DazzleML/MCP-Server-Tutorial
A detailed hands-on tutorial for learning Model Context Protocol (MCP) server development with Python. Includes working examples, VSCode debugging setup, tests, and a step-by-step guide covering everything from basic architecture to deployment. Perfect for anyone wanting to build AI-integrated tools and to understand how Claude extensions work.
A detailed, hands-on tutorial for learning MCP (Model Context Protocol) server development with working examples and debugging tools.
This tutorial project creates a minimal, well-documented MCP server designed to help you understand:
Requirements: Python 3.10+ (MCP library requirement)
# Run the setup script
scripts\setup.bat
# Or manually:
python -m venv venv
venv\Scripts\activate.bat
pip install mcp# Activate virtual environment
venv\Scripts\activate.bat
# Run tests (basic test without Unicode issues)
python tests\test_simple.py
# Or run full test suite (may have Unicode display issues on Windows)
python tests\test_server.py
# Or use the safe test runner for Windows
scripts\test_safe.batNote for Windows users: If you see Unicode encoding errors, use scripts\test_safe.bat which redirects output to test_output.log.
Key breakpoints to set:
server.py:89 - Tool discoveryserver.py:110 - Tool executionhandlers.py:25 - Individual tool handlersconfig\claude_desktop.jsonMCPDebugTest/
├── CLAUDE.md # Detailed learning documentation
├── README.md # This file
├── setup.bat # Windows setup script
├── requirements.txt # Python dependencies
├── simple_mcp_server/ # Main MCP sLoading reviews...