorFailures() method to store assertion failures with generated IDs
- Add createGeneratedFixFeature() to create fix features with lineage tracking
- Add getFeatureLoopSnapshot() to return complete loop state with retryBudgetRemaining
- Add getValidatorRunsByFeature() returning runs ordered by startedAt DESC
- Add getFailuresForRun() returning failures ordered by createdAt ASC
- Add transitionLoopState() with validation for all valid loop state transitions
- Implement retry budget enforcement: blocks implementing when budget exhausted
- Add rowToFailure() and rowToLineage() converters
- Add generateFailureId() and generateLineageId() ID generators
- Add DEFAULT_IMPLEMENTATION_RETRY_BUDGET constant (3 attempts)
- Update MissionFeatureLoopSnapshot interface to include retryBudgetRemaining
Co-authored-by: factory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>
This commit adds the schema migration and types for the mission execution loop validation system:
- Adds loop state tracking columns to mission_features table (loopState, implementationAttemptCount, validatorAttemptCount, lastValidatorRunId, lastValidatorStatus, generatedFromFeatureId, generatedFromRunId)
- Creates mission_validator_runs table for tracking validation runs
- Creates mission_validator_failures table for assertion failure records
- Creates mission_fix_feature_lineage table for tracking fix feature relationships
- Adds workflowStepRetries column to tasks table for retry tracking
- Adds FEATURE_LOOP_STATES and VALIDATOR_RUN_STATUSES enums
- Updates TaskStore to support workflowStepRetries field
- Updates TaskExecutor to handle workflow step failures with retry logic
Co-authored-by: factory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>
- Add projectId parameter to refineText API and TaskForm for proper scoping
- Update mission routes to require projectId and add regression tests
- Add mission interview fallback to mission-level context when target not found
- Add AI refine route scoping tests for project isolation
- Update component tests to expect projectId argument
- Update prompt-keys test counts for new system prompts
- Add FileMemoryBackend with atomic writes, persistence, and conflict resolution
- Add ReadOnlyMemoryBackend for read-only/external memory management
- Add memoryBackendType setting to select backend type (file or readonly)
- Add GET /api/memory/backend endpoint to query current backend status and capabilities
- Update AGENTS.md, README.md, and docs with architecture and settings guidance
- Add memory-backend.test.ts with comprehensive tests for all backends
- Add routes.test.ts coverage for /api/memory/backend endpoint
- Fix settings parity test to include new memoryBackendType key
- Update project memory instructions to encourage selective writes instead of unconditional appends
- Instruct agents to consolidate existing entries rather than add duplicates
- Add guidance to skip memory updates when no durable learnings were discovered
- Clarify what qualifies as durable learnings vs task-specific trivia
- Update user-facing memory documentation to reflect new selective behavior
- Add test coverage for selective memory write instructions
- Add showQuickChatFAB boolean setting to ProjectSettings type
- Add useAppSettings hook function for accessing the setting
- Wire QuickChatFAB visibility in App.tsx based on setting
- Hide QuickChatFAB in MobileNavBar when setting is false
- Add Settings UI toggle for the new setting
- Add CSS for hiding QuickChatFAB on mobile when disabled
- Add comprehensive tests for all affected components
- Fix QuickChatFAB onOpenChange test to use controlled mode
- Update settings reference documentation
Bundles staged work-in-progress modifications across multiple packages
(routes, store, agent-instructions, self-healing, QuickEntryBox, etc.)
plus the dashboard theme-data.css preload fix.
Note: an unstaged 621-line deletion in .fusion/memory.md was deliberately
NOT committed — it appears to be an accidental overwrite of architecture
notes and is left in the working tree for review.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
GET /api/tasks was returning ~69 MB of JSON per call (67.9 MB of agent
logs across 1199 tasks), causing the dashboard to hang for 2+ minutes.
- core: extend listTasks() with slim and includeArchived options
- dashboard: GET /api/tasks now uses slim mode and excludes archived
by default; ?includeArchived=1 opts in
- frontend: lazy-load archived tasks when the archived column is first
expanded via new useTasks.loadArchivedTasks()
- engine: self-healing maintenance now auto-archives done tasks older
than 48h (data stays in SQLite, column flips done -> archived)
- tests: slim mode + includeArchived coverage in store.test.ts;
routes.test.ts assertion updated for new args
Also bundles in-progress test-setup noise filters and pre-existing
QuickEntryBox/routes test work that was already modified locally.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Guard PATCH /api/missions/features/:featureId to reject status
transitions to execution states (triaged, in-progress, done, blocked)
when feature has no taskId
- 'defined' status remains always allowed (initial state)
- Non-status field updates (title, description) are unaffected
- Add 6 new tests covering guard behavior and edge cases
- All 147 mission-e2e tests passing
- Add Routine Engine Integration documentation to project memory
- Fix PROMPT_KEY_CATALOG by adding missing executor role to agent-generation-system and workflow-step-refine entries
- Update test to expect 6 executor keys instead of 4
- Add prompt-overrides module with template resolution and instruction injection
- Wire prompt overrides into agent generation flow via POST /api/agents route
- Support template-based prompt customization with role-based assignments
- Add agent generation tests and routes tests
- Document new prompt override settings in settings reference
- Add ESLint configuration (eslint.config.mjs) for TypeScript/JavaScript linting
- Update executor prompts to include lint instruction before code submission
- Add lint check to triage prompt validation workflow
- Update agent prompts to emphasize lint compliance as quality requirement
- Add lint tool to agent toolset with file-level rule disabling capability
- Include lint in CI workflow with non-blocking status
- Update tests to verify lint-inclusive prompt behavior
- Add documentation for lint integration in contributing.md
- Add changeset for @gsxdsm/fusion minor release
- Add verification runner that executes testCommand then buildCommand before merge completion
- Verification runs on all merge paths (AI resolve, auto-resolve, -X theirs)
- If verification fails, merge is aborted and task stays out of done
- Add comprehensive tests for merger verification logic
- Fix routine-store test variable reference bug (created.id vs routine.id)
- Add changeset for @gsxdsm/fusion patch release
- Add RoutineRunner class for routine execution via heartbeat system
- Add RoutineScheduler class for cron-based routine polling
- Add triggerManual and triggerWebhook methods for API and webhook triggers
- Wire RoutineScheduler into InProcessRuntime lifecycle
- Add routine trigger and webhook API endpoints
- Fix type mismatches between PROMPT and actual FN-1519 types
completeRoutineExecution was reading from the DB outside the per-routine
lock then calling recordRun which acquires the lock internally. This
allowed concurrent operations on the same routine to hit SQLite
simultaneously. Inlined the logic inside a single withRoutineLock call
so the read and write are serialized.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The RoutineRunner and RoutineScheduler were written against a different
interface than what RoutineStore actually implements, causing TypeError
crashes as soon as any routine became due. This adds the missing
agentId/catchUpLimit fields to the Routine type and DB schema, adds
startRoutineExecution/completeRoutineExecution/cancelRoutineExecution
methods to RoutineStore, and fixes all property name mismatches
(lastExecutedAt→lastRunAt, trigger.cron→trigger.cronExpression,
policy value alignment) in the runner, scheduler, and tests.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The dashboard was sporadically hanging for several seconds during task
creation and settings loading due to multiple compounding issues:
- checkForChanges() polled every 1s with SELECT * FROM tasks + full
JSON.stringify comparison, blocking the Node.js event loop. Now uses
incremental polling (only changed tasks via updatedAt filter).
- allocateId() called readConfig() which always ran listWorkflowSteps(),
adding unnecessary DB queries while holding the serialization lock.
Now uses readConfigFast() that skips workflow steps.
- Task creation triggered listWorkflowSteps() up to 3 times per request.
Added in-memory cache with invalidation on create/update/delete.
- Route handlers used getSettings() (slow path) where getSettingsFast()
suffices (POST /tasks, GET /config).
- SSE effect in useTasks had searchQuery and refreshTasks in its
dependency array, causing EventSource teardown/rebuild on every search
change. Moved to refs since the EventSource URL doesn't use searchQuery.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
