A2a MCP Server

🚀 MCP Server Demo

A production-ready task management system built on MCP (Model Control Protocol) and Kafka.

🚀 Quick Start

This project demonstrates a powerful task management system that uses MCP to enable AI agents to interact with a Kafka-based task queue. The system supports task management (creating, updating, and completing tasks), notification handling, and real-time event processing through Kafka.

✨ Features

• Task Management: Create, update, adjust priorities, and complete production tasks.
• Notification System: Real-time notifications with priority levels.
• Kafka Integration: Reliable message queuing and event stream processing.
• MCP Tools: AI-friendly interfaces for task and notification operations.
• Consumer Service: Background processing based on Kafka messages.

📦 Installation

Requirements

• Python 3.13+
• Kafka cluster (local or AWS MSK)
• Confluent Kafka Python client

Steps

# Clone the repository
git clone https://github.com/yourusername/mcp-server-demo.git
cd mcp-server-demo

# Install dependencies
pip install -e .

📚 Documentation

Configuration

Update the Kafka configuration in kafka_config.py with your actual Kafka cluster details:

KAFKA_CONFIG = {
    'bootstrap.servers': 'your-kafka-bootstrap-servers',
    'security.protocol': 'SASL_SSL',
    'sasl.mechanisms': 'SCRAM-SHA-512',
    'sasl.username': 'your-username',
    'sasl.password': 'your-password',
}

Usage

Start the Server

python main.py

Load Test Data

To populate the system with sample tasks and notifications:

python kafka_test_data.py

MCP Tools

The system exposes the following MCP tools for AI agents:

Task Management

• fetch_queue: Get the list of pending tasks.
• change_task_priority: Update the priority of a task.
• pickup_task: Mark a task as in progress.
• complete_task: Mark a task as completed.
• get_task_details: Get detailed information about a task.
• check_task_status: Check the current status of a task.

Notification Management

• check_notification_count: Get the number of unread notifications.
• get_notification_list: Get a filtered list of notifications.
• mark_notification_as_read: Mark a notification as read.

Architecture

The system consists of the following components:

• MCP Server: Provides an interaction interface for AI agents.
• Kafka Producer: Sends messages to Kafka topics.
• Kafka Consumer: Processes messages from Kafka topics.
• Task Service: Business logic for task management.
• Notification Service: Business logic for notification handling.

Development

Project Structure

mcp-server-demo/
├── main.py                # MCP server initialization
├── kafka_config.py        # Kafka configuration
├── consumer_service.py    # Kafka consumer service
├── task_service.py        # Task management logic
├── notification_service.py # Notification handling logic
└── kafka_test_data.py     # Test data generator

📄 License

MIT License

💪 Contribution

Contributions are welcome! Please feel free to submit a Pull Request.