shuvo881/MCP-Server-Client

MCP Server

A Model Context Protocol (MCP) server implementation that provides a stateful server for handling various tools and functionalities.

Introduction

This project implements a Model Context Protocol server using FastMCP, designed to maintain session state and provide various tools through a streamable HTTP transport layer. The server includes a comprehensive logging system and modular tool registration capabilities.

Prerequisites

• Python 3.13 or higher
• uv (Python virtual environment manager)

Installation

1. Clone the repository:

   git clone <repository-url>
   cd MCPServer
   

2. Install uv if you haven't already:

   • On Mac or Linux

     curl -LsSf https://astral.sh/uv/install.sh | sh
     

   • Or on Windows using PowerShell:

     powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
     

3. Create a virtual environment and install dependencies in one command:

   uv sync
   uv lock
   

4. Activate the virtual environment:

   • Windows (PowerShell):

     .\.venv\Scripts\activate
     

   • Linux/macOS:

     source .venv/bin/activate
     

   Note: uv provides faster package installation and better dependency resolution than traditional pip.

Project Structure

MCPServer/
├── src/
│   ├── __init__.py
│   ├── __main__.py
│   ├── server.py         # Main server implementation
│   ├── database/         # Database related modules
│   ├── resources/        # Resource files
│   ├── tools/           # MCP tools implementation
│   └── utils/           # Utility functions including logging
├── tests/               # Test files
├── logs/               # Log files directory
├── docs/               # Documentation
├── requirements.txt    # Project dependencies
└── README.md

Running the Server

1. Ensure your virtual environment is activated

2. Run the server from the project root:

   uv run src
   

The server will start with the following configurations:

• Transport: streamable-http
• Logging: Enabled (both console and file)
• Log file location: logs/mcp_server.log

Logging

The server uses a comprehensive logging system that:

• Logs to both console and file
• Rotates log files (max 10MB per file)
• Maintains up to 5 backup log files
• Includes timestamps, log levels, and module information

You can find the logs in:

• Console output (basic information)
• logs/mcp_server.log (detailed logs)

Development

Adding New Tools

1. Create your tool in the src/tools directory
2. Register your tool in the tools registry
3. Import and register in the main server

Example:

from utils import setup_logger

logger = setup_logger(__name__)

def my_new_tool():
    logger.info("Executing new tool...")
    # Tool implementation

Testing

Run the tests using:

python -m pytest tests/

Troubleshooting

Common issues and solutions:

1. Server won't start:

   • Check if the port is available
   • Ensure all dependencies are installed
   • Check the logs for detailed error messages

2. Logging issues:

   • Verify write permissions in the logs directory
   • Check available disk space

Contributing

...

License

[Add your license information here]

Support

For support and questions, please:

• Check the documentation in the docs directory
• Submit issues through the issue tracker
• Contact the development team