Two protocol standards keep getting confused in architecture meetings, and they solve perpendicular problems. MCP standardizes how one agent talks down to its tools; A2A standardizes how agents talk sideways to each other. Here's the anatomy of both, the mental model that makes the difference stick, and an honest case for why most systems still shouldn't go multi-agent.
Skip the protocol theory. Build a working Model Context Protocol server in TypeScript, wire it into Claude Code, and make a real agent call a real tool in under an hour. Code, traps, and how to know it works.
Model Context Protocol turns your internal services into something an agent can call without bespoke glue. What MCP actually is, what it costs to expose one well, and the patterns that hold up when the caller is an LLM instead of a UI.