Introduction to MCP
Go from zero to a working CLI chatbot with tools, resources, and prompts — all built on the Python SDK.
You're a developer who knows Python and has worked with the Claude API. You want to build MCP servers and clients that give Claude access to external services — without writing mountains of integration code.
Understand what MCP is, why it exists, and how clients and servers communicate through standardised message types.
3 lessons
Set up your project, define tools with the Python SDK, and test them with the MCP Inspector.
4 lessons
Build the client side: implement tool calls, read resources, and invoke server-defined prompts.
5 lessons
Test your knowledge and master the three-primitive decision framework for choosing tools, resources, or prompts.
2 lessons
You should be comfortable with Python and have experience making API calls to Claude. Familiarity with async/await and decorators is helpful but not required — the course explains MCP-specific patterns as you encounter them.
A CLI-based chatbot that implements both an MCP client and an MCP server. You'll define tools (document read/edit), resources (document listing and fetching), and prompts (a /format command) — then test everything with the MCP Inspector and the CLI.
Tool use is how Claude calls functions you define. MCP is about who defines those functions. With MCP, someone else (often the service provider) has already built and tested the tools — you just connect to their MCP server instead of writing everything yourself.
Most developers complete the course in 2-3 hours, including hands-on exercises. The free module (Introduction) takes about 30 minutes.