Phase 3 — Core AI Engineering · Lesson 35 · 20 XP
MCP (Model Context Protocol)
Without a shared standard, every AI app has to write its own bespoke integration for every tool or data source it wants to use. MCP (Model Context Protocol) is an open protocol that standardizes this: an MCP server exposes tools and resources in a consistent way, and any MCP client — Claude Code, or an app you build — can connect to it and use them.
This is the same tool-definition-and-call pattern from Lesson 23, just delivered over a standard protocol instead of a one-off integration — write a tool once as an MCP server, and any MCP-compatible client can use it without custom glue code.
Exercise
Connect an MCP client to a simple MCP server (or stand one up yourself) and call one of its tools, observing the same request/response shape you built by hand in Lesson 23.
Check yourself
1. What problem does a standard protocol like MCP solve compared to every app writing its own custom tool integration?
2. In MCP's client/server model, which side actually defines the available tools?
Workflows vs agents, planning
Answer the check-yourself questions to unlock this