- Move 'Create Refinement Task' button from modal footer into input group adjacent to feedback textarea
- Wrap char count and submit button in new .detail-refine-input-group flex container
- Add CSS for input group with space-between layout and gap
- Add 3 tests verifying button placement, disabled/enabled rules, and sibling structure
- Update dashboard README with refinement modal UX documentation
- Add slugify and generatePresetId utilities with collision handling for unique preset IDs
- Remove manual ID input field from preset editor UI in SettingsModal
- Auto-generate preset IDs server-side from name on creation
- Simplify Header tests by removing ID-related test cases
- Deduplicate tablet header controls tests by importing from shared Header tests
- Remove unused CSS rules for removed ID input field
- Update reviewer to pass through preset ID generation logic
- Replace single overflow terminal button with split-action dropdown showing quick-run scripts
- Add script submenu in header for launching predefined commands from compact view
- Add comprehensive regression tests for split-button behavior and quick-run script actions
- Update dashboard README to document the new split-action terminal overflow UI
- Adjust title summarization threshold and clean up related test/store changes
The body-scan fallback (includes("APPROVE") etc.) would misclassify verdicts
when the reviewer mentioned a keyword in context without it being the actual
verdict — e.g. "this does not yet merit APPROVE". Replaced with a
line-anchored regex that only matches "Verdict: X" lines, and expanded the
primary regex to also catch bold/italic formatted verdicts like **Verdict: APPROVE**.
UNAVAILABLE is now the honest fallback when no structured verdict is found,
triggering a retry rather than silently misclassifying.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Change MIN_DESCRIPTION_LENGTH threshold from 140 to 200 in ai-summarize.ts
- Update shouldAutoSummarize logic and tests for new 200-char threshold
- Update UI copy in SettingsModal and API/routes references
- Update store tests to reflect the new threshold value
- Remove obsolete dashboard activity-log mobile layout test files and unused CSS
- Add changeset for patch bump
- Add mobile responsive CSS for Activity Log modal with adaptive breakpoints
- Add regression test coverage for mobile activity log layout
- Add tests for ActivityLogModal responsive behavior
- Update README with responsive layout documentation note
A deep link handler already existed at line 275 (with proper project
switching and test coverage). The duplicate added earlier caused double
toast rendering in tests.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Fix useActivityLog hook to properly load activity log entries from the store
- Add comprehensive test coverage for useActivityLog hook with 243 lines of tests
- Restore missing ActivityLogModal CSS styles (modal overlay, list, animations)
- Add ActivityLogModal component tests for empty state, loading, and error handling
- Register activity log route in App.tsx
- Update README with activity log data source documentation
- Add changeset for patch release
- Restructure Header.tsx to render Scripts as a submenu of Terminal in compact/tablet overflow menus
- Add submenu toggle state, keyboard navigation, and click-outside-to-close behavior
- Add CSS for nested submenu styling with slide-in animation and depth indicators
- Update Header tests to verify Scripts nested under Terminal with toggle interaction
- Update mobile-header-controls and tablet-header-controls tests for new menu hierarchy
- Update README documentation to reflect the Terminal > Scripts submenu structure
- Replace plain <select> dropdown with styled agent-state-filter container featuring Filter icon, hover/focus states, and consistent dashboard theming
- Add aria-label for accessibility on the state filter dropdown
- Add CSS styles for .agent-state-filter wrapper with border, color transitions, and focus ring
- Add unit test for styled filter container rendering and aria-label presence
- Document Agents view features (state filter, view modes, CRUD, health monitoring) in README
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>
- Close the Scripts modal immediately before making the async script-run API call, preventing the modal from interfering with the terminal handoff
- Add regression tests covering script-to-terminal handoff behavior (modal close timing, API call sequencing, error handling)
- Document the script-run modal handoff behavior in the dashboard README
- Move Plan and Subtask buttons next to the description input in QuickEntryBox, InlineCreateCard, and TaskForm
- Improve action button layout with consistent icon+label pattern and disabled states
- Add CSS styles for new action button group and description-adjacent controls
- Add regression tests for all three task creation surfaces ensuring buttons appear next to description
- Document the description-adjacent AI controls pattern in dashboard README
- Increase inline editor textarea default rows from 1 to 4 for better multi-line editing
- Add mount-time auto-resize to fit content on initial render
- Add regression tests for taller inline editor behavior
- Update documentation to reflect the new taller editor defaults
- Add CSS rules for executor status bar footer with safe-area padding
- Update App.tsx and AgentsView.tsx layout to accommodate footer space
- Add regression tests for footer-safe project layout (App.test.tsx)
- Add unit tests for ExecutorStatusBar component
- Document footer-safe dashboard layout behavior in README
- Redesign ChangedFilesModal as a two-view navigation flow (file list → file detail) with back navigation
- Fix theming and loading/reading ergonomics for the diff viewer
- Enhance useChangedFiles hook with improved loading and error states
- Add comprehensive tests for modal navigation, diff rendering, and edge cases
- Update styles with mobile-friendly layout and diff viewer improvements
- Update dashboard README with mobile navigation description
- Add ScriptRunDialog component with xterm.js terminal emulation and real-time streaming
- Align saved-script launches with backend contract (POST /api/scripts/:id/run)
- Wire dialog into App.tsx with open/close state management
- Add comprehensive styles for dialog overlay, terminal area, and action buttons
- Add unit tests for ScriptRunDialog and App integration
- Document ScriptRunDialog usage in dashboard README
- Share task list data between footer ExecutorStatusBar and board view to ensure consistent counts
- Refactor MissionManager component with improved structure and readability
- Refactor useExecutorStats hook and simplify test coverage
- Update ExecutorStatusBar component and tests for accurate count display
- Consolidate and clean up dashboard styles.css
- Update footer status bar documentation in README
- Rebuild MissionManager component with updated modal shell and list view layout
- Refactor and reorganize dashboard CSS styles (551 lines changed)
- Remove terminal-related test files and simplify terminal service/useTerminal hook
- Clean up global settings: remove unused global-settings test and simplify settings types
- Remove dead code from App.tsx, SettingsModal.tsx, and dashboard README
The ntfyDashboardHost field was incorrectly removed in 6346e22b as part of
an unrelated FN-672 commit. This restores the setting to GlobalSettings (moved
from ProjectSettings where it was misplaced), re-adds the UI field and deep
link handling, and adds schema protection so unknown keys in settings.json
survive code-level schema changes.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Reorder Header overflow menu items so Settings always appears last
- Update Header.tsx to position SettingsMenuItem as the final menu entry
- Add regression tests in Header.test.tsx verifying Settings is last in all overflow states
- Add ordering tests in tablet-header-controls.test.tsx for tablet layout
- Redesign MissionManager component with improved interaction patterns and shell layout
- Refactor Board, Column, and QuickEntryBox components for consistency
- Add comprehensive CSS styles (742 lines) for mission manager and related components
- Add MissionManager test suite (324 lines) covering new interaction behavior
- Update Board, Column tests; remove obsolete ListView and QuickEntryBox tests
- Clean up unused code in App.tsx and ListView.tsx
- Update Missions section in README and dashboard README to reflect actual UI behavior
- Pass autoExpand={false} to QuickEntryBox in Column.tsx so the board view starts collapsed
- Add test verifying autoExpand prop is false for board columns
- Update QuickEntryBox mock in Column tests to capture autoExpand prop
- Wire favorite model state into QuickEntryBox with star toggle and dropdown integration
- Add favorite model selector to Board and Column quick-add components
- Pass favorite models through ListView and App context to entry surfaces
- Add regression tests for QuickEntryBox, Board, Column, and ListView favorite behavior
- Update dashboard README to document quick-add favorite model support
- Remove unused recovery-policy module and dead code from executor/triage/scheduler
- Clean up unused types from core and simplify store methods
- 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 tablet compact-header breakpoint (769px–1024px) with condensed layout
- Implement responsive Header component with mobile/tablet/desktop modes
- Add tablet-header-controls test suite (373 lines) for compact mode behavior
- Update existing Header and mobile-header-controls tests for new responsive tiers
- Refactor server routes and styles.css for three-tier responsive support
- Document responsive header architecture in README
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>
- Simplify TaskCard inline editing to support description-only editing
- Remove terminal/PTY features: useTerminal hook, TerminalModal, xterm integration
- Remove project store resolver caching and session file route handling
- Simplify executor worktree pool and remove terminal-related code
- Clean up dead code from dashboard server, styles, and engine
- Update QuickEntryBox and all related tests for the simplified editing model
- Remove unused .gitignore entries and update dashboard README
- 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
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
- Display resolved executor and validator model names in the Agent Log header of TaskDetailModal
- Add comprehensive UI tests for model resolution in TaskDetailModal
- Add missing fetchSettings mock to PlanningModeModal test
- Remove unused project-store-resolver module, UsageIndicator tests, and usage tests
- Clean up unused styles and API references
- 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
- Add favoriteProviders and favoriteModels state to AppInner
- Initialize favorites from fetchModels response
- Add handleToggleFavorite and handleToggleModelFavorite callbacks with optimistic updates and rollback on error
- Pass favorite state and toggle handlers down to ListView component
- Import updateGlobalSettings API for persisting favorite selections
- Restructure TerminalModal header layout to support mobile viewports with proper flex sizing
- Add CSS styles for responsive terminal header with title truncation and icon wrapping
- Add inline comment documenting mobile header layout contract
- Add comprehensive TerminalModal test suite covering mobile layout, resize behavior, and header interactions
- Clean up unrelated store test and changeset remnants from prior work
- Derive readable refinement titles from source task title or first line of description instead of raw task ID
- Fall back to task ID only when both title and description are empty/whitespace
- Collapse internal whitespace in description-derived labels for clean output
- Update PROMPT.md heading to match the new readable refinement title
- Add comprehensive tests for title precedence, whitespace handling, and edge cases
- Document refinement naming behavior in README
- Add changeset for @gsxdsm/fusion patch release
- Fix mobile visual viewport positioning to handle dynamic viewport changes (address bar, keyboard)
- Improve CustomModelDropdown with better mobile scroll/resize handling
- Add comprehensive visual viewport regression tests for CustomModelDropdown
- Refactor build-exe tests for cleaner structure and coverage
- Harden dashboard startup probe to handle exit race condition in build-exe tests
- Add STANDALONE.md documenting Bun's lack of node:sqlite support in standalone builds
- Refactor build-exe tests with improved wait/retry logic for process lifecycle
- Reduce flakiness in standalone executable integration tests
- Add favorite model selection to QuickEntryBox and InlineCreateCard components
- Wire up model favorites in TaskForm and TaskDetailModal
- Add model favorites management UI to SettingsModal
- Add tests for favorites in InlineCreateCard, QuickEntryBox, TaskForm, ListView, and SettingsModal
- Remove unused CommitDiffTab component and related tests
