Course 01 · Lesson 11 of 12
Course Summary: What You Now Know
You have finished Claude Code Foundations. Here is everything you now know and why it matters for everything that comes next.
What you learned
What Claude Code is: an AI agent that lives inside your computer, has direct access to your files, and builds things through conversation. Not a chatbot. Not copy-paste. An agent that works.
How it is different from Claude AI: Claude AI answers questions. Claude Code builds things. One lives in your browser. The other lives on your computer with direct file access and the ability to take real actions.
Where it runs: Desktop App (best for beginners), Terminal (most powerful), VS Code, JetBrains, and the web. You can switch between them at any time. Your settings and memory files work across all of them.
How to install it: Desktop App via claude.com/download. Terminal via the one-line install script or npm. Both paths are now documented and tested.
How to write a first prompt: describe the outcome, not the code. Give Claude the result you want to see, not the steps to get there.
How to navigate the interface: project files in the sidebar, conversation in the centre, approvals when Claude needs them, and Ctrl+C to stop at any time.
Approval modes: Default asks before every action, Auto-Approve Edits lets files update without asking (faster), and Plan Mode lets you review the full plan before anything runs. Start on Default and move to Auto-Approve when you feel ready.
The context window: Claude's working memory for one session. Keep sessions focused and start fresh for new tasks. CLAUDE.md, which you will set up in the next course, solves the cross-session memory problem permanently.
How to handle mistakes: describe what went wrong, ask Claude to undo or try again, simplify when needed, and always describe outcomes rather than code.
What you built
A personal introduction page with your name, your story, and your design choices, built through conversation with no code written by hand.
What is coming next
Track 1: The Memory System
Right now, every time you start a new Claude Code session, Claude does not know anything about you. It does not know your name, your design style, your audience, or your goals.
The Memory System fixes this permanently.
You will set up a small collection of files that Claude reads at the start of every session. By the time you type your first prompt, Claude already knows everything it needs to know about you. Every build you make from that point forward is tailored specifically to you.
This is one of the most powerful things you will do in this entire course series. It changes the quality of everything Claude builds for you, permanently.
The next course is waiting.
Before you move on
Do not skip the homework lesson. It is the difference between finishing a course and actually building something that exists in the real world.
The homework for this course is in the next lesson.
Do it with other builders
Post your homework. Get unstuck.
Every course ends with homework. Post yours in Claude Codex Mastery, the community this course comes from, and get feedback from Yar and 400+ people building the same way.
Join the community · $29/mo