SpringBoot MCP Server

A SpringBoot project that demonstrates the integration of Spring AI with a simple student management system, allowing AI models to interact with the application through tools.

Overview

This is a demonstration of how to build a Spring Boot application that exposes its functionality as tools for AI models using the Model Context Protocol (MCP). The application provides a simple student management system with CRUD operations, and these operations are exposed as tools that can be called by AI models.

The project consists of two main components:

1. Spring Boot Server: A Java-based backend that manages student data and exposes operations as MCP tools.
2. Python Chatbot: A client that connects to the server and allows interaction with an LLM (Large Language Model) through LM Studio.

Project Structure

Core Components

• Models: Data structures used in the application

  • Student: Represents a student with id, name, email, age, and address
  • Address: Represents an address with pincode, city, and state
  • Response: Standard response format with success flag, message, and data

• Services: Business logic implementation

  • StudentDetailsService: Manages student data with in-memory storage
  • StudentTools: Exposes student operations as MCP tools

• Configuration: Application setup

  • OpenApiConfig: Configuration for API documentation
  • McpServerApplication: Main application class with Spring AI tool configuration

Bot Directory

The bot directory contains a Python-based chatbot that interacts with the MCP server:

• chatbot.py: A comprehensive Python script that:

  • Connects to the MCP server via subprocess
  • Communicates with LM Studio for LLM integration
  • Provides both interactive and stdio modes for interaction
  • Formats responses for better readability
  • Handles tool calls between the LLM and MCP server

• runme.sh: Shell script to launch the chatbot with the correct parameters

  • Starts the MCP server in stdio mode
  • Configures the chatbot for interactive use

• requirements.txt: Lists Python dependencies for the chatbot

  • aiohttp: For asynchronous HTTP requests
  • asyncio-mqtt: For MQTT communication
  • sseclient-py: For Server-Sent Events

Usage

Running the Spring Boot Server

1. Build the project using Maven:

   mvn clean package
   

2. Run the server:

   java -jar target/mcp-server-0.0.1-SNAPSHOT.jar
   

The server will start on port 8081 by default.

LM Studio Pre-requisites

Before using the chatbot, ensure that:

1. LM Studio is running in dev mode on port 1234
2. The LLM model qwen/qwen2.5-vl-7b is loaded in LM Studio

Using the Chatbot

1. Install Python dependencies:

   cd bot
   pip install -r requirements.txt
   

2. Run the chatbot using the provided script:

   cd bot
   ./runme.sh
   

3. Interact with the chatbot in the terminal. You can:

   • Ask questions about students
   • Add new student records
   • Retrieve student information
   • Delete student records

API Documentation

The application includes OpenAPI documentation, which can be accessed at:

http://localhost:8081/swagger-ui.html

Available MCP Tools

The following tools are exposed through the MCP protocol:

• LIST_ALL_STUDENT_RECORDS: Returns a list of all student records
• ADD_STUDENT_RECORD: Adds a new student record
• DELETE_STUDENT_RECORD: Deletes a student record by ID
• GET_STUDENT_RECORD: Retrieves a student record by ID

Technical Details

• Java Version: 24
• Spring Boot Version: 3.5.0
• Spring AI Version: 1.0.0
• Data Storage: In-memory (ConcurrentHashMap)
• API Documentation: OpenAPI 2.3.0

Development

This project demonstrates how to:

1. Create a Spring Boot application with Spring AI integration
2. Expose application functionality as MCP tools
3. Build a Python client that connects to the MCP server
4. Integrate with LLMs through LM Studio

License

This project is for demonstration purposes.

Author

Sumit Gaur