Smart Code Search MCP Server

# Changelog All notable changes to SCS-MCP will be documented in this file. The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/), and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html). ## [Unreleased] ### Added - **Media Server Documentation** - Comprehensive `docs/MEDIA_SERVER.md` explaining screenshot and recording features - Complete voice commands reference in `docs/voice-assistant/COMMANDS.md` - Quick start guide `docs/voice-assistant/MEDIA_QUICKSTART.md` - Updated feature documentation with media capture details - **Testing Infrastructure** - Active `tests/` directory with unit and integration tests - `pytest.ini` configuration for test discovery and markers - `requirements-dev.txt` for development dependencies - Moved key tests from dev-archive to active test suite - **Graph Visualization** - New `generate_dependency_graph` tool for visualizing code dependencies - Support for DOT (Graphviz), Mermaid, and JSON output formats - Graph types: imports, function calls, class inheritance - Circular dependency detection - No external dependencies required (text-based formats) - **Documentation Improvements** - `CREDITS.md` with full attribution for all dependencies - `NOTICE` file for Apache 2.0 license compliance - `THIRD_PARTY_LICENSES.md` with complete license texts - `MANIFEST.in` for proper package distribution - README.md auto-generation in `.claude-symbols/` directory ### Fixed - Documentation files now properly indexed and searchable - Agent parameter mismatches in orchestration flows - Username attribution updated throughout project ### Changed - Enhanced `.env.example` with comprehensive documentation (284 lines) - Improved indexing to include documentation files (`.md`, `.json`, `.yaml`, etc.) ## [1.0.0] - 2024-01-08 ### 🎉 Initial Release #### Added - **Core Features** - Semantic code search using sentence transformers - Hybrid search combining semantic and keyword matching - Multi-language support (Python, JavaScript, TypeScript, Java, Go, Rust, Ruby) - Thread-safe SQLite database operations - Intelligent caching system - **Voice Assistant** - Web-based voice interface with real-time recognition - Natural language command processing - Media history for screenshots and recordings - VS Code extension for editor integration - ElevenLabs TTS integration support - **Code Analysis Tools** - Instant code review with best practices checking - Complexity analysis (cyclomatic and cognitive) - Test gap detection and suggestions - Security vulnerability scanning - Dependency analysis and circular detection - **Git Integration** - Search through commit history - Analyze code changes over time - Find related commits by code or issue - **Model Information Tools** - List available AI models and capabilities - Compare model specifications - Suggest optimal models for tasks - Estimate token usage and costs - **Orchestration Framework** - Technical debt analysis and prioritization - Refactoring coordination - Migration planning - Quality assessment - Performance optimization #### Performance - Search response time < 100ms for most queries - Support for codebases with 100,000+ files - Efficient incremental indexing - Memory usage optimized to < 500MB #### Documentation - Comprehensive API documentation - Detailed installation guide - Architecture overview - Contributing guidelines ### Known Issues - Voice recognition requires Chrome/Edge browser - Some Tree-sitter parsers may need manual compilation on certain platforms - Browser MCP integration requires Playwright installation ### Contributors - Initial implementation by SCS-MCP team - Special thanks to the Claude Desktop and MCP communities --- ## [Unreleased] ### Planned Features - Cloud deployment support (AWS, GCP, Azure) - Enhanced ML models (CodeBERT integration) - Real-time collaborative features - IDE plugins for IntelliJ and Sublime Text - Advanced refactoring automation ### Under Consideration - GraphQL API endpoint - Kubernetes operator - Multi-tenant support - Custom model training --- ## Version History - **1.0.0** (2024-01-08) - Initial release with full feature set - **0.9.0** (2023-12-15) - Beta release with voice assistant - **0.8.0** (2023-11-20) - Added orchestration framework - **0.7.0** (2023-10-30) - Git integration complete - **0.6.0** (2023-10-01) - Multi-language support - **0.5.0** (2023-09-15) - Code analysis tools - **0.4.0** (2023-08-20) - Database optimization - **0.3.0** (2023-07-25) - Hybrid search implementation - **0.2.0** (2023-06-30) - Basic semantic search - **0.1.0** (2023-06-01) - Initial prototype --- ## Upgrade Guide ### From 0.x to 1.0.0 1. **Database Migration** ```bash python scripts/migrate_db.py --from 0.9 --to 1.0 ``` 2. **Configuration Changes** - Add `ENABLE_VOICE=true` to `.env` for voice features - Update `MODEL_INFO_ENABLED=true` for model tools 3. **API Changes** - `search_code()` renamed to `search()` - New required parameter `threshold` in similarity searches 4. **Breaking Changes** - Python 3.7 no longer supported (minimum 3.8) - Changed embedding model to all-MiniLM-L6-v2 --- ## Security ### Reporting Vulnerabilities Please report security vulnerabilities to security@scs-mcp.org ### Security Updates - **1.0.0**: Input sanitization for all user queries - **0.9.5**: Fixed SQL injection vulnerability in search - **0.8.3**: Added path traversal protection --- ## Deprecation Notice The following features will be removed in v2.0.0: - Legacy JSON-RPC 1.0 support - Python 3.8 support (moving to 3.9+) - Deprecated `find_code` tool (use `search` instead)