Centralized routing policy helpers in core and aligned both direct assignment and delegation to route engineer-role agents consistently, ensuring assigned tasks respect explicit engineer routing the same way delegated tasks do. Added comprehensive test coverage across core, CLI, dashboard, and engin
Fusion-Task-Id: FN-3935
Fusion-Task-Lineage: 069f4d54-f7a8-4bd3-a2c3-4de830de042f
Documents mesh lease recovery semantics across the agents, architecture, and multi-project reference files, adding 32 lines of clarifying documentation to explain how mesh leases are recovered in the system.
Fusion-Task-Id: FN-3452
Adds agent creation and deletion tools to the pi extension, with engine-side implementation in `agent-tools.ts`, core store integration for agent lifecycle management, and corresponding tests and documentation updates across the workspace.
Fusion-Task-Id: FN-3777
Non-ephemeral agents now default to `active` state on creation (FN-3773), with aligned test coverage across the agent store and dashboard routes. The even-realities plugin gains a polling notification system including a notification store, card rendering, diff logic, and API routes. Review state per
Fusion-Task-Id: FN-3773
Adds shared state snapshots for mesh sync (FN-3451) along with 143-line mesh route registration, unifies agent action gating classifications (FN-3742), and patches the merger engine with two autostash fixes (capture stderr with AI fallback and TOCTOU race defense). Also ships a new quick-capture fea
Fusion-Task-Id: FN-3755
This merge completes the mesh sync shared-state pipeline (FN-3451 Steps 1–3) by adding agent run snapshot limits, applying inbound snapshots, and returning shared state snapshots, while hardening agent store cache initialization during sync. It also unifies agent action gating classifications across
Fusion-Task-Id: FN-3451
- Add shared agent role policy helpers for implementation task detection and executor-role enforcement
- Provide a standardized role-mismatch error formatter for delegation and assignment paths
- Update test isolation leak detection to ignore ephemeral fusion-test-home-root temp directories while preserving baseline checks
Fusion-Task-Id: FN-3716
Introduces a multi-agent permission policy model in `@fusion/core` (types, store persistence, API exposure) with corresponding dashboard UI wiring in the agent onboarding modal and task detail modal, plus a distributed task ID overlap retry mechanism for cluster creation, an inline fast mode toggle
Fusion-Task-Id: FN-3545
This merge restores the canonical agent lifecycle with termination scoped at the run level (FN-3580, 4 steps), adds sender-side wake recipient override for messages, and introduces test isolation CI enforcement with a stuck-requeue race fix. UI changes remove terminated-agent indicators from AgentDe
Fusion-Task-Id: FN-3580
This merge completes the agent terminated state alignment (FN-3535), adding "running → terminated" transition support with consistent styling and lifecycle controls across the heartbeat engine, agent store, and dashboard UI, plus plugin author documentation improvements (FN-3537) and plugin loader t
Fusion-Task-Id: FN-3535
Drops "terminated" from AGENT_STATES. The agent lifecycle now runs through
idle | active | running | paused | error. paused (carrying a pauseReason)
absorbs every former terminated use case — manual stop, heartbeat run
termination, spawned-child cleanup. Run status (agentRuns.status) is
unchanged: "terminated" stays a valid run-status value.
AGENT_VALID_TRANSITIONS allows direct any→idle transitions so resetAgent
no longer needs the intermediate hop.
Stack-wide:
- core/agent-store: lastError clearing + resetAgent simplified.
- engine/agent-heartbeat, executor, in-process-runtime: terminated state
writes → paused; halt-state listener fires on paused/error.
- dashboard: AgentsView/AgentListModal/AgentDetailView lose the Terminated
badge/option/state-block; agent pickers no longer filter terminated;
agentHealth drops the Terminated branch; routes/state cast widened to
the new AgentState union.
Tests across core and engine updated to assert paused for AgentState and
left "terminated" intact for run-status assertions.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
When the dashboard crashes mid-heartbeat, the agentRuns row is left in
status='active' forever. HeartbeatTriggerScheduler.onTimerTick treats
any active run as "still running" and skips every subsequent tick, so
agents go silent indefinitely (observed: 6+ hours). The existing
in-memory missed-heartbeat watchdog can't help — its trackedAgents map
is wiped on process restart.
SelfHealingManager.recoverStaleHeartbeatRuns now reconciles these on
startup and during periodic maintenance: terminates active runs whose
processPid does not match the current process, has no recorded pid, or
has been active for more than 6 hours.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Added auto-claim setting UI to the Agent detail view with corresponding documentation in agents.md and test coverage for the new component behavior. The changeset marks this as a minor feature for the published `@runfusion/fusion` package.
Fusion-Task-Id: FN-3250
Adds a regression test sentinel in the engine's in-process runtime test suite to guard against runtime guard behavior regressions (FN-3324).
Fusion-Task-Id: FN-3324
Companion to the per-rootDir Database cache: when an AgentStore is
closed, drop it from the cache so the next instantiation opens a
fresh connection instead of handing out the closed handle. Also
tolerate "database is not open" on close, since multiple AgentStore
instances now share one underlying Database.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The dashboard re-instantiates AgentStore on every /api/agents request, and
each instance was opening a fresh Database — re-running schema migrations
and PRAGMA integrity_check on the full file. On a multi-hundred-MB DB that
was 3-7s per request and leaked file handles. Sharing a cached Database
keyed by rootDir reduces it to a one-time process cost.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The merge delivers three major features: a droid CLI path reconciliation extension that resolves workspace path mismatches for spawned agents, session-first quick chat with improved heartbeat prompts and a dramatically simplified QuickChatFAB component, and canonical agent asset directory naming wit
Fusion-Task-Id: FN-3033
Added a new `droid-cli-probe` module to the dashboard package with test coverage, implementing a CLI probe capability for the droid system.
Fusion-Task-Id: FN-2979
Each agent now gets its own .fusion/agents/<id>/HEARTBEAT.md procedure
file instead of sharing a single project-wide file. A one-shot
migration in AgentStore.init() re-points existing agents off the legacy
shared path and copies the legacy file's contents into each agent's
new per-agent location so operator edits are preserved.
The HeartbeatTriggerScheduler now phase-aligns the first tick to
lastHeartbeatAt + intervalMs so a process restart resumes each agent's
existing schedule rather than waiting up to a full interval before
firing again. Overdue ticks fire promptly within a small jitter window
to avoid a thundering herd at boot.
Also fixes three pre-existing QuickChatFAB test failures introduced by
ad4db8243: auto-select default model now switches to model mode whether
or not agents are present, the model tag only renders in model mode,
and one test scopes its option lookup to role="option" to disambiguate
the in-header tag from the dropdown entry.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds an opt-in `inMemory` flag to `Database`/`ArchiveDatabase` (and
`{ inMemoryDb }` to TaskStore, AgentStore, RoutineStore,
AutomationStore, PluginStore) that swaps the on-disk fusion.db /
archive.db for SQLite's `:memory:` connection. Production callers
never set the flag, so behavior is unchanged.
Test files for each store now flip the flag in `beforeEach`. The
handful of tests that exercise cross-instance persistence (open store
A, close, open store B on same dir, expect data) construct disk-backed
stores explicitly inside the test body, marked with a comment at each
site.
Wall-clock impact:
- core: 69.4s → 18.5s (3.7× faster, 3038 tests)
- dashboard: 156.6s → 30.0s (5.2× faster — improvement ripples through
any test that constructs a TaskStore)
The refactor eliminates the per-test SQLite open + WAL fsync + tmp
dir cleanup loop that dominated setup cost: ~50ms/test → ~5ms/test.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Parallel subagent pass: four typescript-pro agents on non-overlapping scopes.
Patterns applied:
- catch (err: any) { ... err.message ... } → catch (err) { ... getErrorMessage(err) ... }
using the new @fusion/core helper. Bare catch {} where the error was unused.
- SQLite row types: defined typed XxxRow interfaces per table and cast
.all()/.get() results via `as unknown as XxxRow[]` (the double cast is
required because better-sqlite3 returns Record<string, SQLOutputValue>).
- rowToX(row: any) converters: typed argument with the matching row interface.
- Dynamic settings key writes: (settings as Record<string, unknown>)[key].
- React event handlers and setState callbacks: inferred types or concrete
React.{Mouse,Change,Form}Event<...> where needed.
- pi-claude-cli: local PiMessage / PiContext duck types to avoid re-typing
pi-ai concrete shapes; typed Claude stream event message fields.
72 files changed, ~400 anys eliminated. Typecheck passes across the workspace.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Drops orphaned imports and the never-referenced V4 migration SQL constant
(V4 was inlined into runMigrations). Also drops the unused TypedEventEmitter
helper type and unused destructured values from stores.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- HeartbeatMonitor.executeHeartbeat() now checks if resolved task is done/archived
and exits before session creation with reason 'task_closed'
- When stale task came from persisted agent.taskId, clears the linkage to prevent
repeated stale activations
- HeartbeatTriggerScheduler.watchAssignments() now skips assignment callback dispatch
when assigned task is done/archived (when taskStore is available)
- Added comprehensive tests for stale-task activation guard behavior
- Add constructor guards in AgentStore and ReflectionStore that throw when rootDir is omitted under Vitest
- Keep default .fusion root resolution for non-test execution paths
- Prevent accidental test writes to unintended filesystem locations by forcing explicit paths
- Default AgentStore rootDir to resolve(".fusion") so agent data paths are absolute by default
- Default ReflectionStore rootDir to resolve(".fusion") for consistent absolute root resolution
- Validate Database kbDir is absolute and throw a descriptive error when a relative path is provided
The perf logs were temporary diagnostics used to identify that slow
reloads were caused by a registered remote node timing out in
/projects/across-nodes. Root cause is resolved and the short-circuit
for zero-remote setups (already committed in 7ea60382a) remains.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Expose getAgentStore from runtime and engine packages
- Add includeSystem filter to getOrgTree API endpoint
- Wire AgentStore into SSE endpoint for event forwarding
- Forward agent events through the SSE pipeline
- Exclude ephemeral agents from org tree API and UI
- Filter ephemeral agents in useAgents hook
- Update AgentsView test to expect includeSystem filter
- Document agent SSE event forwarding architecture
- 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
- Update AGENTS.md with authoritative run lifecycle semantics documentation
- Add API-level regression test for repeated manual run prevention
- Update run lifecycle tests to match corrected behavior
- Refactor AgentStore to use structured run records as the authoritative source for run state
- Improve run status queries with better filtering and ordering
- Add RunMutationContext type to track which agent run caused a mutation
- Thread runContext through TaskStore.logEntry, addComment, addSteeringComment, and pauseTask
- Propagate runContext from HeartbeatMonitor.executeHeartbeat to task store operations
- Propagate runContext from TaskExecutor.execute to task store operations
- Add GET /api/agents/:id/runs/:runId/mutations endpoint to query mutations by runId
- Add createTaskLogToolWithContext for heartbeat tools with run context support
- Add comprehensive tests for RunMutationContext across store and heartbeat modules
- Update memory.md with RunMutationContext usage convention
- Add BlockedStateSnapshot typing/export and AgentStore persistence APIs for last blocked heartbeat state
- Deduplicate blocked-task heartbeat comments using blockedBy + context hash, and clear blocked snapshots when tasks are no longer blocked
- Thread triggeringCommentIds/triggeringCommentType through heartbeat execution, wake context, scheduler assignment triggers, and runtime wiring
- Trigger immediate heartbeat runs from task/steering comment routes for assigned immediate-response agents, with validation for comment wake fields on /api/agents/:id/runs
- Expand core, engine, and dashboard tests to cover blocked dedup logic, comment-triggered wakes, validation, and skip scenarios
- Add checkout lease types and conflict error exports, plus DB schema v20 migration for checkedOutBy/checkedOutAt
- Persist checkout lease fields in TaskStore and add AgentStore checkout/release/force-release/get-holder operations
- Add dashboard checkout API routes for acquire/release/force-release/status with explicit 409 conflict and 403 holder enforcement
- Enforce checkout ownership in heartbeat execution with graceful checkout_conflict exits when another agent holds the lease
- Expand core and dashboard test coverage for schema, store behavior, API routes, and leasing workflows, and document leasing behavior in AGENTS.md
- Rename all pi extension tool registrations from kb_* to fn_* across task, mission, and agent tool families
- Update extension and skill-sync tests plus Fusion skill docs/workflows to assert and document the new fn_* tool names
- Align product-identity strings across core, dashboard, and CLI references, including GitHub import examples, User-Agent headers, and terminal TERM_PROGRAM
- Add a @gsxdsm/fusion minor changeset describing the extension tool-prefix rename
- Add soul and memory fields to agent types and AgentStore with persistence/update test coverage
- Add dashboard routes and API helpers to fetch and update agent soul/memory data
- Extend AgentDetailView with Soul, Memory, and Employees tabs and rename children labels to employees
- Normalize employee route params for type safety and add focused route/component tests for the new flows
- Add AgentBudgetConfig and AgentBudgetStatus types to runtime config and export them from @fusion/core
- Implement AgentStore.getBudgetStatus to compute usage, thresholds, over-budget flags, and next reset timestamps for daily/weekly/monthly/lifetime periods
- Implement AgentStore.resetBudgetUsage to zero token counters, record budgetResetAt, and emit agent updates while preserving other runtime settings
- Add comprehensive AgentStore budget management tests covering no-budget behavior, threshold calculations, reset behavior, and reset-day edge cases
- Add AgentRating, AgentRatingSummary, and AgentRatingInput types and re-export them from @fusion/core
- Bump core schema to v14 with an agentRatings table plus agentId/createdAt indexes
- Implement AgentStore rating methods for add/list/filter/summary/delete with score validation and rating:added events
- Extend database and agent-store tests to cover migration, rating CRUD behavior, filtering, limits, and trend calculation
