🌐 MCP-Use-TS is the open source TypeScript library to connect any LLM to any MCP server and build custom agents that have tool access, without using closed source or application clients.

💡 Let developers easily connect any LLM to tools like web browsing, file operations, and more with full TypeScript support.

Features

✨ Key Features

| Feature | Description | |---------|-------------| | 🔄 Ease of use | Create your first MCP capable agent with just 6 lines of TypeScript code | | 🤖 LLM Flexibility | Works with any LangChain supported LLM that supports tool calling (OpenAI, Anthropic, Groq, LLama etc.) | | 🌐 HTTP Support | Direct connection to MCP servers running on specific HTTP ports | | ⚙️ Dynamic Server Selection | TODO Agents can dynamically choose the most appropriate MCP server for a given task from the available pool | | 🧩 Multi-Server Support | TODO Use multiple MCP servers simultaneously in a single agent | | 🛡️ Tool Restrictions | TODO Restrict potentially dangerous tools like file system or network access | | 📝 Type Safety | TODO Full TypeScript support with type definitions for all APIs and configurations |

Quick start

With npm:

npm install mcp-use-ts

Or install from source:

git clone https://github.com/dforel/mcp-use-ts.git
cd mcp-use-ts
npm install
npm run build

Installing LangChain Providers

mcp-use-ts works with various LLM providers through LangChain. You'll need to install the appropriate LangChain provider package for your chosen LLM. For example:

# For OpenAI
npm install @langchain/openai

# For Anthropic
npm install @langchain/anthropic

# For other providers, check the [LangChain chat models documentation](https://js.langchain.com/docs/integrations/chat/)

and add your API keys for the provider you want to use to your .env file.

OPENAI_API_KEY=
ANTHROPIC_API_KEY=

Important: Only models with tool calling capabilities can be used with mcp-use-ts. Make sure your chosen model supports function calling or tool use.

Spin up your agent:

import { config } from 'dotenv';
import { ChatOpenAI } from '@langchain/openai';
import { MCPAgent, MCPClient } from 'mcp-use-ts';

async function main() {
    // Load environment variables
    config();

    // Create configuration object
    const config = {
      mcpServers: {
        playwright: {
          command: 'npx',
          args: ['@playwright/mcp@latest'],
          env: {
            DISPLAY: ':1'
          }
        }
      }
    };

    // Create MCPClient from configuration object
    const client = MCPClient.fromConfig(config);

    // Create LLM
    const llm = new ChatOpenAI({ modelName: 'gpt-4' });

    // Create agent with the client
    const agent = new MCPAgent({
        llm,
        client,
        maxSteps: 30
    });

    try {
        // Run the query
        const result = await agent.run(
            'Find the best restaurant in San Francisco'
        );
        console.log('\nResult:', result);
    } finally {
        // Clean up resources
        await client.closeAllSessions();
    }
}

main().catch(console.error);

You can also add the servers configuration from a config file like this:

const client = MCPClient.fromConfigFile(
    path.join(__dirname, 'browser_mcp.json')
);

Example configuration file (browser_mcp.json):

{
  "mcpServers": {
    "playwright": {
      "command": "npx",
      "args": ["@playwright/mcp@latest"],
      "env": {
        "DISPLAY": ":1"
      }
    }
  }
}

For other settings, models, and more, check out the documentation.

Example Use Cases

Web Browsing with Playwright

import { config } from 'dotenv';
import { ChatOpenAI } from '@langchain/openai';
import { MCPAgent, MCPClient } from 'mcp-use-ts';
import path from 'path';

async function main() {
    // Load environment variables
    config();

    // Create MCPClient from config file
    const client = MCPClient.fromConfigFile(
        path.join(__dirname, 'browser_mcp.json')
    );

    // Create LLM
    const llm = new ChatOpenAI({ modelName: 'gpt-4' });
    // Alternative models:
    // const llm = new ChatAnthropic({ modelName: 'claude-3-sonnet-20240229' });
    // const llm = new ChatGroq({ modelName: 'llama3-8b-8192' });

    // Create agent with the client
    const agent = new MCPAgent({
        llm,
        client,
        maxSteps: 30
    });

    try {
        // Run the query
        const result = await agent.run(
            'Find the best restaurant in San Francisco USING GOOGLE SEARCH',
            { maxSteps: 30 }
        );
        console.log('\nResult:', result);
    } finally {
        // Clean up resources
        await client.closeAllSessions();
    }
}

main().catch(console.error);

This example demonstrates how to connect to an MCP server running on a specific HTTP port. Make sure to start your MCP server before running this example.

Multi-Server Support

MCP-Use-TS allows configuring and connecting to multiple MCP servers simultaneously using the MCPClient. This enables complex workflows that require tools from different servers, such as web browsing combined with file operations or 3D modeling.

Configuration

You can configure multiple servers in your configuration file:

{
  "mcpServers": {
    "airbnb": {
      "command": "npx",
      "args": ["-y", "@openbnb/mcp-server-airbnb", "--ignore-robots-txt"]
    },
    "playwright": {
      "command": "npx",
      "args": ["@playwright/mcp@latest"],
      "env": {
        "DISPLAY": ":1"
      }
    }
  }
}

Usage

The MCPClient class provides methods for managing connections to multiple servers. When creating an MCPAgent, you can provide an MCPClient configured with multiple servers.

By default, the agent will have access to tools from all configured servers. If you need to target a specific server for a particular task, you can specify the serverName when calling the agent.run() method.

// Example: Manually selecting a server for a specific task
const airbnbResult = await agent.run(
    'Search for Airbnb listings in Barcelona',
    { serverName: 'airbnb' } // Explicitly use the airbnb server
);

const googleResult = await agent.run(
    'Find restaurants near the first result using Google Search',
    { serverName: 'playwright' } // Explicitly use the playwright server
);

Agent-Specific Verbosity

If you only want to see debug information from the agent without enabling full debug logging, you can set the verbose parameter when creating an MCPAgent:

// Create agent with increased verbosity
const agent = new MCPAgent({
    llm,
    client,
    verbose: true  // Only shows debug messages from the agent
});

This is useful when you only need to see the agent's steps and decision-making process without all the low-level debug information from other components.

Roadmap

Contributing

We love contributions! Feel free to open issues for bugs or feature requests.

Requirements

• Node.js 18+
• TypeScript 5.0+
• MCP implementation (like Playwright MCP)
• LangChain and appropriate model libraries (OpenAI, Anthropic, etc.)

Citation

If you use MCP-Use-TS in your research or project, please cite:

@software{mcp_use_ts,
  author = {dforel},
  title = {MCP-Use-TS: MCP Library for TypeScript},
  year = {2025},
  publisher = {GitHub},
  url = {https://github.com/dforel/mcp-use-ts}
}

Other

this project is a fork of mcp-use

i hope you enjoy it

License

MIT