CGM MCP Server

# Clone the repository git clone https://github.com/your-org/cgm-mcp.git cd cgm-mcp # Install dependencies pip install -r requirements.txt # Or install in development mode pip install -e .

# No additional setup needed! # MPS (Metal Performance Shaders) is automatically detected and enabled pip install torch torchvision torchaudio # Usually already installed

# Install CUDA-enabled PyTorch pip install torch --index-url https://download.pytorch.org/whl/cu118 # Optional: Enhanced GPU features pip install cupy-cuda11x # For CUDA 11.x # or pip install cupy-cuda12x # For CUDA 12.x

# Linux (ROCm) pip install torch --index-url https://download.pytorch.org/whl/rocm5.6 # Windows (DirectML) pip install torch-directml

# Run setup script ./scripts/setup.sh # Copy example environment file cp .env.example .env

Option A: Use Cloud Models (OpenAI/Anthropic)

# Edit .env with your API keys export CGM_LLM_PROVIDER=openai export CGM_LLM_API_KEY=your-openai-api-key export CGM_LLM_MODEL=gpt-4

Option B: Use Local Models (Recommended)

# Install and start Ollama curl -fsSL https://ollama.ai/install.sh | sh ollama serve # Download recommended model ollama pull deepseek-coder:6.7b # Start with local model ./scripts/start_local.sh --provider ollama --model deepseek-coder:6.7b

# Download and start LM Studio # Load deepseek-coder-6.7b-instruct model # Start local server # Start CGM with LM Studio ./scripts/start_local.sh --provider lmstudio

# Start MCP server (cloud models) python main.py # Start with local models ./scripts/start_local.sh # Or with custom config python main.py --config config.local.json --log-level DEBUG

# Run example usage python examples/example_usage.py # Check GPU acceleration status python check_gpu_dependencies.py

# Run the GPU dependency checker python check_gpu_dependencies.py

🎉 OPTIMAL: Apple Silicon GPU acceleration is active! • MPS backend enabled • No additional dependencies needed • CuPy warnings can be ignored

- TTL Cache: 1-hour expiration for analysis results
- LRU Cache: 500 most recent files kept in memory
- AST Cache: 200 parsed syntax trees cached
- Embedding Cache: GPU-accelerated similarity vectors

- Real-time Monitoring: Track GPU and system memory usage
- Automatic Cleanup: Clear caches when memory usage > 80%
- Unified Memory: Apple Silicon's shared CPU/GPU memory
- Memory Pools: Efficient GPU memory allocation

- Async File I/O: Non-blocking file operations
- Batch Processing: Process multiple files simultaneously
- Semaphore Control: Limit concurrent operations (default: 10)
- GPU Queuing: Intelligent GPU task scheduling

# GPU Configuration export CGM_USE_GPU=true # Enable GPU acceleration export CGM_GPU_BATCH_SIZE=1024 # Batch size for GPU operations export CGM_SIMILARITY_THRESHOLD=0.1 # Entity similarity threshold export CGM_CACHE_EMBEDDINGS=true # Cache embedding vectors # Memory Management export CGM_MAX_CACHE_SIZE=500 # Maximum cached files export CGM_MEMORY_CLEANUP_THRESHOLD=80 # Memory cleanup trigger (%) export CGM_GPU_MEMORY_FRACTION=0.8 # GPU memory usage limit

{ "gpu": { "use_gpu": true, "batch_size": 1024, "max_sequence_length": 512, "similarity_threshold": 0.1, "cache_embeddings": true, "gpu_memory_fraction": 0.8 }, "performance": { "max_concurrent_files": 10, "cache_ttl_seconds": 3600, "max_file_cache_size": 500, "memory_cleanup_threshold": 80 } }

{ "llm": { "provider": "openai", "model": "gpt-4", "temperature": 0.1, "max_tokens": 4000 } }

{ "llm": { "provider": "ollama", "model": "deepseek-coder:6.7b", "api_base": "http://localhost:11434", "temperature": 0.1, "max_tokens": 4000 }, "graph": { "max_nodes": 5000, "max_edges": 25000, "cache_enabled": true }, "server": { "log_level": "INFO", "max_concurrent_tasks": 3 } }

The server provides the following MCP tools:

Analyze repository structure and extract code entities with GPU acceleration.

- repository_path: Path to the repository
- query: Search query for relevant code
- analysis_scope: Scope of analysis (full,focused,minimal)
- max_files: Maximum number of files to analyze

Get detailed file content and analysis with concurrent processing.

- repository_path: Path to the repository
- file_paths: List of file paths to analyze

Find code entities related to a specific entity using GPU-accelerated similarity matching.

- repository_path: Path to the repository
- entity_name: Name of the entity to find relations for
- relation_types: Types of relations to include (optional)

Extract structured context for external model consumption.

- repository_path: Path to the repository
- query: Query for context extraction
- format: Output format (structured,markdown,prompt)

Process a repository issue using the CGM framework.

- task_type: Type of task (issue_resolution,code_analysis,bug_fixing,feature_implementation)
- repository_name: Name of the repository
- issue_description: Description of the issue
- repository_context: Optional repository context

{ "task_type": "issue_resolution", "repository_name": "my-project", "issue_description": "Authentication fails with special characters in password", "repository_context": { "path": "/path/to/repository", "language": "Python", "framework": "Django" } }

- cgm://health: Server health information
- cgm://tasks: List of active tasks

- cgm://cache: Cache statistics and hit/miss ratios
- cgm://performance: Server performance and memory usage metrics
- cgm://gpu: GPU acceleration status and memory usage

# Check GPU status curl "cgm://gpu" # Monitor cache performance curl "cgm://cache" # View performance metrics curl "cgm://performance"

CGM MCP Server- Bringing graph-integrated code intelligence to your development workflow! 🚀

This is a web browser that enables your coding agent, such as Claude Code, to visit websites on your behalf and assist you in identifying bugs or creating UI test cases.

Local-first knowledge graph for developers. Watches project files, extracts entities and relationships via LLMs, and lets you query across projects with natural language and source citations.

An MCP server that indexes local code into a graph database to provide context to AI assistants.

Graph-powered code intelligence MCP server with semantic search, knowledge graph, and dependency analysis for Claude Code, Cursor, and Copilot.

Graph-based long-term memory skill for AI (LLM) coding agents — faster context, fewer tokens, safer refactors

Developer context continuity system that builds a temporal knowledge graph of your codebase — modules, symbols, decisions, and open problems — and serves it to AI coding agents via 12 MCP tools, so every agent session starts knowing your architecture without manual context pasting.

Memtrace gives AI coding agents structural memory — your codebase as a live knowledge graph so agents stop re-deriving code structure from scratch and start reasoning from fact.

MCP bridge that lets Claude Code delegate heavy tasks to the Antigravity CLI (agy) — purpose-built tools, model routing with fallback, session continuity, and output truncation to save Claude's context and tokens.