MCP-Spring-Python-AI: Multi-Protocol Chatbot

A Model Context Protocol (MCP)-based chatbot integrating Spring Boot (AI + Ollama), Python/NodeJS MCP servers, and an Angular frontend.

Author: ER-RAHOUTI ACHRAF

---

Table of Contents

• Overview
• Key Features
• Architecture
• Installation
• Usage
• Technologies
• Configuration
• Testing
• Screenshots

---

Overview

A modular chatbot demonstrating MCP protocol implementation with:

• Spring Boot (SSE-based MCP client/server)
• Python/NodeJS (STDIO-mode MCP tools)
• Angular frontend for interactive chat.

---

Key Features

• Multi-Backend Integration:
  • Spring AI + Ollama (Qwen3 model)
  • Python/NodeJS MCP servers for tool execution.
• Real-Time Communication: SSE for streaming responses.
• Agent-Based UI: Angular frontend with thinking/reply distinction.
• Tool Ecosystem: Stock lookup, file ops, employee info, etc.

---

Architecture

The project consists of four main components:

• mcp-client/: Spring Boot MCP Client

  • Implements AI agents and REST controllers.
  • Connects to MCP servers and manages AI interactions.

• mcp-server/: Spring Boot MCP Server

  • Provides MCP tools and services.
  • Implements backend logic for tool execution.

• mcp-frontend/: Angular Frontend

  • Interactive chat UI with agent thinking and reply views.
  • Built with Angular CLI.

• python-mcp-server/: Python MCP Server

  • Additional MCP tools implemented in Python.
  • Provides extended functionality via MCP protocol.

• Screenshots/: Directory containing project screenshots for reference.
  gi

---

Installation

Prerequisites

• Ollama (ollama pull qwen3)
• Node.js 18+
• Angular CLI (npm install -g @angular/cli)
• Postman (for API testing)
• Java 17+ and Maven (for Spring Boot projects)

Steps

1. Clone the repository and navigate to the project directory:

   git clone https://github.com/your-username/mcp-spring-python-ai.git  
   cd mcp-spring-python-ai  

2. Build and run the Spring Boot client and server:

   cd mcp-client
   ./mvnw clean install
   ./mvnw spring-boot:run

   In a separate terminal:

   cd mcp-server
   ./mvnw clean install
   ./mvnw spring-boot:run

3. Install and run the Angular frontend:

   cd mcp-frontend
   npm install
   ng serve

   Open your browser at http://localhost:4200/.
4. Run the Python MCP server (optional):

   cd python-mcp-server
   python main.py

---

Usage

• Use the Angular frontend to interact with the chatbot.
• The Spring Boot client communicates with MCP servers for AI and tool execution.
• The Python MCP server provides additional tools accessible via MCP protocol.

---

Technologies

• Java 17+, Spring Boot 3.x
• Angular 20.x
• Python 3.x
• Node.js 18+
• Ollama Qwen3 AI model
• Model Context Protocol (MCP)

---

Configuration

• Configure MCP servers in mcp-client/src/main/resources/mcp-servers.json.
• Application properties are in src/main/resources/application.properties for client and server.
• Adjust Angular environment and configuration files as needed.

---

Testing & Debugging

• Run unit tests for Spring Boot projects:

  ./mvnw test

• Run Angular unit tests:

  ng test

• Use Postman or similar tools for API testing.

---

Screenshots

See the Screenshots/ directory for visual references of the application in action.

---

License

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