MCP (Machine-Curated Programming) Tools and Utilities

This repository contains a collection of tools and utilities for working with the MCP (Model-Context Protocol) framework, which enables AI agents to interact with tools through a standardized protocol.

Project Overview

This project demonstrates various examples of:

• Creating MCP servers with tools for mathematical operations and Paint automation
• Building MCP clients that connect to these servers
• Implementing agentic AI workflows with Google's Gemini model
• Illustrating interactive drawing capabilities through Paint application automation

Key Components

MCP Servers

• example2-3.py: A comprehensive MCP server with mathematical tools (add, subtract, multiply, etc.) and Paint drawing utilities.
• mcp_server.py: A simple MCP server that provides a string reversal tool.
• example_mcp_server.py: Another MCP server implementation with various functionalities.

MCP Clients

• talk2mcp-2.py: Client that connects to the example2-3 server and uses Gemini AI to iteratively solve problems using the MCP tools.
• mcp_client.py: Simple client that connects to the string reverser server.
• AgenticMCPUse.py: An agentic implementation that uses Gemini to interact with MCP tools.

Utility Scripts

• getCoords.py: Helper script to get mouse coordinates for Paint automation.
• Various test scripts (test1.py through test7.py) exploring different MCP functionalities.

Setup Instructions

1. Environment Setup:

   # Create a virtual environment (Python 3.13+ required)
   python -m venv .venv
   source .venv/bin/activate  # On Windows: .venv\Scripts\activate
   
   # Install dependencies
   pip install -e .
   

2. API Key Configuration:

   • Create a .env file in the root directory
   • Add your Gemini API key:

     GEMINI_API_KEY=your_api_key_here
     

Usage Examples

Basic String Reverser

Run the server and client in separate terminals:

# Terminal 1: Start the server
python mcp_server.py

# Terminal 2: Run the client
python mcp_client.py

Math Operations with AI Agent

# Run the agentic MCP client
python AgenticMCPUse.py

Paint Automation with MCP

# First ensure MS Paint is installed on your Windows system
python talk2mcp-2.py

Project Structure

• /Agentic_AI_MCPPaint_Tool: Additional agentic AI tools
• /react_frontend_python_serverClient: Frontend React components with Python backend integration
• Various Python scripts demonstrating MCP capabilities

Dependencies

• mcp[cli]: The MCP framework
• pyautogui: For screen automation
• pywinauto: For Windows application automation
• PIL: For image processing
• google-generativeai: For Gemini AI integration

Notes

• This project primarily targets Windows systems due to MS Paint automation features.
• Some scripts contain experimental features and may require adjustments based on your environment.
• The agentic implementations demonstrate how AI can be used to orchestrate tools via the MCP protocol.