What you are going to build
You will finish this course with an MCP server for a shop's support team, and a small agent that connects to it, reads its tools and uses them to answer customers.
MCP, the Model Context Protocol, is an open standard for connecting AI applications to tools and data. Write a server once, and Claude Code, Claude Desktop, Cursor, VS Code and your own agents can all use it, with no custom code for each one.
No API key
Servers do not talk to models, so most of the course needs no model at all. Part 6 adds an agent, and you write a small stand-in model for it that chooses tools from the user's message. Every output is from a real run on the pinned SDK.
What you need
- Python 3.10 or later.
- Functions, type hints, Pydantic and async from Python for AI. Part 5 uses FastAPI from APIs for AI.
pip install "mcp==2.2.0" "fastapi==0.141.1" pytestmcp is the official Python SDK, version 2. Much older example code online uses version 1, where the server class was called FastMCP; the ideas carry over, the names do not always.
Every expert started right here.