- Added token estimation and batching functionality in LiteLLMEmbedder to handle large text inputs efficiently.
- Updated embed method to support max_tokens_per_batch parameter for better API call management.
- Introduced new API routes for managing custom CLI endpoints, including GET, POST, PUT, and DELETE methods.
- Enhanced CLI history component to support source directory context for native session content.
- Improved error handling and logging in various components for better debugging and user feedback.
- Added internationalization support for new API endpoint features in the i18n module.
- Updated CodexLens CLI commands to allow for concurrent API calls with a max_workers option.
- Enhanced embedding manager to track model information and handle embeddings generation more robustly.
- Added entry points for CLI commands in the package configuration.
- Added JSON-based settings management in Config class for embedding and LLM configurations.
- Introduced methods to save and load settings from a JSON file.
- Updated BaseEmbedder and its subclasses to include max_tokens property for better token management.
- Enhanced chunking strategy to support recursive splitting of large symbols with improved overlap handling.
- Implemented comprehensive tests for recursive splitting and chunking behavior.
- Added CLI tools configuration management for better integration with external tools.
- Introduced a new command for compacting session memory into structured text for recovery.
- Updated embedding_manager.py to include backend parameter in model configuration.
- Modified model_manager.py to utilize cache_name for ONNX models.
- Refactored hybrid_search.py to improve embedder initialization based on backend type.
- Added backend column to vector_store.py for better model configuration management.
- Implemented migration for existing database to include backend information.
- Enhanced API settings implementation with comprehensive provider and endpoint management.
- Introduced LiteLLM integration guide detailing configuration and usage.
- Added examples for LiteLLM usage in TypeScript.
- Create ccw-litellm Python package with AbstractEmbedder and AbstractLLMClient interfaces
- Add BaseEmbedder abstraction and factory pattern to codex-lens for pluggable backends
- Implement API Settings dashboard page for provider credentials and custom endpoints
- Add REST API routes for CRUD operations on providers and endpoints
- Extend CLI with --model parameter for custom endpoint routing
- Integrate existing context-cache for @pattern file resolution
- Add provider model registry with predefined models per provider type
- Include i18n translations (en/zh) for all new UI elements
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Add context_cache MCP tool for caching files by @patterns:
- pattern-parser.ts: Parse @expressions using glob
- context-cache-store.ts: In-memory cache with TTL/LRU
- context-cache.ts: MCP tool with pack/read/status/release/cleanup
Simplify CLI cache options:
- --cache now uses comma-separated format instead of JSON
- Items starting with @ are patterns, others are text content
- Add --inject-mode option (none/full/progressive)
- Default: codex=full, gemini/qwen=none
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Introduced a new CodexLens Manager item in the dashboard for easier access.
- Implemented GPU management commands in the CLI, including listing available GPUs, selecting a specific GPU, and resetting to automatic detection.
- Enhanced the embedding generation process to utilize GPU resources more effectively, including batch size optimization for better performance.
- Updated the embedder to support device ID options for GPU selection, ensuring compatibility with DirectML and CUDA.
- Added detailed logging and error handling for GPU detection and selection processes.
- Updated package version to 6.2.9 and added comprehensive documentation for Codex Agent Execution Protocol.
- Added accelerator and providers fields to SemanticStatus interface.
- Updated checkSemanticStatus function to retrieve ONNX providers and accelerator type.
- Introduced detectGpuSupport function to identify available GPU modes (CUDA, DirectML).
- Modified installSemantic function to support GPU acceleration modes and clean up ONNX Runtime installations.
- Updated package requirements in PKG-INFO for semantic-gpu and semantic-directml extras.
- Added new source files for GPU support and enrichment functionalities.
- Updated tests to cover new features and ensure comprehensive testing.
- Updated the Chunker class to adjust the window movement logic, ensuring proper handling of overlap lines.
- Introduced a new smart search tool with features including intent classification, CodexLens integration, multi-backend search routing, and index status checking.
- Implemented various search modes (auto, hybrid, exact, ripgrep, priority) with detailed metadata and error handling.
- Added support for progress tracking during index initialization and enhanced output transformation based on user-defined modes.
- Included comprehensive documentation for usage and parameters in the smart search tool.
- Add SQLite table and CRUD methods for tracking update history
- Create freshness calculation service based on git file changes
- Add API endpoints for freshness data, marking updates, and history
- Display freshness badges in file tree (green/yellow/red indicators)
- Show freshness gauge and details in metadata panel
- Auto-mark files as updated after CLI sync
- Add English and Chinese i18n translations
Freshness algorithm: 100 - min((changedFilesCount / 20) * 100, 100)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Updated the dashboard template to hide the Code Graph Explorer feature.
- Enhanced the `executeCodexLens` function to use `exec` for better cross-platform compatibility and improved command execution.
- Changed the default `maxResults` and `limit` parameters in the smart search tool to 10 for better performance.
- Introduced a new `priority` search mode in the smart search tool, replacing the previous `parallel` mode, which now follows a fallback strategy: hybrid -> exact -> ripgrep.
- Optimized the embedding generation process in the embedding manager by batching operations and using a cached embedder instance to reduce model loading overhead.
- Implemented a thread-safe singleton pattern for the embedder to improve performance across multiple searches.
- Introduced styles for the help view including tab transitions, accordion animations, search highlighting, and responsive design.
- Implemented core memory styles with modal base styles, memory card designs, and knowledge graph visualization.
- Enhanced dark mode support across various components.
- Added loading states and empty state designs for better user experience.
- Updated command patterns across documentation and templates to reflect the new CLI syntax.
- Enhanced CLI tool implementation to support reading prompts from files and multi-line inputs.
- Modified core components and views to ensure compatibility with the new command structure.
- Adjusted help messages and internationalization strings to align with the updated command format.
- Improved error handling and user notifications in the CLI execution flow.
- Added core-memory.js and core-memory-graph.js for managing core memory views and visualizations.
- Introduced functions for viewing knowledge graphs and evolution history of memories.
- Implemented modal dialogs for creating, editing, and viewing memory details.
- Developed core-memory.ts for backend operations including list, import, export, and summary generation.
- Integrated Zod for parameter validation in core memory operations.
- Enhanced UI with dynamic rendering of memory cards and detailed views.
- MCP server outputs progress to stderr during smart_search init
- Progress format: [Progress] {percent}% - {message}
- Does not interfere with JSON-RPC protocol (stdout)
- Added executeInitWithProgress for external progress callback
- Added executeToolWithProgress to tools/index.ts
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Fixed API endpoint mismatches in mcp-manager.js to ensure global install/update buttons function correctly.
- Corrected undefined function references in mcp-manager.js for project installation.
- Refactored event handling to eliminate global scope pollution in mcp-manager.js.
- Added comprehensive debugging guide for MCP installation issues.
- Implemented a session path resolver to infer content types from filenames and paths, improving usability.
- Introduced tests for embeddings improvements in init and status commands to verify functionality.
This commit addresses multiple issues in session management and command documentation:
Session Management Fixes:
- Add auto-inference of location from type parameter in session.ts
- When --type lite-plan/lite-fix is specified, automatically set location accordingly
- Preserve explicit --location parameter when provided
- Update session-manager.ts to support type-based location inference
- Fix metadata filename selection (session-metadata.json vs workflow-session.json)
Command Documentation Fixes:
- Add missing --mode analysis parameter (3 locations):
* commands/memory/docs.md
* commands/workflow/lite-execute.md (2 instances)
- Add missing --mode write parameter (4 locations):
* commands/workflow/tools/task-generate-agent.md
- Remove non-existent subcommands (3 locations):
* commands/workflow/session/complete.md (manifest, project)
- Update session command syntax to use simplified format:
* Changed from 'ccw session manifest read' to 'test -f' checks
* Changed from 'ccw session project read' to 'test -f' checks
Documentation Updates:
- Update lite-plan.md and lite-fix.md to use --type parameter
- Update session/start.md to document lite-plan and lite-fix types
- Sync all fixes to skills/command-guide/reference directory (84 files)
All ccw command usage across the codebase is now consistent and correct.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Previously, embeddings were only generated for root directory files (1.6% coverage, 5/303 files).
This fix implements recursive processing across all subdirectory indexes, achieving 100% coverage
with 2,042 semantic chunks across all 303 files in 26 index databases.
Key improvements:
1. **Recursive embeddings generation** (embedding_manager.py):
- Add generate_embeddings_recursive() to process all _index.db files in directory tree
- Add get_embeddings_status() for comprehensive coverage statistics
- Add discover_all_index_dbs() helper for recursive file discovery
2. **Enhanced CLI commands** (commands.py):
- embeddings-generate: Add --recursive flag for full project coverage
- init: Use recursive generation by default for complete indexing
- status: Display embeddings coverage statistics with 50% threshold
3. **Smart search routing improvements** (smart-search.ts):
- Add 50% embeddings coverage threshold for hybrid mode routing
- Auto-fallback to exact mode when coverage insufficient
- Strip ANSI color codes from JSON output for correct parsing
- Add embeddings_coverage_percent to IndexStatus and SearchMetadata
- Provide clear warnings with actionable suggestions
4. **Documentation and analysis**:
- Add SMART_SEARCH_ANALYSIS.md with initial investigation
- Add SMART_SEARCH_CORRECTED_ANALYSIS.md revealing true extent of issue
- Add EMBEDDINGS_FIX_SUMMARY.md with complete fix summary
- Add check_embeddings.py script for coverage verification
Results:
- Coverage improved from 1.6% (5/303 files) to 100% (303/303 files) - 62.5x increase
- Semantic chunks increased from 10 to 2,042 - 204x increase
- All 26 subdirectory indexes now have embeddings vs just 1
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Introduced a detailed guide for intelligent tools selection strategy, including quick reference, tool specifications, prompt templates, and best practices for CLI execution.
- Established a coding philosophy document outlining core beliefs, simplicity principles, and guidelines for effective coding practices.
- Created context requirements documentation emphasizing the importance of understanding existing patterns and dependencies before implementation.
- Developed a file modification workflow detailing the use of edit_file and write_file MCP tools, along with priority logic for file reading and editing.
- Implemented CodexLens auto hybrid mode, enhancing the CLI with automatic vector embedding generation and default hybrid search mode based on embedding availability.
- Introduced `help-i18n.js` for managing translations in Chinese and English for the help view.
- Created `help.js` to render the help view, including command categories, workflow diagrams, and CodexLens quick-start.
- Implemented search functionality with debounce for command filtering.
- Added workflow diagram rendering with Cytoscape.js integration.
- Developed tests for write-file verification, ensuring proper handling of small and large JSON files.
- Implement tests for migration 005 to verify removal of deprecated fields in the database schema.
- Ensure that new databases are created with a clean schema.
- Validate that keywords are correctly extracted from the normalized file_keywords table.
- Test symbol insertion without deprecated fields and subdir operations without direct_files.
- Create a detailed search comparison test to evaluate vector search vs hybrid search performance.
- Add a script for reindexing projects to extract code relationships and verify GraphAnalyzer functionality.
- Include a test script to check TreeSitter parser availability and relationship extraction from sample files.
- Implemented tests for the QueryParser class, covering various identifier splitting methods (CamelCase, snake_case, kebab-case), OR expansion, and FTS5 operator preservation.
- Added parameterized tests to validate expected token outputs for different query formats.
- Created edge case tests to ensure robustness against unusual input scenarios.
- Developed tests for the Reciprocal Rank Fusion (RRF) algorithm, including score computation, weight handling, and result ranking across multiple sources.
- Included tests for normalization of BM25 scores and tagging search results with source metadata.
- Added a cleanup function to reset the state when navigating away from the graph explorer.
- Updated navigation logic to call the cleanup function before switching views.
- Improved internationalization by adding new translations for graph-related terms.
- Adjusted icon sizes for better UI consistency in the graph explorer.
- Implemented impact analysis button functionality in the graph explorer.
- Refactored CLI tool configuration to use updated model names.
- Enhanced CLI executor to handle prompts correctly for codex commands.
- Introduced code relationship storage for better visualization in the index tree.
- Added support for parsing Markdown and plain text files in the symbol parser.
- Updated tests to reflect changes in language detection logic.
- Added a new Storage Manager component to handle storage statistics, project cleanup, and configuration for CCW centralized storage.
- Introduced functions to calculate directory sizes, get project storage stats, and clean specific or all storage.
- Enhanced SQLiteStore with a public API for executing queries securely.
- Updated tests to utilize the new execute_query method and validate storage management functionalities.
- Improved performance by implementing connection pooling with idle timeout management in SQLiteStore.
- Added new fields (token_count, symbol_type) to the symbols table and adjusted related insertions.
- Enhanced error handling and logging for storage operations.
- Implemented unit tests for the Tokenizer class, covering various text inputs, edge cases, and fallback mechanisms.
- Created performance benchmarks comparing tiktoken and pure Python implementations for token counting.
- Developed extensive tests for TreeSitterSymbolParser across Python, JavaScript, and TypeScript, ensuring accurate symbol extraction and parsing.
- Added configuration documentation for MCP integration and custom prompts, enhancing usability and flexibility.
- Introduced a refactor script for GraphAnalyzer to streamline future improvements.
- Added main JavaScript functionality for CLAUDE.md management including file loading, rendering, and editing capabilities.
- Created a test HTML file to validate the functionality of the CLAUDE.md manager.
- Introduced CLI generation examples and documentation for rules creation via CLI.
- Enhanced error handling and notifications for file operations.
- Implement full coverage tests for Embedder model loading and embedding generation
- Add CRUD operations and caching tests for VectorStore
- Include cosine similarity computation tests
- Validate semantic search accuracy and relevance through various queries
- Establish performance benchmarks for embedding and search operations
- Ensure edge cases and error handling are covered
- Test thread safety and concurrent access scenarios
- Verify availability of semantic search dependencies
Fixes CLI tool detection failure after server restart.
The previous caching implementation cached negative results (errors,
timeouts) which caused persistent failures if the first check failed
for any transient reason (environment, timing, etc.).
Changes:
- Only cache positive tool availability results
- Don't cache errors or timeouts - they may be transient
- Subsequent checks will retry if a tool wasn't found
This ensures that once a tool is found, lookups are fast, but
transient failures don't persist in the cache.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Added `skills-manager.js` for managing Claude Code skills with functionalities for loading, displaying, and editing skills.
- Introduced a Notifier module in `notifier.ts` for CLI to server communication, enabling notifications for UI updates on data changes.
- Created comprehensive documentation for the Chain Search implementation, including usage examples and performance tips.
- Developed a test suite for the Chain Search engine, covering basic search, quick search, symbol search, and files-only search functionalities.
- Implemented insights history section in the dashboard with cards displaying analysis results.
- Added CSS styles for insights cards, detail panel, and empty state.
- Enhanced i18n support for insights-related strings in English and Chinese.
- Developed backend methods for saving, retrieving, and deleting insights in the CLI history store.
- Integrated insights loading and rendering logic in the memory view.
- Added functionality to refresh insights and handle detail view interactions.
