Your first conversation

This is the lesson where it stops being abstract.

A conversation with Claude Code is a back-and-forth in your terminal. You type a question or a request, Claude runs whatever tools it needs to answer, and tells you what it found.

Pick a folder. Any folder you don’t mind it poking around in — your downloads, a side project, the recipes-site from Lesson 1.10 if it’s still around. Open a terminal there.

Say hi

claude

The terminal transforms. The familiar prompt is gone; instead, you have an empty box waiting for input. Type:

what's in this folder?

Press enter. Watch what happens — don’t skip past this part.

You’ll see Claude do something like this:

1. A short “thinking” indicator.
2. A line announcing a tool — usually something like Bash(ls).
3. The raw output of that command, right there in your terminal.
4. A short summary in Claude’s voice: “This folder has three subdirectories and a README…”

That sequence is the turn loop. Every interaction is some version of it: you ask, Claude thinks, Claude maybe uses a tool, Claude responds.

Try one more

Don’t quit yet. Ask a follow-up that builds on what just happened:

what does the README say?

This time Claude won’t re-run ls — it already knows the folder layout from a second ago. It’ll go straight to reading the README and summarizing. The conversation has memory: Claude remembers what it just did, like a person you’re chatting with would.

Try a couple of questions of your own. Get a feel for the rhythm. Notice how the answer is about your files — not a generic answer that could apply to any folder.

Getting out

When you’re done, press Ctrl + C twice. You’re back at your normal prompt.

The conversation is over. Next time you run claude, you start fresh — a new turn loop, no memory of this one.

What’s next

You’ve had a conversation. But it’s worth asking: what exactly can Claude see while you’re chatting? And what can’t it? That’s the next lesson.