Nestjs MCP Server Langchainjs Demo

🚀 Example of Using NestJS MCP Server - Model Context Protocol

Author: @LiusDev

This repository demonstrates the implementation of the Model Context Protocol (MCP) and microservice architecture using NestJS. It consists of two main services:

1. mcp-server: Provides the functionality to obtain the current time context for use by Large Language Models (LLMs).
2. mcp-backend: A client that uses LangChain.js and integrates the MCP client SDK to connect to the MCP server.

🚀 Quick Start

Prerequisites

• Node.js (v20 or higher)
• npm

📦 Installation

1. Clone the repository.
2. Install dependencies:

npm install

3. Copy .env.example to .env and add your OpenAI API key:

cp .env.example .env

Then edit the .env file to include your OpenAI API key:

OPENAI_API_KEY=your_openai_api_key_here
OPENAI_API_URL=https://api.openai.com/v1
PORT=3001

Running the Services

You need to run both services simultaneously to achieve full functionality:

Starting the MCP Server

npm run start:dev mcp-server

This will start the MCP server on port 3000 (by default).

Starting the MCP Backend

npm run start:dev mcp-backend

This will start the MCP backend on port 3001 (by default).

💻 Usage Examples

Basic Usage

Once both services are running, you can test the functionality by sending a POST request:

Sample Request

Send a POST request to http://localhost:3001 with the following JSON body:

{
  "message": "What time is it in Viet Nam?"
}

Using curl

curl -X POST http://localhost:3001 -H "Content-Type: application/json" -d '{"message":"What time is it in Viet Nam?"}'

Using Postman

1. Open Postman and create a new request.
2. Set the URL to http://localhost:3001.
3. Select the POST method.
4. Add Content-Type: application/json to the request headers.
5. Enter the following JSON in the request body:

{
  "message": "What time is it in Viet Nam?"
}

6. Send the request and view the response.

Advanced Usage

To connect to multiple MCP servers in mcp-backend, make the following configuration in the src/main.ts file:

import { NestFactory } from '@nestjs/core';
import { AppModule } from './app.module';

async function bootstrap() {
  const app = await NestFactory.create(AppModule);

  // Add the hosts and ports of other MCP servers
  const otherMcpServers = ['http://localhost:3002', 'http://localhost:3003'];
  
  // Register the MCP service provider (optional)
  // app.registerMcpProvider(otherMcpServers);

  await app.listen(3001);
}

🔧 Technical Details

Architecture

• mcp-server: Responsible for processing and returning time-related requests.
• mcp-backend: Acts as a client to send requests to the MCP server and receive responses.

Technologies Used

• NestJS: Used for building the microservice architecture.
• LangChain.js: Used for interacting with large language models.
• Axios: Used for HTTP request handling.

📄 License

This project is distributed under the UNLICENSED license. For more details, please refer to the LICENSE file.