Your First Conversation | Claude Code Guide

You have a terminal. Claude Code is installed. You've typed commands and the computer responded. Now comes the part where it gets fun. You're about to have a conversation with an AI that can actually touch your files. Not just talk about them. Touch them. Create, edit, move, organize. This is the moment it stops being a concept and becomes a tool. ## Starting Claude Code First, let's create a practice folder. You don't want Claude Code working in your entire Documents folder. Give it a specific room to work in. **Windows (PowerShell):** ```bash mkdir "$HOME\Documents\practice" cd "$HOME\Documents\practice" ``` **Mac/Linux:** ```bash mkdir ~/Documents/practice cd ~/Documents/practice ``` Now start Claude Code: ```bash claude ``` You'll see a welcome message and a prompt. The prompt is where you type. This is a conversation. Talk normally. ## Three Starter Prompts Let's do three things, each one building on the last. **Prompt 1: Ask a question** Type this: ```text What files are in this folder? ``` Claude will look around and tell you the folder is empty. Because it is. We just created it. But notice what happened: Claude actually checked your folder. It didn't guess. It looked. **Prompt 2: Create something** Type this: ```text Create a file called hello.txt that says "This is my first file created by AI" ``` Claude will ask your permission before creating the file. This is important. Claude never makes changes without asking first. You'll see something like: ```text I'll create the file hello.txt with that content. > Create file: hello.txt Allow? [y/n] ``` Type **y** and press Enter. Done. Open your practice folder in your file explorer. There's the file. **Prompt 3: Change something** Type this: ```text Read hello.txt and change the message to say "I told an AI to create this file, and it actually did" ``` Claude reads the file, shows you the change it wants to make, and asks permission. Approve it. Open the file again. The text changed. Three prompts. No code. Claude read a folder, created a file, and edited it. All because you asked in plain English. ## Notice What Just Happened Take a second. You just: - Created a real file on your computer by describing what you wanted - Modified that file by describing the change in English - Didn't write a single line of code, HTML, or anything technical If you've used web-based AI chats like ChatGPT, Gemini, or even Claude on claude.ai, you know the workflow: copy content into the chat, attach a document, get a response, copy it back out, paste it into your file, repeat. It works, but it's clunky. Especially when you're working with multiple files or making iterative changes. Claude Code skips all of that. It reads your files directly. It writes changes directly. No copy-pasting. No attachments. No "here's the updated version, please replace your existing code with this." You describe what you want, and it happens in your actual files. This is the entire workflow. It scales from "create a text file" to "build me a website with five pages." The only thing that changes is the complexity of your description. ## The Permission System You noticed Claude asked before making changes. This is by design. Claude Code operates on a trust-but-verify model. Every time Claude wants to: - Create a file - Edit a file - Delete a file - Run a command It asks first. You see exactly what it wants to do. You approve or decline. This means you can't accidentally break things. Claude might suggest something you don't want. You say no. It suggests something different. It's a conversation, not a runaway train. As you get more comfortable, you can tell Claude to stop asking about certain things. But for now, approve everything manually. It's how you learn what Claude does. ## Exiting Claude Code When you're done, you have three options: 1. Type `/exit` and press Enter 2. Press **Ctrl + C** (this works everywhere) 3. Type **exit** and press Enter Your files stay. The conversation disappears. Next time you type `claude` in this folder, Claude Code starts fresh. It reads your files again, but the previous conversation is gone. This is actually a feature. Fresh conversations don't carry baggage from earlier misunderstandings. ## Try a Few More Things Before we move on, try these prompts. Each one shows you something Claude Code can do: - "Create an HTML page with a dark background and the text 'Hello World' in large white letters" - "What's in hello.txt?" - "Create a folder called notes and put a file in it called ideas.txt with three random project ideas" - "List all the files in this folder and its subfolders" Don't worry about understanding HTML or folder structures. Just notice that Claude does what you ask. You describe, it executes. When something isn't right, you say so, and Claude adjusts. ## My First Session First time I used Claude Code, I asked it to make a simple HTML page. Dark background, some text, nothing fancy. It created the file. I opened it in my browser. There was a web page. I said "make the text bigger and add my name at the top." Five seconds later, updated. I said "add a section listing three things I'm good at." Done. Thirty minutes in, I had a basic personal page. Not because I learned HTML. Because I learned to describe what I wanted clearly enough for Claude to build it. That's the skill this guide teaches. Not coding. Communicating. > Next up: how Claude Code sees your files. Understanding this changes how you work with it.

Your First Conversation with Claude Code

Starting Claude Code

Sign In to Unlock

Check Your Email