Business Analyst AI Assistant

A Spring AI-powered client application that demonstrates how to create AI assistants for business analysts by connecting to a Jira MCP (Model Context Protocol) server. This project serves as a learning example of how to leverage Spring AI to create MCP servers and clients that can effectively communicate with LLMs through defined tools.

Overview

This project is a client application that connects to a Jira MCP server, enabling AI assistants to interact with Jira through a standardized protocol. It demonstrates the practical implementation of:

• Spring AI integration
• MCP (Model Context Protocol) client implementation
• Server-Sent Events (SSE) communication
• Integration with Jira API through MCP server (developed in a separate project : https://github.com/AzamMehr/jira-mcp-server)

Prerequisites

• Java 21 or later
• Gradle
• Access to a running Jira MCP server (see jira-mcp-server)

Dependencies

The project has a clear dependency on the Jira MCP server. Key dependencies include:

• Spring AI
• Spring Boot
• Spring WebFlux
• Spring AI Core

Project Structure

src/
├── main/
│   ├── java/
│   │   └── com/ba/copilot/
│   │       ├── config/
│   │       ├── controller/
│   │       ├── model/
│   │       └── service/
│   └── resources/
│       └── application.properties

Configuration

Configure the application using environment variables or application.properties:

# MCP Server Configuration
mcp.server.url=${MCP_SERVER_URL}

# Application Configuration
server.port=8080

Building the Project

./gradlew build

Running the Application

./gradlew bootRun

Features

• Connects to Jira MCP server using SSE communication
• Enables AI assistants to interact with Jira through defined tools
• Demonstrates Spring AI integration with MCP protocol
• Provides a learning example of MCP server-client architecture

Learning Objectives

This project serves as a practical example of:

1. Creating MCP servers using Spring AI
2. Implementing MCP clients that connect to these servers
3. Enabling LLMs to use tools defined in MCP servers
4. Implementing SSE-based communication between client and server

Contributing

1. Fork the repository
2. Create your feature branch
3. Commit your changes
4. Push to the branch
5. Create a new Pull Request

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

Acknowledgments

• jira-mcp-server - The MCP server implementation this client connects to
• Spring AI team for providing the framework
• Atlassian for Jira API