Before we build anything, you need to know what the pieces are and how they fit together. This lesson is the map. Every lesson after this is walking the territory.

There are exactly five primitives in the Claude Code extensibility system:

1. Skills    — "Here's how to do X"
2. Agents    — "Go do X using these skills"
3. Rules     — "Always/never do Y"
4. Hooks     — "When Z happens, run this script"
5. Commands  — "When I type /foo, do this workflow"

That's the whole thing. Every customization you'll ever make to your AI coding environment is a combination of these five.