Mcphub Trw

🚀 MCPHub Project Documentation

MCPHub is a tool that simplifies the integration of Model Context Protocol (MCP) servers into AI applications. It achieves this through the following four core components:

✨ Features

Core Components

1. Params Hub

   • Manages configurations read from the .mcphub.json configuration file.
   • Defines the MCP servers to be used and how to set them up.
   • Stores server parameters such as commands, arguments, and environment variables.

2. MCP Server Manager

   • Responsible for server installation and startup.
   • Supports two types of servers:
     • TypeScript-based servers (installed via npx).
     • Python-based servers (installed from GitHub via uv).
   • Manages server lifecycle and environment.

3. MCP Client

   • Establishes communication with the MCP server.
   • Interacts using standard input/output (stdio).
   • Handles two main operations:
     • list_tools: Discovers available server tools.
     • call_tool: Executes server tools.

4. Framework Adapter

   • Converts MCP tools into a specific framework's format.
   • Supports multiple AI frameworks:
     • OpenAI Agents
     • LangChain
     • Autogen

Workflow

1. Configuration and Setup

   • Params Hub reads the configuration file.
   • The server manager starts the required servers.
   • The servers start running and become available for use.

2. Communication

   • The MCP client connects to the server via stdio.
   • Tools are discovered and become available for use.
   • Requests and responses flow between the client and the server.

3. Integration

   • The framework adapter converts MCP tools into a specific framework's format.
   • AI applications use these tools through their preferred frameworks.
   • Tools are executed through the established communication channel.

This architecture provides a seamless way to integrate MCP functionality into any AI application while maintaining clean separation of concerns and framework flexibility.

📦 Installation

Before you start, make sure you have Python 3.10 or higher and poetry installed. Then run the following command:

poetry install

Edit the .mcphub.json configuration file to specify the MCP servers and parameters to use.

💻 Usage Examples

Basic Usage

Get a list of tools from the MCP server and execute a tool:

from mcphub import MCPHub

async def basic_usage():
    hub = MCPHub()
    
    # Get all servers
    servers = hub.list_servers()
    
    # Get tools for a specific MCP server
    tools = await hub.list_tools(mcp_name="sequential-thinking-mcp")
    
    # Print tool information
    for tool in tools:
        print(f"Tool Name: {tool.name}")
        print(f"Description: {tool.description}")
        print(f"Parameters: {tool.parameters}")
        print("---")
    
    # Tools can:
    # - Be cached using cache_tools_list=True for better performance.
    # - Be automatically converted to a specific framework's format.
    # - Be used directly with AI frameworks through the adapter.

asyncio.run(basic_usage())

Advanced Usage

Integrate MCP tools into popular frameworks such as OpenAI, LangChain, and Autogen:

from mcphub import MCPHub, get_openai_proxy, get_langchain_proxy, get_autogen_proxy

async def advanced_usage():
    hub = MCPHub()
    
    # Get the OpenAI proxy
    openai_proxy = await get_openai_proxy(hub)
    
    # Get the LangChain proxy
    langchain_proxy = await get_langchain_proxy(hub)
    
    # Get the Autogen proxy
    autogen_proxy = await get_autogen_proxy(hub)

asyncio.run(advanced_usage())

🤝 Contribution Guidelines

How to Contribute Code

1. Fork this repository.
2. Create a new feature branch in your local copy: git checkout -b feature/your-feature.
3. Commit your changes: git commit -m "Add new feature".
4. Push to the remote repository: git push origin feature/your-feature.
5. Create a Pull Request.

How to Report Issues

Create a new issue on the Issues page and describe the problem you encountered and the steps to reproduce it.

⚙️ Development Environment Setup

Install Dependencies

Use poetry to manage dependencies:

poetry install

Run Tests

Run unit tests:

poetry run pytest tests/

Perform Format Checks

Run code format checks (assuming you have black and flake8 installed):

poetry run black .
poetry run flake8 .

📄 License

The MCPHub project is open-source under the MIT License.

Note: The above content is an example. Please adjust it according to the actual project requirements.