- Add tokenCap to ProjectSettings with tokenCapEnabled flag, default 500K tokens
- Create TokenCapDetector class with usage tracking and cap enforcement
- Wire TokenCapDetector into executor agentWork() to auto-fail tasks exceeding cap
- Add token cap input fields to Settings Modal dashboard UI
- Add comprehensive unit tests for TokenCapDetector (174 lines)
- Executor skips redundant moveTask call on manual retry (was causing race with scheduler)
- Scheduler now triggers scheduling on todo column transitions to pick up retried tasks
- Add tests for retry race condition fix covering both executor and scheduler paths
- Add spawn settings (maxSpawnedAgentsPerParent, maxSpawnedAgentsGlobal) to ProjectSettings type
- Implement AgentStore access and spawned agent tracking in TaskExecutor
- Add spawn_agent tool with child agent execution in isolated worktrees
- Wire up child termination on parent session end and system prompt generation
- Add comprehensive test suite (554 lines) covering spawning, limits, termination, and error handling
- Document agent spawning architecture, IPC protocol, and usage in AGENTS.md
When tasks fail and retry, the old worktree lingered on disk and the new
worktree got a random name causing worktrunk to report it as unassigned.
Now: retry clears worktree/branch fields, removes old worktree from disk,
and createWorktree returns the actual branch name (including -2 suffixes)
so task.branch always matches reality.
Also fixes pre-existing test failures: Terminal WebSocket mock sessions
missing lastActivityAt, and TerminalModal keyboard overlap tests leaking
cached _initialViewportHeight between tests.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add ContextLimitDetector to detect agent loops via repeated tool call patterns
- Implement compact-and-resume strategy: summarize conversation and restart agent from current step
- Add loop recovery to StuckTaskDetector with configurable attempt tracking and retry limits
- Extend executor with automatic loop recovery on context limit detection
- Add loop recovery support to pi executor with same compact-and-resume pattern
- Add comprehensive tests for context-limit-detector, stuck-task-detector loop detection, executor, and pi recovery
- Add changeset for patch bump to @gsxdsm/fusion
- Update README with loop detection and recovery documentation
- Add phase field to workflow step definitions (pre-merge vs post-merge) with persistence and API
- Execute pre-merge steps in executor before merge; post-merge steps in merger after successful merge
- Pre-merge failures block merge and keep task in in-review; post-merge failures are logged only
- Expose phase controls and phase-aware results in the dashboard UI
- Add changeset for the published @gsxdsm/fusion package
- Add memoryEnabled boolean setting to core types and default settings
- Add Memory section toggle to SettingsModal dashboard UI
- Integrate memoryEnabled check in executor and triage engine prompts
- Add tests for settings UI, executor prompt behavior, and triage prompt behavior
- Update SettingsModal section count test for the new Memory section
When the stuck task detector killed a task and immediately called
moveTask("todo"), the scheduler could re-dispatch the task before the
old execution's finally block cleared this.executing. The new execute()
call hit the guard and silently returned, stranding the task in
"in-progress" with no active session or worktree (seen on FN-810/FN-912).
Move the requeue responsibility from StuckTaskDetector.killAndRetry to
the executor's finally block, which runs after this.executing.delete().
The beforeRequeue budget check now runs before session.dispose() and its
result is passed via StuckTaskEvent.shouldRequeue → markStuckAborted().
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Fix resolveBaseBranch to use stored branch name and consistent fusion/ prefix
for both explicit deps and blockedBy paths (was using kb/ for blockedBy)
- Add main branch checkout verification in merger before squash merge to prevent
feature code from landing on wrong branch lineage
- Align all branch prefix references from stale kb/ to fusion/ across executor,
merger, store, and routes
- Fix executor test OOM by mocking merger fully, adding fake timers to retry
tests, and switching vitest pool to vmThreads
- Update all test assertions to use fusion/ branch prefix
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Instead of immediately failing when an agent finishes without calling
task_done, spawn a fresh session with a recovery prompt that asks the
agent to review the worktree state and complete the task. Only fail
if the retry also doesn't produce a task_done call.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add explicit no-progress loop detection signals to executor agent sessions
- Enhance stuck-task detector with configurable strategies (timeout, no-progress, combined)
- Add comprehensive test coverage for stuck-task detector (371 lines of tests)
- Remove modelFilter test utilities (198 lines of dead test code)
- Simplify modelFilter.ts by removing unused filtering logic
- Clean up dashboard server startup and minor README fix
- Extend WorkflowStep type with optional modelProvider, modelId, validatorModelProvider, validatorModelId fields
- Add API endpoints (PATCH /api/workflow-steps/:id) to update model overrides on workflow steps
- Update executor to use per-step model overrides when running workflow step agents
- Add AgentDetailView support for displaying workflow step model configuration
- Add store and executor tests for new model override behavior
- Auto-resume in-progress tasks that were paused and then unpaused, picking up from their last step
- Add event-driven scheduler trigger on task unpause so resumed tasks are immediately scheduled
- Update Store and TaskStore with unpause handling, type changes for resume tracking
- Refactor executor, merger, and scheduler tests for improved reliability
- Remove dead code: mission-interview, pr-comment-handler, taskStuck util, OpenRouter model sync changeset
- Update AGENTS.md documentation for pause/unpause behavior
- Fix executor bug where tasks paused during execution with a graceful session exit were silently dropped instead of moved back to todo
- When session.dispose() resolves the prompt without throwing, the paused-aborted flag now triggers moveTask(todo) so the scheduler can resume after unpause
- Add executor test suite covering pause via graceful exit, pause via abort error, and pause during session disposal
- Document corrected pause/unpause behavior for in-progress tasks in README and AGENTS.md
1. Scoping rules in agent prompts: executor, triage template, and reviewer
now explicitly forbid deleting/gutting modules, settings, interfaces,
exports, or test files outside the task's declared File Scope. Reviewer
will REVISE if out-of-scope removals are detected.
2. Pre-merge diffstat scope check: merger.ts validates the git diffstat
against the task's PROMPT.md File Scope before merging. Large deletions
outside scope are logged as warnings on the task (soft guardrail).
3. Changeset requirement for feature removal: triage template now requires
a .changeset/ entry when removing existing functionality. Executor and
reviewer enforce this requirement.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add state to tasks: persisted in DB, gates scheduler pickup to prevent immediate retry of transient failures
- Introduce shared recovery-policy module with bounded exponential backoff (1s → 60s, max 5 attempts)
- Wire recovery policy into executor, scheduler, and triage so all agents respect the same retry cadence
- Persist retry state (attempt count, next eligible time) in task metadata via store and DB schema
- Add DB migration for new retry columns and update schema tests
- Update README with recovery policy documentation
- Refactor dashboard Header component and styles, consolidate header tests
- Fix session-files route tests to align with updated route signatures
- Add git worktree prune as first recovery step before branch deletion to release stale locks
- Add git update-ref -d fallback when git branch -D fails on corrupted references
- Handle stale references in the fallback conflict path (not just primary path)
- Expand invalid-reference pattern matching to cover additional error outputs (unable to resolve reference, stale file handle, not a valid ref, unable to delete ref)
- Add detailed logging at each recovery step for operational traceability
- Add describeModel() helper in pi.ts to format provider/model info for logging
- Log resolved model details in executor, reviewer, and triage agent creation
- Update executor and reviewer to call describeModel before session start
- Add unit tests for describeModel covering all input combinations
- Fix test mocks to account for new describeModel dependency
- Fix addSteeringComment to skip auto-refinement and write directly to steeringComments array
- Update formatCommentForInjection to accept SteeringComment type instead of raw string
- Export SteeringComment type from @fusion/core for executor usage
- Add tests verifying steering comments are persisted to task store
- Fix executor import to use exported SteeringComment type
- Add 'branch' field to Task and ArchivedTaskEntry types with DB migration
- Executor stores branch name on task after worktree assignment
- Merger reads branch from task metadata instead of relying on worktree state
- Implement non-destructive conflict recovery in prepareForTask with reset/clean
- Add tests for branch storage, merger branch reading, and worktree recovery
- Add branch conflict recovery to worktree-pool prepareForTask when checkout fails
- Add fallthrough in executor to create fresh worktree when pool preparation fails
- Add tests for branch conflict recovery in worktree-pool
- Add tests for executor fallthrough on pool preparation failure
- Add rate-limit retry utility with exponential backoff, jitter, and configurable limits
- Integrate retry logic into executor, triage, and merger agent sessions
- Export retryOnRateLimit from engine package index
- Remove unused DirectoryPicker component, stale CSS, and dead dashboard routes
- Update SetupWizardModal and usage tests to reflect cleanup
- Add mission CLI commands and pi extension tools for creating missions, milestones, slices, features, and task links
- Extend core mission storage and schema migrations with auto-advance support, task slice linkage, and comment normalization coverage
- Wire mission-aware scheduler and executor behavior so linked features progress with task execution and completed slices can auto-activate follow-on work
- Add regression tests for mission CLI parsing, extension behaviors, scheduler mission semantics, and executor integration
- Add `setupScript` property to ProjectSettings to reference a named script
- Execute the referenced setup script after worktreeInitCommand on fresh worktrees
- Log success/failure of setup script execution; warn and skip if script name not found
- Skip setup script for pooled/recycled worktrees (fresh only)
- Add changeset (minor) for the new setting
- Add mission CLI commands: create, list, show, delete, activate-slice
- Add pi extension tools for mission management
- Add MissionStore helper methods and Mission types
- Add scheduler mission awareness for automatic slice activation
- Add engine integration wiring for mission-aware task execution
- Add dashboard mission routes and API endpoints
- Link tasks to missions via sliceId field with autoAdvance support
- Add report_build_failure tool to merger agent for explicit build failure signaling
- Run buildCommand before finalizing merge, abort if build fails with git reset --merge
- Treat build failures as fatal errors (no retry) and keep task in in-review
- Add build verification instruction to executor system prompt
- Update executor to use steeringComments field for mid-execution guidance
- Merge steeringComments and comments into unified comments field in Task type
- Update TaskStore to use single comments array instead of separate steeringComments
- Add database migration to convert existing steeringComments to comments
- Update executor to inject all comments into AI execution context
- Update dashboard SteeringTab to use unified comments API
- Update CLI task steer command to use comments field
- Update PR comment handler to add comments via unified API
- Add project subcommands: list, add, remove, show, set-default, detect
- Add --project flag support for all task commands
- Create project-context.ts utilities for project resolution
- Add defaultProjectId to GlobalSettings type
- Implement project auto-detection from cwd
- Update CLI argument parsing for global --project flag
- Add multi-project CLI documentation to AGENTS.md
- Add modifiedFiles and baseCommitSha fields to task schema for tracking changes
- Capture list of modified files during task execution in the executor
- Add GET /tasks/:id/diff API endpoint to retrieve file list and patches
- Create TaskChangesTab component with expandable file diffs
- Integrate Changes tab into TaskDetailModal for in-progress, in-review, and done tasks
- Add CSS styles for diff viewer with syntax highlighting
- Update API client with fetchTaskDiff function and Project Management types
- Optimize backup tests using fake timers instead of real timeouts
- Enable parallel file execution in core, engine, CLI, and dashboard packages
- Add inline test helpers to reduce dependencies in dashboard routes tests
- Update executor tests with exact command matching and improved assertions
- Update AGENTS.md with test optimization patterns (fake timers, unique temp dirs)
- Add database migration to merge steeringComments into comments field
- Update SQLite schema and queries to use unified comments column
- Refactor TaskStore to handle single comments field instead of dual fields
- Update dashboard components (SteeringTab, TaskCard) for unified comments
- Update engine executor and PR comment handler for new field structure
- Remove deprecated steeringComments from types and interfaces
- Add modifiedFiles and baseCommitSha tracking during task execution
- Create GET /tasks/:id/diff API endpoint for file list and patches
- Build TaskChangesTab component with expandable file diffs
- Integrate Changes tab into TaskDetailModal for in-progress, in-review, and done tasks
- Add database columns and types for diff tracking
- Update executor to capture modified files during agent sessions
- Add project API methods and types (fetchProjects, registerProject, fetchProjectHealth, etc.)
- Add ProjectCard component with health metrics, status badges, and pause/resume actions
- Add server-side project management routes for multi-project orchestration
- Add ActivityFeed component with grouped entries and project badges
- Add SetupWizard component with 5-step project creation flow
- Fix TypeScript errors in server routes for listProjects and getGlobalConcurrencyState
