The agent log SSE endpoint was calling getOrCreateProjectStore(projectId)
and attaching the listener to the returned store. But the in-process
TaskExecutor is always bound to the default store passed to createServer,
so agent:log events were emitted on that store, never reaching the
project-scoped store's EventEmitter.
Fix: remove the getOrCreateProjectStore call and always listen directly
on the default store. Also simplifies the handler by removing the
unnecessary async IIFE (the listener can now be attached synchronously)
and eliminates the detachListener indirection.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Concurrent SSE and API requests for the same projectId both missed the
cache (storeCache.set ran after await store.watch()), creating independent
TaskStore instances with separate EventEmitters. SSE listeners attached
to one instance while mutations fired on the other, so no events reached
the browser.
Fix: add a pendingCreations promise map that deduplicates concurrent
calls, ensuring all callers share the same in-flight promise and thus
the same store instance. Also clear pendingCreations in evictProjectStore
and evictAllProjectStores. Adds a concurrent-call regression test.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Reset isDisclosureExpanded in QuickEntryBox resetForm so disclosure collapses after task creation
- Remove sticky disclosure persistence from QuickEntryBox component
- Preserve terminal sessions across transient WebSocket disconnects with buffer/replay
- Buffer and replay initial terminal state for prompt visibility on reconnect
- Update tests for non-persistent disclosure, terminal reconnect, and server routes
- Remove unused UsageIndicator tests and TaskDetailModal test cleanup
- Preserve Claude reset timestamps in the usage data model (usage.ts)
- Display session reset time in the UsageIndicator dashboard component
- Add component comments explaining reset-time rendering logic
- Add styles for the reset time display in the usage modal
- Add tests for usage data model and UsageIndicator component
- Add project-store-resolver to correctly scope SSE and WebSocket connections to the active project's TaskStore
- Fix shared resolver so multi-project setups deliver task events only to the right dashboard clients
- Add regression tests for project store resolver (routing, missing project, SSE integration)
- Remove unused import and clean up stale changeset
- Add changeset for published package patch bump
- Replace hard error returns with fetchClaudeUsageViaCli() fallback when token refresh fails
- Cover three failure paths: refresh failure, no refresh token, exhausted retry attempts
- Add comprehensive tests for all token expiry and refresh fallback scenarios
- Fix SSE event relay to properly broadcast task store events to dashboard clients
- Use named heartbeat events instead of SSE comments for reliable keep-alive
- Add missing event emission in core task store for state changes
- Add comprehensive tests for SSE pipeline, event emission, and UI hooks
- Remove broken useTerminal hook and AgentLogViewer tests, fix flaky test suites
- Fix refresh token request to use correct endpoint URL
- Set proper content-type header for token refresh requests
- Include client_id in refresh token payload
- Add tests verifying correct endpoint, content-type, and client_id in refresh flow
- Add token expiry detection with 60-second buffer before API calls
- Implement refresh_token grant flow against Anthropic OAuth endpoint
- Cache refreshed access tokens in-memory only (never written to disk)
- Retry token refresh on 401/403 responses before failing
- Add comprehensive tests for token refresh, expiry, and error scenarios
- Increase Claude fetch timeout from 10s to 75s to accommodate retries and CLI fallback
- Include response body snippet in HTTP error messages for better debugging
- Improve Claude CLI timeout error message with actionable guidance
- Export withTimeout and CLAUDE_FETCH_TIMEOUT_MS for testability
- Add comprehensive tests for withTimeout, timeout constant, and error diagnostics
- Remove DirectoryPicker component and its tests (unused multi-project UI)
- Simplify SetupWizardModal by removing directory picker integration
- Remove unused multi-project API endpoints, SSE handlers, and route definitions
- Clean up usage-limit detector and related engine references (executor, scheduler, triage)
- Fix 6 merge conflicts in usage.test.ts and update tests for simplified usage tracking
- Add browse-directory API endpoint and client for native folder selection
- Create DirectoryPicker component with manual input, browse button, and validation
- Redesign SetupWizardModal to use DirectoryPicker for project path selection
- Add comprehensive tests for DirectoryPicker and SetupWizardModal
- Remove bundled fusion skill and related extension/sync code from CLI package
- Compute live activeTaskCount, inFlightAgentCount, and totalTasksCompleted from the task store when the requested project matches the current dashboard
- Fall through to cached central DB health data for remote projects
- Add project existence check before health lookup, returning proper 404s
- Remove unused CSS styles (20 lines deleted from styles.css)
- Updated /tasks/:id/diff endpoint to use merge-base comparison
- Changed from HEAD~1..HEAD to ...HEAD syntax
- Added fallback to HEAD when merge-base fails
- Verified fix matches /tasks/:id/file-diffs endpoint pattern
- Add scripts field to ProjectSettings type for custom dashboard scripts
- Update ScriptsModal with improved UI for script management
- Add QuickScriptsDropdown component for quick script access
- Implement scripts API routes for CRUD operations
- Update CLI with multi-project commands (project add/remove/list/show/set-default/detect)
- Various UI improvements and bug fixes across dashboard components
- Migration support for multi-project architecture
- Add git diff for session-files using baseCommitSha with double-dot syntax
- Add merge-base fallback to find common ancestor when baseCommitSha is missing
- Add HEAD~1 fallback when merge-base fails (e.g., no remote main)
- Add 10-second in-memory cache for session-files computation
- Fix scheduler to immediately schedule tasks created via dashboard
- Add task:moved to done handler for immediate scheduling on status change
- Add comprehensive unit tests for scheduler immediate scheduling
- Add unit tests for session-files caching and base ref resolution
- Add fn init command for initializing new kb projects
- Add auto-migration check at CLI startup
- Add /api/setup-state and /api/complete-setup dashboard endpoints
- Add SetupState and CompleteSetupInput types to dashboard API
- Add migration and rollback documentation to AGENTS.md
- Create changeset for multi-project migration feature
- Updated mocking strategy for node:fs and node:fs/promises in file-service tests.
- Removed redundant test files and improved test implementations for better clarity.
- Enhanced GitHub polling service tests with rate limiter integration.
fix: update scheduler filesystem path from .kb to .fusion for task validation
style: clean up TaskDetailModal styling with reusable CSS classes
test: add comprehensive tests for file-service operations with mocked filesystem
- Add API endpoint to fetch git diffs for tasks in review
- Implement file diff routes with support for changed files view
- Add comprehensive test suite for file diffs endpoint
- Update dashboard API client for diff fetching capabilities
- Add core task fields for base commit and modified files to support diff computation
- Implement dashboard and API wiring for task changes, including Task Detail Changes tab integration
- Update executor flow to capture modified files during execution for accurate diff display
- Add tests and changeset covering the diff viewer feature
- Optimize backup tests with fake timers and deterministic timestamps
- Enable Vitest fileParallelism in core, engine, cli, and dashboard packages
- Fix SettingsModal scope initialization regression covered by merge resolution
- Update affected dashboard and engine tests for the new parallel-friendly behavior
- Add a changeset and document test optimization patterns in AGENTS.md
- Add task file diff API support with cached changed-file snapshots and robust diff generation for added, deleted, modified, and renamed files
- Integrate the dashboard changed files viewer into task flows without reintroducing removed board project context UI
- Expand route coverage for project-aware task listing and changed-file diff behavior, including rename diff regression cases
- Include the published package changeset for the new changed files diff viewer feature
- Add deeper MissionStore integration tests for hierarchy integrity, rollups, reorders, and task link cleanup
- Extend mission API end-to-end coverage for feature patching, validation, and cascade scenarios
- Fix TaskStore task upsert placeholders to preserve mission slice linking fields after merge resolution
- Tidy mission CLI help text and include the published package changeset for the missions launch
- Add project settings support and dashboard API helpers for custom scripts
- Add scripts CRUD and run routes backed by terminal sessions with validation and duplicate checks
- Add Scripts modal and header integration for managing and launching project scripts
- Add regression coverage for scripts routes and API client helpers
- Add activity feed fallback to the local project log when central activity is empty or unavailable
- Pass configured build commands into executor and merger prompts with hard-gate instructions
- Add merger build-failure reporting via report_build_failure and stop commits when verification fails
- Expand engine and dashboard regression coverage for activity feed fallback and merge build verification
- resync dashboard tasks after stream reconnects and when the tab becomes visible again
- add regression coverage for stale board recovery and usage-fetch edge cases
- replace direct Claude usage API calls with the CLI-based usage command
- document the dashboard resync safeguards and include a published package changeset
- Tighten mission route status validation with generic typing to preserve exact status unions
- Expand mission API end-to-end coverage for feature patching, invalid statuses, and task unlink support in the mock store
- Point dashboard Vitest core alias at the package index so mission route tests resolve the full mission store API
- Add a changeset for the sticky New Task modal action button position fix
- Add dual-source fallback: try CentralCore first, fall back to store.getActivityLog()
- Adapt per-project ActivityLogEntry[] to ActivityFeedEntry[] format with projectId="local"
- Handle CentralCore init failures gracefully without returning 500
- Add comprehensive tests covering central data, empty state, init failure, and query params
- Add JSDoc documenting the fallback strategy and data source priority
- Replace direct HTTPS requests to api.anthropic.com with `claude usage --json` CLI subprocess call, eliminating 429 rate-limit errors
- Remove retry/backoff logic (_sleep, CLAUDE_MAX_RETRIES, exponential delay) that is no longer needed
- Add specific error handling for CLI-not-found (ENOENT) and timeout scenarios
- Update tests to use setupClaudeMocks helper that handles both keychain and CLI mocking
- Add changeset for the patch fix
- Fix SQL parameter mismatch in db-migrate task INSERT (add baseCommitSha, modifiedFiles, sliceId columns)
- Fix dashboard routes type predicate to guard against null entries from git diff parsing
- Stabilize central-core activity cleanup test with fake timers
- Align executor test assertions with updated steering comment log messages
- Align store test comment log assertion with new action/outcome format
