Acoste616/MCP-SERVER

MCP Server

A local Model Context Protocol server that stores project context (bookmarks, files, notes) and exposes a REST + WebSocket API so multiple AI models (OpenAI GPT-4o, Anthropic Claude 3, local LM Studio) can share the same information.

Features

• Store and manage project context (files, bookmarks, notes)
• Integrate with multiple AI model providers:
  • OpenAI (GPT-4o, GPT-4o-mini)
  • Anthropic Claude
  • LM Studio (local models)
• WebSocket API for real-time updates
• Token usage tracking and budget management
• Async/await architecture with FastAPI

Project Structure

mcp-server/
├── backend/               # Python backend code
│   ├── app/               # Application code
│   │   ├── api/           # API endpoints
│   │   ├── core/          # Core functionality
│   │   ├── models/        # SQLModel definitions
│   │   └── services/      # Business logic services
│   ├── tests/             # Test suite
│   └── alembic/           # Database migrations
├── docker/                # Docker configuration
└── docs/                  # Documentation

Prerequisites

• Python 3.12+
• Poetry for dependency management
• Optional: LM Studio for local models

Installation

1. Clone the repository:

   git clone https://github.com/yourusername/mcp-server.git
   cd mcp-server
   

2. Install dependencies with Poetry:

   poetry install
   

3. Create a .env file with your configuration:

   OPENAI_API_KEY=your_openai_key
   ANTHROPIC_API_KEY=your_anthropic_key
   LM_STUDIO_BASE_URL=http://localhost:1234/v1
   

Running the Server

Start the development server:

cd backend
poetry run uvicorn app.main:app --host 0.0.0.0 --port 8000 --reload

The API will be available at http://localhost:8000

API Documentation

Once the server is running, you can access the interactive API documentation at:

• Swagger UI: http://localhost:8000/docs
• ReDoc: http://localhost:8000/redoc

License

MIT

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.