MCP Server
A Model Context Protocol (MCP) server implementation with LLM integration and chat memory capabilities.
Features
- MCP Server: Full Model Context Protocol server implementation
- LLM Integration: Support for OpenAI and Anthropic models
- Chat Memory: Persistent conversation storage and retrieval
- Tool System: Extensible tool framework for various operations
Installation
- Clone this repository:
git clone <repository-url>
cd MCP
- Install dependencies:
pip install -r requirements.txt
Or using the development environment:
pip install -e .[dev]
Configuration
Set up your API keys as environment variables:
export OPENAI_API_KEY="your-openai-api-key"
export ANTHROPIC_API_KEY="your-anthropic-api-key"
Or create a .env file:
OPENAI_API_KEY=your-openai-api-key
ANTHROPIC_API_KEY=your-anthropic-api-key
Usage
Running the MCP Server
Start the server using the command line:
python -m mcp
Or run directly:
python mcp.py
Available Tools
The server provides the following tools:
Echo Tool
Simple echo functionality for testing.
{
"name": "echo",
"arguments": {
"text": "Hello, world!"
}
}
Chat Memory Tools
Store Memory
{
"name": "store_memory",
"arguments": {
"conversation_id": "conv_123",
"content": "User preferences: dark mode enabled",
"metadata": {"type": "preference"}
}
}
Get Memory
{
"name": "get_memory",
"arguments": {
"conversation_id": "conv_123"
}
}
LLM Chat Tool
{
"name": "llm_chat",
"arguments": {
"message": "What is the capital of France?",
"model": "gpt-3.5-turbo"
}
}
Supported Models
OpenAI Models:
- gpt-3.5-turbo
- gpt-4
- gpt-4-turbo
- gpt-4o
Anthropic Models:
- claude-3-haiku-20240307
- claude-3-sonnet-20240229
- claude-3-opus-20240229
Development
Running Tests
pytest
Code Formatting
black .
isort .
Type Checking
mypy .
Architecture
Components
- mcp.py: Main MCP server implementation and tool registration
- llmintegrationsystem.py: LLM provider integration and chat completions
- chatmemorysystem.py: Persistent conversation storage with SQLite
Database Schema
The chat memory system uses SQLite with two main tables:
memories: Individual conversation messages and metadataconversation_summaries: Conversation overviews and statistics
Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests for new functionality
- Submit a pull request
License
MIT License - see LICENSE file for details.
Troubleshooting
Common Issues
API Key Errors Ensure your API keys are properly set in environment variables.
Database Permissions
The server creates a chat_memory.db file in the current directory. Ensure write permissions.
Port Conflicts The MCP server uses stdio communication by default. No port configuration needed.
Logging
Enable debug logging:
PYTHONPATH=. python -c "import logging; logging.basicConfig(level=logging.DEBUG); import mcp; mcp.main()"