Aisdk MCP Bridge

🚀 AISDK-MCP Communication Bridge

AISDK-MCP is a bridge tool designed to connect AI SDKs with MCP (Machine Learning Platform). It enables you to leverage the powerful capabilities of MCP directly within your AI applications.

✨ Features

• Support for multiple MCP services.
• Easy to configure and extend.
• Built - in error handling and logging.
• High - performance communication.

🚀 Quick Start

📦 Installation

npm install aisdk-mcp

💻 Usage Examples

Basic Usage

const { AISDKMCP } = require('aisdk-mcp');

async function main() {
    // Initialize the MCP bridge
    const mcpBridge = new AISDKMCP({
        configPath: 'config/mcp-config.json', // Configuration file path
        debug: true // Enable debug logging
    });

    // Get the MCP toolset
    const tools = await mcpBridge.getMCPTools({ serverName: 'twitter-mcp' });

    // Execute a specific function
    const result = await executeMCPFunction('twitter-mcp', 'postTweet', {
        text: 'Hello from MCP!'
    });
}

main().catch(console.error);

📚 Documentation

Configuration

Configuration File Format

{
  "mcpServers": {
    "twitter-mcp": {
      "command": "node",
      "args": ["server.js"],
      "env": {
        "PORT": "3000"
      },
      "mode": "stdio", // or 'sse'
      "sseOptions": {
        "endpoint": "ws://localhost:3000/ws",
        "headers": {},
        "reconnectTimeout": 5000
      }
    },
    "fire-crawl": {
      "command": "python",
      "args": ["script.py"],
      "env": {
        "API_KEY": "your_api_key"
      }
    }
  }
}

Core Interfaces

getMCPTools(options)

Retrieve AI tools compatible with MCP.

interface ToolOptions {
  debug?: boolean;
  serverName?: string;
}

async function getMCPTools(options: ToolOptions): Promise<ToolSet>;

executeMCPFunction(serverName, functionName, args)

Execute a function directly on a specific MCP server.

async function executeMCPFunction(
    serverName: string,
    functionName: string,
    args: Record<string, unknown>
): Promise<MCPToolResult>;

Error Handling

The bridge includes comprehensive error handling for:

• Server initialization failures.
• Communication errors.
• Tool execution errors.
• Configuration issues.
• Server connection problems.

Logging

Debug Logging

Enable detailed debug logging by setting environment variables:

# Enable all debug logs
DEBUG=* npm start

# Enable MCP debug logs
DEBUG=mcp npm start

# Enable all MCP namespace logs
DEBUG=mcp:* npm start

Log Types

The logging system supports the following types:

• Operation logs.
• Error logs.
• Warning logs.
• Information logs.

Log Files

By default, logs are output to the console. You can configure to write logs to a file.

Development Guide

Install Dependencies

npm install uuid express ws

Create a New Server

Refer to the server.js or script.py examples to create a custom MCP server.

Contribution Guide

1. Fork the repository.
2. Create a feature branch.
3. Commit your changes and create a Pull Request.

Support Information

Encountered an issue? Visit GitHub Issues or send an email to support@aisdk-mcp.com.

Changelog

1.0.0 (2023 - 10 - 05)

• Initial version released.
• Basic features added.
• Documentation completed.

Security Notice

If you discover a security vulnerability, please report it immediately to security@aisdk-mcp.com.

Author Information

• Developer: AI SDK Team
• Email: contact@aisdk-mcp.com

📄 License

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