A standard protocol for exposing tools and data to any model.
Made for you now: slide lessons from basics to production, usually with code and a quiz. See a finished one
Without a standard, every app wires every tool to every model by hand. The Model Context Protocol (MCP) defines one way for an AI application to discover and use external tools and data.
An MCP server exposes capabilities: tools (actions), resources (readable data) and prompts (reusable templates). An MCP client inside the host application connects to servers, lists what they offer, and relays the model’s tool calls.
A connected server can influence the model through its tool descriptions and results. Connect only servers you trust, scope their permissions, and keep a human approval step for sensitive actions.
Step 1
Host app + model
Chat app or agent that needs tools.
Step 2
MCP client
Connects, lists tools and resources.
Step 3
MCP server
Wraps a service: files, database, API.
Step 4
Result back
Returned to the model as context.