In module one, I covered why the terminal is the most powerful abstraction for an AI agent architecture:

In module two, we’ll see some of the things you can do within that abstraction.

Module 2 moves from abstraction to architecture. Where Module 1 explained what Claude Code is, Module 2 explains how it works as a workflow engine — the universal pattern behind every task, the seven tools the agent uses under the hood, and the mental models you need to start building effective daily workflows.

The session runs ~22 minutes and remains at an abstract level — the practical sessions come later, but the conceptual framework laid here is what makes the practical work stick.