Multifunctional Computing Platform (MCP) Server for Qwen Integration

Overview

The Multifunctional Computing Platform (MCP) Server is a robust backend service designed to support a variety of functionalities including file access, database connections, API integrations, and vector database access. This project is specifically tailored for integration with large language models like Qwen, providing a complete Docker deployment configuration and Qwen invocation examples.

Project Structure

mcp_server/
├── src/                    # Source code directory
│   ├── __init__.py         # Initialization module
│   ├── config.py           # Configuration management
│   ├── server.py           # Server main class
│   └── modules/            # Functional modules
│       ├── __init__.py     # Module registration
│       ├── file_module.py  # File access module
│       ├── database_module.py # Database connection module
│       ├── api_module.py   # API integration module
│       └── vector_module.py # Vector database module
├── docker/                 # Docker configuration
│   ├── Dockerfile          # Docker image configuration
│   └── docker-compose.yml  # Docker Compose configuration
├── examples/               # Example code
│   ├── qwen_client.py      # Qwen client library
│   └── qwen_example.py     # Qwen usage example
├── docs/                   # Documentation
│   ├── user_guide.md       # User guide
│   ├── api_docs.md         # API documentation
│   └── qwen_examples.md    # Qwen example instructions
├── test_server.py          # Server test script
├── test_qwen_client.py     # Qwen client test script
├── main.py                 # Main entry file
└── requirements.txt        # Dependency list

Features

• File Access: Upload, download, list, and delete files
• Database Connection: MongoDB integration, supporting CRUD operations on documents
• API Integration: Support for calling external API services
• Vector Database: Support for vector storage and similarity search
• Docker Deployment: Complete Docker configuration, supporting one-click deployment
• Qwen Integration: Provides Qwen client and examples for invoking the MCP server

Quick Start

Installation

1. Clone the project code:

git clone https://github.com/ningwenjie/mcp_server
cd mcp_server

1. Start the service using Docker Compose:

docker-compose -f docker/docker-compose.yml up -d

1. Verify that the service is running:

curl http://localhost:8000/health

Using the Qwen Client

from examples.qwen_client import QwenMCPClient

# Initialize the client
client = QwenMCPClient("http://localhost:8000")

# Upload a file
file_info = client.upload_file("example.txt")

# Store a vector
vector = [0.1, 0.2, 0.3] * 512  # 1536-dimensional vector
metadata = {"text": "This is an example text", "source": "Qwen"}
vector_info = client.store_vector("qwen_embeddings", vector, metadata)

# Search vectors
query_vector = [0.15, 0.25, 0.35] * 512
search_results = client.search_vector("qwen_embeddings", query_vector, top_k=3)

Documentation

For detailed documentation, please refer to:

• User Guide: Installation, configuration, and basic usage
• API Documentation: Detailed API reference
• Qwen Example Instructions: Qwen integration examples

Testing

Run the server tests:

python test_server.py

Run the Qwen client tests:

python test_qwen_client.py

Dependencies

• Python 3.10+
• FastAPI
• Uvicorn
• PyMongo
• FAISS
• Docker (for deployment)

For a detailed list of dependencies, please refer to requirements.txt.

License

MIT License