Skip to main content

​
Quickstart Guide

This guide will help you get started with easy-mcp-use quickly. We’ll cover installation, basic configuration, and running your first agent.

​
Installation

You can install easy-mcp-use using npm: 
npm install easy-mcp-use
Or install from source: 
git clone https://github.com/dforel/easy-mcp-use.git
cd easy-mcp-use
npm install
npm run build

​
Installing LangChain Dependencies

easy-mcp-use works with various LLM providers through LangChain. The core LangChain dependencies are included, but you might need to install additional provider-specific packages: 
# For OpenAI
npm install @langchain/openai

# For Anthropic
npm install @langchain/anthropic

# For other providers, check the [LangChain documentation](https://js.langchain.com/docs/integrations/chat/)
Important: Only models with tool calling capabilities can be used with easy-mcp-use. Make sure your chosen model supports function calling or tool use.

​
Environment Setup

Create a .env file in your project root: 
OPENAI_API_KEY=your_api_key_here
ANTHROPIC_API_KEY=your_api_key_here

​
Your First Agent

Here’s a simple example to get you started: 
import { config } from 'dotenv';
import { ChatOpenAI } from '@langchain/openai';
import { MCPAgent, MCPClient } from 'easy-mcp-use';

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

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

    // Create MCPClient from configuration object
    const client = MCPClient.fromDict(mcpConfig);

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

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

    // Run the query
    const result = await agent.run(
        'Find the best restaurant in San Francisco USING GOOGLE SEARCH'
    );
    console.log('\nResult:', result);
}

main().catch(console.error);

​
Configuration Options

You can also add the servers configuration from a config file: 
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"
      }
    }
  }
}

​
Using Multiple Servers

The MCPClient can be configured with multiple MCP servers, allowing your agent to access tools from different sources. This capability enables complex workflows spanning various domains. Configuration: Define multiple servers in your configuration file (multi_server_config.json): 
{
  "mcpServers": {
    "airbnb": {
      "command": "npx",
      "args": ["-y", "@openbnb/mcp-server-airbnb", "--ignore-robots-txt"]
    },
    "playwright": {
      "command": "npx",
      "args": ["@playwright/mcp@latest"],
      "env": {
        "DISPLAY": ":1"
      }
    }
  }
}
Usage: 
// Assuming MCPClient is initialized with the multi_server_config.json
const client = MCPClient.fromConfigFile('multi_server_config.json');
const agent = new MCPAgent({ llm, client }); // Server manager not enabled by default

// Manually specify the server if needed
const result = await agent.run(
    'Search for Airbnb listings in Barcelona',
    { serverName: 'airbnb' }
);

​
Enabling Dynamic Server Selection (Server Manager)

To improve efficiency when working with multiple servers, you can enable the Server Manager: 
// Enable server manager during agent creation
const agent = new MCPAgent({
    llm,
    client,
    useServerManager: true
});

// The agent will automatically select the appropriate server
const result = await agent.run(
    'Search for a place in Barcelona on Airbnb, then Google nearby restaurants.'
);

​
Restricting Tool Access

You can control which tools are available to the agent: 
import { config } from 'dotenv';
import { ChatOpenAI } from '@langchain/openai';
import { MCPAgent, MCPClient } from 'easy-mcp-use';

async function main() {
    config();

    const mcpConfig = {
      mcpServers: {
        playwright: {
          command: 'npx',
          args: ['@playwright/mcp@latest'],
          env: {
            DISPLAY: ':1'
          },
          allowedTools: ['navigate', 'click'] // Only allow these tools
        }
      }
    };

    const client = MCPClient.fromDict(mcpConfig);
    const llm = new ChatOpenAI({ modelName: 'gpt-4' });
    const agent = new MCPAgent({ llm, client });

    const result = await agent.run(
        'Navigate to example.com and click the first link'
    );
    console.log('\nResult:', result);
}

main().catch(console.error);
⌘I