Model Context Protocol Server

A FastAPI-based server that implements the Model Context Protocol to provide relevant information to AI models when processing user queries.

Environment Configuration

The server can be configured using environment variables. Create a .env file in the root directory with the following variables:

Deployment

Local Development

1. Create and activate a virtual environment:

2. Install dependencies:

3. Run the server:

Production Deployment

1. Set up environment variables for production:

2. Run the server:

Docker Deployment

1. Build the Docker image:

2. Run the container:

API Documentation

When running in development mode (DEBUG=true), API documentation is available at:

• Swagger UI: http://your-server:8000/api/v1/docs

• ReDoc: http://your-server:8000/api/v1/redoc

• OpenAPI JSON: http://your-server:8000/api/v1/openapi.json

Security Considerations

1. In production:

   • Set DEBUG=false to disable API documentation

   • Use HTTPS

   • Configure proper authentication

   • Use secure database credentials

   • Set appropriate CORS policies

2. For Ollama server:

   • Ensure Ollama server is properly secured

   • Use internal network for communication if possible

   • Consider using API keys or other authentication methods

Monitoring and Logging

The server includes built-in logging with different levels based on the environment:

• Development: Debug level logging

• Production: Info level logging

Logs can be configured to output to files or external logging services.

Features

• Intelligent query routing based on query analysis

• Support for multiple data sources (Database, GraphQL, REST)

• Integration with Ollama models (Mistral, Qwen, Llama2)

• Environment-aware configuration (Development/Production)

• Comprehensive logging and error handling

• Health check endpoints

• Mock data support for development

Prerequisites

• Python 3.8+

• Ollama installed and running locally

• Required Ollama models:

  • mistral

  • qwen

  • llama2

Installation

1. Clone the repository:

2. Create and activate a virtual environment:

3. Install dependencies:

4. Create a .env file:

5. Update the .env file with your configuration:

Running the Server

1. Start Ollama (if not already running):

2. Start the MCP server:

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

API Endpoints

Get Context

List Available Models

Health Check

Project Structure

Development

Adding New Providers

1. Create a new provider class in the appropriate directory

2. Implement the required interface methods

3. Register the provider in the factory

Adding New Models

1. Add the model to the AVAILABLE_MODELS dictionary in model_providers/ollama.py

2. Update the model validation logic if needed

Contributing

1. Fork the repository

2. Create a feature branch

3. Commit your changes

4. Push to the branch

5. Create a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.