MCP Manager

A comprehensive toolkit for building, managing, and testing Model Context Protocol (MCP) servers and tools.

!MCP Manager

Overview

MCP Manager provides a streamlined interface to:

• Launch and monitor MCP servers locally
• Generate MCP tool code using LLMs (OpenAI, Mistral, or OpenRouter)
• Test and verify MCP tools
• Configure client applications to use your MCP servers

This application bridges the gap between MCP development and deployment, making it easier for developers to create, test, and integrate MCP tools into applications like Claude, Cursor, Continue, and other MCP-compatible clients.

Features

🚀 Server Management

• Create and launch MCP servers from templates
• Monitor server status and logs in real-time
• Terminate servers gracefully when done
• Scan and view port usage for running servers

🤖 LLM-Powered Tool Generation

• Generate MCP tool code using AI (OpenAI, Mistral, or OpenRouter)
• Describe tools in natural language and get production-ready code
• Auto-inject generated code into server templates

📊 Monitoring & Diagnostics

• Real-time server logs
• Port scanning and process monitoring
• Integration status checking

🔌 Client Setup Instructions

• Guided setup for Claude Desktop, Cursor, Continue, and Zed
• Configuration snippets and examples

Getting Started

Prerequisites

• Python 3.12+
• API keys for at least one LLM provider:
  • OpenAI
  • Mistral AI
  • OpenRouter (recommended, provides access to many models)

Installation

1. Clone the repository:

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

2. Set up the environment:

   python -m venv .venv
   source .venv/bin/activate  # On Windows: .venv\Scripts\activate
   uv pip install -e .
   

3. Configure your API keys:

   • Copy .env.example to .env (or create one)
   • Add your API keys:

     OPENAI_API_KEY=your_openai_key
     MISTRAL_API_KEY=your_mistral_key
     OPENROUTER_API_KEY=your_openrouter_key
     

Running the Application

Start the Streamlit app:

streamlit run streamlit_app.py

The application will open in your default web browser.

Usage Guide

Creating a New MCP Server

1. Navigate to the "Launch Server" tab
2. Enter a server name (e.g., "weather_service")
3. Toggle security options if needed
4. Click "Setup & Launch Server"
5. Your server will be created in the mcp_servers/ directory

Generating Tool Code

1. Go to the "Generate Tool" tab
2. Describe your tool functionality in natural language, e.g.:

   A tool that takes a city name and returns the current temperature using the OpenWeatherMap API.
   

3. Click "Generate Tool Code"
4. Review the generated code
5. Use it when launching a new server or copy it for your own use

Monitoring Servers

1. Visit the "Monitor Servers" tab
2. View running server status
3. Scan ports to check what's running
4. View server logs in real-time

Setting Up Client Applications

1. Go to the "Client Setup" tab
2. Select your client application (Claude, Cursor, etc.)
3. Follow the detailed instructions to configure the client

Project Structure

• streamlit_app.py - Main Streamlit application
• mcp_launcher.py - Server management functionality
• tool_generator.py - LLM-based code generation
• client_config_helper.py - Client configuration templates
• port_scanner.py - Network utilities for server monitoring
• utils/ - Support modules:
  • llm_client.py - Unified LLM client for code generation
  • logger.py - Logging utilities
• server_templates/ - MCP server templates

Contributing

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

License

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

Acknowledgments

• Anthropic for the Model Context Protocol
• FastMCP for the server framework
• Streamlit for the UI framework