MCP: Model Context Protocol
MCP is one common way to connect AI apps to tools and data. Instead of writing special code for every tool, you plug the tool in. This lesson explains the parts, how a call flows, and how to stay safe.
New words
- MCP
- Model Context Protocol: an open standard to connect AI apps to outside systems.
- Host
- The AI application that the user talks to, such as a chat app.
- Client
- A part inside the host that keeps one connection to one server.
- Server
- A program that offers tools, data or prompts to the AI app.
- Tool
- A function that the model can ask the app to run.
- Transport
- The way client and server send messages to each other.
What you will learn
- The problem: too many cables
- Host, client and server
- What can a server offer?
- How one call flows
- Local and remote servers
- Try it: connect and disconnect servers
- MCP and tool calling
- Safety: connect only what you trust
- Trade-offs
- Interview tip
- Key takeaways
10 quiz questions at the end.
The full lesson, with interactive demos and the quiz, opens after you sign in with Google. It is free.