Skip to main content
Glama

Smart Code Search MCP Server

FEATURES.md7.63 kB
# SCS-MCP Features Comprehensive documentation of all features available in the Smart Code Search MCP Server. ## Table of Contents - [Core Search Capabilities](#core-search-capabilities) - [Code Analysis](#code-analysis) - [Graph Visualization](#graph-visualization) - [Voice Assistant](#voice-assistant) - [Model Management](#model-management) - [Orchestration Framework](#orchestration-framework) - [Git Integration](#git-integration) ## Core Search Capabilities ### Semantic Search - **Technology**: Sentence transformers with all-MiniLM-L6-v2 model - **Features**: - Natural language queries - Code similarity matching - Context-aware ranking - Multi-language support ### Hybrid Search - Combines semantic and keyword matching - Weighted scoring system - Metadata filtering (TODOs, comments, types) ### Symbol Analysis - Deep analysis of functions, classes, and variables - Usage tracking across codebase - Test coverage identification - Call graph generation ## Code Analysis ### Instant Code Review Provides immediate feedback on code quality: - Best practices violations - Security issues - Performance problems - Style inconsistencies - Complexity metrics ### Complexity Analysis - **Cyclomatic Complexity**: Measures decision points - **Cognitive Complexity**: Measures mental effort to understand - **Halstead Metrics**: Volume, difficulty, effort - **Maintainability Index**: Overall code health score ### Test Gap Detection - Identifies untested functions and classes - Suggests test cases based on code patterns - Coverage analysis without running tests - Test file association ### Import Optimization - Detects unused imports - Identifies missing imports - Suggests import reorganization - Circular dependency detection ## Graph Visualization ### Dependency Graphs Generate visual representations of code dependencies in multiple formats: #### Output Formats 1. **DOT (Graphviz)** - Industry-standard graph description language - Can be rendered with Graphviz tools - Suitable for documentation 2. **Mermaid** - Renders directly in GitHub/GitLab markdown - No external tools required - Interactive in supporting viewers 3. **JSON** - Structured data for custom tools - Includes metadata and metrics - Suitable for programmatic analysis #### Graph Types - **Import Graph**: File and module dependencies - **Call Graph**: Function call relationships - **Inheritance Tree**: Class hierarchy visualization #### Features - Circular dependency detection - Node clustering by module - Edge labeling with relationship types - Configurable depth and filtering ### Example Usage ```json { "tool": "generate_dependency_graph", "arguments": { "output_format": "mermaid", "graph_type": "imports", "file_pattern": "src/**/*.py", "detect_cycles": true } } ``` ## Voice Assistant ### Web Interface - Real-time speech recognition - Natural language processing - Visual feedback and waveforms - Media history with screenshots ### Voice Commands - **Search**: "Find authentication code" - **Review**: "Review the current function" - **Navigate**: "Go to the user model" - **Analyze**: "What does this code do?" - **Generate**: "Create a test for this function" ### Media Capture - **Screenshots**: Voice-triggered or manual capture with annotation tools - **Screen Recording**: Full session recording with WebM format - **Automatic Capture**: Context-aware screenshots during debugging/review - **Media Management**: Gallery view with search, filter, and export - **Annotations**: Drawing tools, text overlay, and highlighting - **Storage**: Organized file system with SQLite metadata - See [Media Server Documentation](MEDIA_SERVER.md) for complete details ### VS Code Integration - Extension for editor context - Current file awareness - Selection handling - Quick actions ## Model Management ### Multi-Model Support Track and compare different AI models: - **Capabilities Matrix**: What each model can do - **Cost Analysis**: Token pricing and estimates - **Performance Metrics**: Speed and quality comparisons - **Context Windows**: Token limits per model ### Model Recommendation - Task-based model selection - Cost optimization suggestions - Quality vs. speed trade-offs - Fallback strategies ### Dynamic Model Loading - On-demand model switching - Memory management - Cache optimization - Parallel processing support ## Orchestration Framework ### Multi-Agent Coordination Coordinate multiple specialized agents for complex tasks: #### Available Orchestrations 1. **Technical Debt Analysis** - Code smells detection - Complexity assessment - Refactoring prioritization - Migration planning 2. **Instant Review** - Best practices check - Security scan - Performance analysis - Documentation review 3. **Test Gap Analysis** - Coverage calculation - Test case generation - Edge case identification - Integration test planning 4. **Import Optimization** - Dependency cleanup - Module reorganization - Circular dependency resolution - Package structure optimization ### Context Management - Large file handling with chunking - Context window optimization - Relevant snippet extraction - Cross-file analysis ## Git Integration ### History Search - Search through commit messages - Find code changes over time - Author-based filtering - Date range queries ### Blame Analysis - Line-by-line authorship - Change frequency tracking - Hot spot identification - Contributor statistics ### Diff Analysis - Smart diff interpretation - Change impact assessment - Breaking change detection - Migration path suggestions ## Performance Features ### Caching System - Embedding cache for repeated queries - Result caching with TTL - Git history caching - Incremental indexing ### Thread Safety - Concurrent request handling - Database connection pooling - Lock-free operations where possible - Graceful degradation ### Scalability - Handles projects with 100k+ files - Efficient memory usage - Lazy loading strategies - Background indexing ## Security Features ### Code Scanning - Basic vulnerability detection - Hardcoded credential detection - SQL injection patterns - XSS vulnerability patterns ### Access Control - API key authentication (optional) - Rate limiting support - Audit logging - Secure configuration ## Configuration ### Customization Options - Embedding model selection - Index update intervals - Cache sizes and TTLs - Performance tuning ### Language Support Currently supported languages: - Python (.py) - JavaScript (.js, .jsx, .mjs) - TypeScript (.ts, .tsx) - Java (.java) - Go (.go) - Rust (.rs) - Ruby (.rb) - C/C++ (.c, .cpp, .h, .hpp) ### File Type Support - Source code files - Configuration files (JSON, YAML, TOML) - Documentation (Markdown, RST) - Shell scripts - Docker files ## Integration Points ### MCP Protocol Full compliance with Model Context Protocol: - Tool discovery - Async execution - Error handling - Progress reporting ### IDE Integration - VS Code extension - Command-line interface - REST API (planned) - Language server protocol (planned) ### CI/CD Integration - GitHub Actions support - Pre-commit hooks - Quality gates - Automated reviews ## Future Roadmap ### Planned Features - [ ] Cloud deployment support - [ ] Team collaboration features - [ ] Custom rule definitions - [ ] Machine learning-based suggestions - [ ] Real-time collaboration - [ ] Plugin system - [ ] Web dashboard - [ ] Mobile app support ### Under Consideration - GraphQL API - Kubernetes operator - Terraform provider - Browser extension - Slack/Discord integration

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/stevenjjobson/scs-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server