- Introduced architecture document template for Phase 4, including structure and individual ADR records.
- Added epics & stories template for Phase 5, detailing epic breakdown and dependencies.
- Created product brief template for Phase 2, summarizing product vision, problem statement, and target users.
- Developed requirements PRD template for Phase 3, outlining functional and non-functional requirements with traceability.
- Implemented spec command for project spec management with subcommands for loading, listing, rebuilding, and initializing specs.
- Added `findAllCoreMemoryDatabases` function to recursively locate core-memory databases in nested project structures.
- Updated `listAllProjects` to utilize the new recursive function for improved project listing.
- Enhanced `getMemoriesFromProject` and `findMemoryAcrossProjects` to support nested project structures.
feat: Introduce spec context injection in hooks configuration
- Added a new hook configuration for "Spec Context Injection" to load project specs based on prompt keywords.
chore: Add gray-matter dependency for YAML frontmatter parsing
- Included `gray-matter` package in `package.json` for parsing YAML frontmatter in markdown files.
feat: Create Spec Index Builder tool for managing project specs
- Implemented `spec-index-builder.ts` to scan markdown files, extract YAML frontmatter, and generate index cache files for different spec dimensions.
feat: Develop Spec Init tool for initializing spec directories and seed documents
- Created `spec-init.ts` to set up the directory structure and seed documents for the spec system.
feat: Build Spec Keyword Extractor for keyword extraction from prompts
- Added `spec-keyword-extractor.ts` to extract keywords from user prompts, supporting both English and Chinese text.
feat: Implement Spec Loader for loading and filtering specs based on keywords
- Developed `spec-loader.ts` to handle loading of specs, filtering by read mode and keyword matches, and formatting output for CLI or hooks.
- Replace GitHubTreeEntry with GitHubContentEntry matching actual API response
- Add runtime array validation to prevent unexpected response formats
- Normalize GitHub API types ('file'/'dir') to internal types ('blob'/'tree')
- Validate required fields (name, path, type) for each entry
This fixes potential runtime errors when GitHub API response structure
differs from expected type definition.
- Add hook quick templates component with configurable templates
- Refactor NativeSessionPanel to use new SessionTimeline component
- Add OpenCode session parser for parsing OpenCode CLI sessions
- Enhance API with session-related endpoints
- Add locale translations for hooks and native session features
- Update hook commands and routes for better hook management
- Implemented `fetchCodexConfigPreview` and `fetchGeminiConfigPreview` functions in the API layer to retrieve masked configuration files.
- Added new interfaces `CodexConfigPreviewResponse` and `GeminiConfigPreviewResponse` to define the structure of the API responses.
- Created utility functions to read and mask sensitive values from `config.toml` and `auth.json` for Codex, and `settings.json` for Gemini.
- Updated CLI settings routes to handle new preview endpoints.
- Enhanced session content parser to support Claude JSONL format.
- Updated UI components to reflect changes in history page and navigation, including new tabs for observability.
- Localized changes for English and Chinese languages to reflect "CLI History" terminology.
- Implemented exportSettings and importSettings APIs for CLI settings.
- Added hooks useExportSettings and useImportSettings for managing export/import operations in the frontend.
- Updated SettingsPage to include buttons for exporting and importing CLI settings.
- Enhanced backend to handle export and import requests, including validation and conflict resolution.
- Introduced new data structures for exported settings and import options.
- Updated localization files to support new export/import features.
- Refactored CLI tool configurations to remove hardcoded model defaults, allowing dynamic model retrieval.
Decouple CLI settings architecture from Claude-only to support multiple
providers. Each provider has independent settings UI and backend handling.
- Add CliProvider type discriminator ('claude' | 'codex' | 'gemini')
- Add CodexCliSettings (profile, authJson, configToml) and GeminiCliSettings types
- Update EndpointSettings with provider field (defaults 'claude' for backward compat)
- Refactor CliSettingsModal with provider selector and provider-specific forms
- Remove includeCoAuthoredBy field across all layers
- Extend CliConfigModal to show Config Profile for all tools (not just claude)
- Add provider-aware argument injection in cli-session-manager (--settings/--profile/env)
- Rename addClaudeCustomEndpoint to addCustomEndpoint (old name kept as deprecated alias)
- Replace providerBasedCount/directCount with per-provider counts in useCliSettings hook
- Update CliSettingsList with provider badges and per-provider stat cards
- Add Codex and Gemini test cases for validateSettings and createDefaultSettings
- Read URL tab parameter using useSearchParams
- Initialize locationFilter to 'hub' when tab=hub is present
- Allows /skill-hub route to correctly display hub content
- Fixed GITHUB_CONFIG.skillIndexPath from 'index.json' to 'skill-hub/index.json'
- Added RemoteSkillEntry.path field for directory-based skills
- Added buildDownloadUrlFromPath() helper function
- Added GitHub API-based directory download with downloadSkillDirectory()
- Added installSkillFromRemotePath() for full skill directory installation
- Modified install route to support both downloadUrl and path-based installation
- Create centralized useCliStreamWebSocket hook with module-level
message tracking to ensure each WebSocket message is processed only once globally
- Replace position:absolute with position:sticky + flexbox wrapper
for scroll-to-bottom buttons to fix viewport positioning
- Remove duplicate WebSocket handling from CliViewerPage, CliStreamMonitorLegacy,
and CliStreamMonitorNew components
Fixes:
- CLI output no longer duplicated when multiple components subscribe
to the same WebSocket feed
- Scroll-to-bottom button now stays fixed at bottom-right corner
of viewport instead of scrolling with content
Extend HookTriggerType from 5 to 12 official events (Notification, SubagentStart,
SubagentStop, PreCompact, SessionEnd, PostToolUseFailure, PermissionRequest).
Add templates: stop-notify, auto-format-on-write, auto-lint-on-write,
block-sensitive-files, git-auto-stage, post-edit-index, session-end-summary
across notification/automation/indexing categories. Fix install pipeline to
use correct nested settings.json format with _templateId metadata for precise
detection. Redesign templates UI as responsive card grid with per-template icons.
- Add tags: memory.tags to importMemories upsertMemory call to prevent
tag loss during cross-project import
- Guard buildCompressionMetadata against division by zero when
originalSize is 0
Implement 5 interconnected memorycore enhancements:
1. Tags backend: add tags TEXT column to memories table with migration,
JSON array storage, full CRUD support via upsertMemory/getMemory/getMemories
2. LLM auto-tag extraction: extend extraction prompt to produce tags,
parse and validate in pipeline, create CMEM from extraction results
3. Session summary API: expose rollout_summary via new REST endpoints
GET /api/core-memory/sessions/summaries and sessions/:id/summary
4. Hook injection: increase SESSION_START_LIMIT to 1500, add Component 5
(Recent Sessions) to UnifiedContextBuilder with 300-char budget
5. Tag filtering: add getMemoriesByTags() with json_each() for safe
SQL matching, wire through MCP tool, CLI --tags flag, REST ?tags= param
6. Solidify compress mode: add --type compress to solidify.md with
getRecentMemories(), archiveMemories(), buildCompressionMetadata()
Security fixes: safeParseTags() for corrupt DB data, json_each() instead
of LIKE injection, ESCAPE clause for searchSessionsByKeyword, singleton
store in unified-context-builder.
When both prompt and target flags (--uncommitted/--base/--commit) are
provided together for codex review mode, warn user and drop the prompt
to avoid CLI argument conflict error.
Only log "Loaded X pending questions" when X > 0 to reduce noise
in CLI output. The message was appearing on every ccw command due to
module import chain, but is only meaningful for ccw view startup.
- Implemented Skill Hub page with tabs for remote, local, and installed skills.
- Added localization support for Chinese in skill-hub.json.
- Created API routes for fetching remote skills, listing local skills, and managing installed skills.
- Developed functionality for installing and uninstalling skills from both remote and local sources.
- Introduced caching mechanism for remote skills and handling updates for installed skills.
1. Race condition fix: Removed frontend direct lockSession call in
useOrchestratorExecution.ts - session locking now handled purely
via backend WebSocket broadcast (CLI_SESSION_LOCKED)
2. WebSocket handlers: Added handleSessionLockedMessage and
handleSessionUnlockedMessage to sessionManagerStore.ts
3. useWebSocket integration: Added case handlers for
CLI_SESSION_LOCKED and CLI_SESSION_UNLOCKED messages
4. API input validation: Added validation for sessionConfig,
stepTimeout, and errorStrategy in execute-in-session endpoint
5. Fixed wsBroadcast reference: Changed to broadcastToClients
from context
- Remove AgentList component from left sidebar
- Integrate orchestration plans display into ExecutionMonitorPanel
- Execution Monitor now shows both workflow executions and orchestration plans
- Cleaner sidebar with only session tree
- Add tag/name input field to CliConfigModal with auto-generation
- Auto-generate format: {tool}-{HHmmss} (e.g., gemini-143052)
- Add regenerate button for quick name changes
- Add i18n keys for new fields (en/zh)
Sessions are now grouped by tag in the sidebar for better organization.
- Replace dropdown menu with direct dialog button
- Remove unused state variables (selectedTool, launchMode, selectedShell)
- Update button icon to Plus and label to "New Session"
- Clean up i18n keys (remove unused tool/mode/shell options)
- Add confirmation dialog when closing pane with active session
- Add explicit "Close Session" button to terminate backend PTY
- Handle "session not found" scenario with user-friendly message
- Add i18n keys for new UI elements (en/zh)
- Implement tests for KeywordDetector including keyword detection, sanitization, and priority handling.
- Add tests for SessionStateService covering session validation, loading, saving, and state updates.
- Create tests for UserAbortDetector to validate user abort detection logic and pattern matching.
- Implemented `explorer` role for parallel codebase exploration using `cli-explore-agent`.
- Created `explore.md` command documentation detailing exploration strategy and execution steps.
- Established `synthesizer` role for integrating insights from explorations, analyses, and discussions.
- Developed `synthesize.md` command documentation outlining synthesis strategy and output format.
- Configured team settings in `team-config.json` to support new roles and pipeline modes.
- Added regression test for CodexLens bootstrap fallback to ensure robustness in error handling.
- Added new staged settings in config.py for coarse_k, lsp_depth, stage2_mode, and clustering strategy.
- Updated config-handlers.ts to handle new staged settings and map environment variables.
- Enhanced codexlens.json localization files for English and Chinese to include new staged settings.
- Modified astgrep_js_ts_processor.py to improve import handling for named imports.
- Updated JavaScript and TypeScript patterns to support new import formats.
- Added tests for staged settings loading and performance benchmarks for stage-2 expansion.
