Merge starvation recovery was hardened across the engine: self-healing now detects and clears blocked-in-review tasks that starve the merger, the project engine gains defensive recovery hooks for stalled merges, and tests cover the new recovery paths.
Fusion-Task-Id: FN-4084
Adds ntfy access token support to Fusion's notification system, wiring the token through the core settings schema, dashboard UI (SettingsModal), engine notifier, and notification pipeline, with corresponding tests across core, dashboard, and engine packages; also updates settings and storage documen
Fusion-Task-Id: FN-4078
FN-4093 introduced a reusable `setup-node-pnpm` composite GitHub Action and refactored the PR checks workflow to use it, reducing the workflow file by over 60 lines while adding a contract test for the action. A small follow-up stabilizes workspace quality gates by adding an export in the core store
Fusion-Task-Id: FN-4093
Restores green verification gates for executor review verdicts (FN-4095) with a targeted core store change, and adds a regression test ensuring the QuickChatFAB scrolls to the bottom when opened.
Fusion-Task-Id: FN-4095
Adds a squash-merge audit gate that verifies duplicate-cherry-pick commits and file-overlap losses after any squash merge into main, restoring the post-squash audit step to the merge workflow with test coverage for both the audit logic and the broader merger lifecycle.
Fusion-Task-Id: FN-4067
Fusion-Task-Lineage: 593aa917-5640-495b-b1ae-80c3eaa09d01
Hardened SQLite concurrent-write safety in `@fusion/core` by auditing WAL enforcement and adding immediate write transactions to prevent stale reads under load, with executor semaphore limit tests in `@fusion/engine` validating the concurrency bounds; added a new `store-concurrent-writes.test.ts` in
Fusion-Task-Id: FN-4083
Fusion-Task-Lineage: adabe51a-a9c4-481a-850d-985c373df51a
The merge replaces all real `setTimeout`/`sleep` polling with deterministic event-based waits across the executor pause integration tests, restart integration tests, dashboard tests, and chat-store tests, making the suite faster and more reliable.
Fusion-Task-Id: FN-4087
Implements workflow revision forking with a new per-project setting, allowing follow-up tasks to inherit a fork of the calling task's workflow rather than the original revision. The feature adds classification logic, fork execution, and the settings UI, with docs and tests covering the full flow.
Fusion-Task-Id: FN-4070
Restores the bound reports lookup in the heartbeat executor (FN-4056), updating the relevant logic in `agent-heartbeat.ts` with a corresponding test and documentation clarification in `agents.md`.
Fusion-Task-Id: FN-4056
Adds reviewer prompt compaction and automatic retry logic when the reviewer hits context-window limits, with matching test coverage and a changeset for `@runfusion/fusion`.
Fusion-Task-Id: FN-4082
Adds task priority support to the task creation flow, spanning the engine runtime (`agent-tools.ts`, `triage.ts`), the CLI extension interface (`extension.ts`), and corresponding documentation and tests. The changeset bumps `@runfusion/fusion` as a minor version.
Fusion-Task-Id: FN-4074
Fusion-Task-Lineage: 726c0ae6-90fb-4c13-ace1-215a2becb2a9
Added delegate collision detection and error surfacing across the task creation and delegation pipeline. The core store now throws a typed `DelegateCollisionError` when `createTask` encounters a duplicate task ID, the engine's `delegate_task` tool propagates these errors with context, and the CLI ex
Fusion-Task-Id: FN-4045
Adds defensive recovery logic for stale review sessions: the executor now guards against resuming in an invalid worktree context, the self-healing manager recovers stale session paths and unblocks dependent tasks, and the restart recovery coordinator is updated to classify these cases correctly. Inc
Fusion-Task-Id: FN-4031
Adds a global `thinkingLogEnabled` setting that gates AI thinking log persistence across the engine (executor, reviewer, merger, triage, step-session) and exposes the control in the dashboard Settings modal, with tests verifying settings parity and modal behavior.
Fusion-Task-Id: FN-4062
Enforces memory write contracts for triage and planning agents, adding tests in both `project-memory.test.ts` and `triage.test.ts` to guard the contract boundaries established in `project-memory.ts`.
Fusion-Task-Id: FN-4063
Fusion-Task-Lineage: 5f86295b-d583-4757-9aa5-30a64b2cc2d6
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
Added memoization for startup slim task list reads (`store.ts`), with dashboard server routing startup reads through the memo window and tests covering the store watcher consumers in scheduler and worktree-pool, plus performance documentation.
Fusion-Task-Id: FN-4027
Codifies the worktree database open-path hydration failure with a recovery bootstrap mechanism, documented the behavior in storage docs, and added test coverage for the executor hydration degradation path — all gated behind a patch changeset for `@runfusion/fusion`.
Fusion-Task-Id: FN-4039
Hardened empty `rawResponse` handling in the memory insights module by adding defensive checks to prevent errors when the response is empty, accompanied by tests to cover those edge cases (FN-4041 Step 1).
Fusion-Task-Id: FN-4041
Added self-healing logic to prevent merge-confirmed tasks from re-entering the executor, with test coverage for both merge error recovery and self-healing behavior, plus corresponding architecture and task-management docs.
Fusion-Task-Id: FN-4019
Adds stale timer repair handling to the heartbeat scheduler with test coverage, and updates the agents documentation accordingly.
Fusion-Task-Id: FN-4033
The monolithic `merger.test.ts` (8,229 lines) has been decomposed into nine focused test suites covering conflict resolution, diff scope, merge details, merge lifecycle, post-merge, prompt and utils, session recovery, skills, and verification, along with a shared `merger-test-helpers.ts` module for
Fusion-Task-Id: FN-4014
FN-4021 refactors the core store test suite by splitting the monolithic `store.test.ts` (6,865 lines) into nine focused domain suites — activity, comments, create, movement, ops, parsing, resilience, scheduling, snapshots, update, and upsert — and introduces shared test helpers to support the split
Fusion-Task-Id: FN-4021
Fusion-Task-Lineage: 359f2dca-f7f7-4735-bcbd-92d569d2bcd5
Merges FN-4015 to add stale module-resolution incident detection and self-healing in the engine (`transient-error-detector.ts`, `self-healing.ts`) with mobile regression coverage for the shared `AgentErrorDetailsModal` layout.
Fusion-Task-Id: FN-4015
Adds health-surfacing logic for timer-triggered stale heartbeats in the engine, a new `agentHealth.tsx` dashboard utility with tests, and documentation updates for the repair behavior.
Fusion-Task-Id: FN-4020
Merged FN-4014 to fix dual-scope GitHub tracking settings persistence and validation (mission vs task scope), scoped repo saves to the relevant section, and documented the dual defaults. Also landed FN-4018 to isolate split merger temp workspaces and add regression checks for split-suite test isolat
Fusion-Task-Id: FN-4018
Adds frontend-UX auto-skip behavior for workflow steps in the executor, with tests covering the flow and documentation of the recovery logic. A changeset is included for this patch release.
Fusion-Task-Id: FN-3906
Fusion-Task-Lineage: 3c50b82b-dd03-4ba3-8d77-5aaabb8734da
Stabilizes blocked-by overlap stamping in the scheduler and adds a standalone `recover-stale-blocked-by` utility script to clear stale upstream blockers, with corresponding scheduler tests and architecture documentation.
Fusion-Task-Id: FN-3899
The merge implements post-merge model resolution, using the merger's own session model for executing post-merge prompt workflow steps instead of falling back to executor model defaults. Tests cover the resolution hierarchy, and documentation clarifies the precedence order for downstream consumers.
Fusion-Task-Id: FN-3905
Adds bounded room thread context and mailbox reply context to the chat system, implemented in `agent-tools.ts` with corresponding dashboard room handling in `chat.ts`; includes expanded test coverage for both modules and updated documentation in `agents.md`.
Fusion-Task-Id: FN-4004
Implements worktree database hydration for the executor (`worktree-db-hydrate.ts`), enabling the task executor to restore its database state from a worktree on resume, with integration in `executor.ts` and comprehensive test coverage across the new module and worktree scenarios. Documentation and a
Fusion-Task-Id: FN-3841
FN-3979 adds test coverage validating the FN-3834 overlap detection fix: Step 1 introduces overlap validation tests in the merger and self-healing test suites, and Step 2 cleans up a type-safety issue in the overlap test.
Fusion-Task-Id: FN-3979
Merged FN-3834 to add a no-op merge recovery flow to the merger, including a branch-ahead detector, short-circuit logic to exclude no-op merges from the mergeable sweep, and finalization of no-op review tasks — backed by substantial test coverage across merger and self-healing modules.
Fusion-Task-Id: FN-3834
Adds heartbeat scope-discipline guardrails to the executor's heartbeat procedure, ensuring agents stay focused on their assigned scope rather than drifting to unrelated work. Includes tests for the updated heartbeat prompts and documentation in agents.md, shipped as a patch.
Fusion-Task-Id: FN-3884
Adds task lineage storage infrastructure (FN-3990 Step 2), introducing a dedicated `task-lineage.ts` module and related types to track task identity and association relationships, with corresponding database schema and store support plus tests and documentation for the lineage reconciliation process
Fusion-Task-Id: FN-3990
Completes FN-3976 by reclassifying GitHub import action gating in the engine's `gating-classifications.ts`, updating documentation in `docs/agents.md` accordingly, and adding test coverage across four gating test suites.
Fusion-Task-Id: FN-3976
Refined self-healing recovery logic for tasks that lack a `fn_task_done` signal, aligned detection in the engine with matching test coverage and architecture documentation for the recovery behavior.
Fusion-Task-Id: FN-3962
- Add self-healing recovery to detect in-review tasks that lost their worktree and clear stale blockedBy state
- Extend restart recovery coordinator handling and tests for missing-worktree classification behavior
- Add focused self-healing and restart recovery test coverage for the new recovery path
- Document missing-worktree review recovery behavior in architecture docs
Fusion-Task-Id: FN-3972
Adds scheduler heartbeat timer reconciliation with automatic self-healing when timers drift, including tests for tracked-only monitor recovery and documentation in the agents reference.
Fusion-Task-Id: FN-3958
Implements approval guards for agent provisioning (FN-3791), adding policy-gated create/delete flows with dedupe logic, CLI tool alignment, engine run-audit coverage, and corresponding test suites, plus documentation updates and a regression fix for verification/tool docs sync.
Fusion-Task-Id: FN-3791
Adds test coverage for task environment propagation through the spawn flow, covering child agents inheriting task env, workflow step task env, and spawn hook env isolation cases in `pi-create-fn-agent.test.ts`, with minor additions to `executor-step-session.test.ts` and `executor-pause.test.ts`.
Fusion-Task-Id: FN-3968
