Memuat
Memuat
Want to go deeper?
Join one of our courses or ask us directly on WhatsApp.
Picked from the software, topic, and level of “Claude from Zero to Expert Class #73: Your First Commands in the Terminal”.
Other ways we can help with Claude.
Learn with a mentor
Scheduled AI & Machine Learning courses with live guidance and a certificate.

Midjourney Course: Create Stunning AI Images from Text
★ 4.9

DALL-E Course: Creating Professional AI Images from Text
★ 4.9

Stable Diffusion Course: AI Art from Beginner to Professional
★ 4.9
Have it built for you
No time to learn Claude? Our team can do the project for you.

AI Image Creation with Stable Diffusion — Unlimited Custom Visual Production
★ 4.9

AI Image & Video Creation with Leonardo AI — All-in-One Visuals for Business
★ 4.9

AI Image Creation with Midjourney — Cinematic AI Art for Business
★ 4.9
Get notified about new Claude classes
At most one email a week. Unfollow anytime with one click.
The full narration of this video. The current sentence is highlighted while the video plays; click any sentence to jump there.
Hi, and welcome back. Claude Code is installed on the computer. Now it's time to give our first real instructions. In this episode we recognize the parts of the Claude Code screen, create and change files, stop Claude when it heads the wrong way, and use the most useful slash commands.
Before giving instructions, get to know five parts of the screen. First, the prompt box at the bottom. This is where we type our request, then press Enter to send it. Second, Claude's replies. Each reply starts with a white dot and contains an explanation, a plan, or a summary of the work. Third, tool lines with a green dot. Read means reading a file, Write creates a new file, Edit changes part of a file, and Bash runs a computer command. Fourth, the orange-bordered permission box. When this box appears, Claude stops and waits for our choice. Read it before choosing. Fifth, the status line at the very bottom. On the left is the active folder name, and on the right is a help hint. This screen is a simulation; details may differ slightly because Claude Code is updated often.
Raka starts with a simple instruction: create a menu file with three Sunset Coffee products, complete with sizes and prices. Notice that Raka writes out all the data, so Claude doesn't have to guess. Claude shares its plan first: one text file called menu, with three lines for the product name, size, and price. Then a tool line appears. Write means Claude is writing a new file. Below it, we see that the file has three lines. The permission box asks: create the menu file? That matches the request, so Raka chooses Yes by pressing Enter. Claude finishes with a short summary. Get into the habit of reading summaries like this, because this is where we learn what was actually done. Raka doesn't just take it on trust. Raka asks Claude to show the file's contents so each price can be checked one by one. Claude uses the Read tool. Reading doesn't change anything, so there's no permission box here. The contents appear neatly in three lines. Raka compares them with Laras's price list, and everything matches. Checking numbers like this is a must, especially prices that customers will see.
Now we change a file that already exists. Raka asks to add the gift hamper at one hundred fifty thousand rupiah, and adds a limit: don't change any other lines. This time the tool is Edit, not Write. Edit means changing part of a file, not creating a new file from scratch. Then a change box appears. Gray lines are old content that hasn't changed. Green lines marked with a plus are new additions. If there's a red line marked with a minus, something was removed. Raka reads the gist: just one green line and no red lines. That's what was asked for, so Raka chooses Yes. Claude confirms that no other lines changed. A small limit like don't change any other lines really helps keep the work under control. Compare a vague instruction with a clear one. A vague instruction like fix the menu makes Claude guess what we mean. A clear instruction says what to change. A clear instruction also says where the change goes, for example in the menu file, on the last line. Finally, state the limits. A short sentence like don't change any other lines prevents changes we don't want. These three elements, what, where, and limits, are ones we'll keep using throughout building the Sunset Coffee website.
Sometimes Claude gets a little too eager. Raka asks for product descriptions for the whole menu, without saying how many. Claude plans to write twenty variations for each product. That's far more than Laras needs. Raka presses the Esc key. The work stops immediately, and the terminal notes that the user interrupted it. Nothing is broken. Raka gives a new, more precise direction: one description per product, two sentences max, casual tone. Claude adjusts its plan. So don't hesitate to press Esc. Stopping early saves far more effort than cleaning up results that went the wrong way. Here are the keys to remember. Enter sends a command, and also picks an option in the permission box. Esc stops Claude while it's working, as we just saw. The up arrow brings back the previous command. Handy when we want to resend a command with a small change. And slash help shows the help list, including how to exit Claude Code. This list always matches the latest version, so it's more reliable than memorizing.
Commands that start with a slash are special Claude Code commands, not messages to Claude. Raka tries the first one: slash help. The terminal shows a list of commands. On your screen the list is longer and may be different, because it keeps being updated. The two beginners use most are help and clear. Raka is done with the menu and wants to switch topics, so Raka types slash clear to start a new conversation. The conversation is cleared, but the menu and description files are still safe in the folder. Clear only makes Claude forget the earlier chat, so it isn't confused by old context. With a fresh conversation, Raka tries an instruction that combines two steps: read the menu file, then create a welcome page with that menu list. Claude reads the menu file first. Because the conversation was cleared, Claude takes the data from the file, not from memory of the chat. Then Claude writes a page file called hello. The html extension means this file opens in a browser, such as Chrome or Edge. The permission box appears again. Raka reads the file name, makes sure no other files are touched, and chooses Yes. Claude suggests how to see the result: double-click the file in the folder, and the browser will open it. Here's Raka's first page. There's a welcome message and the opening hours at the top. In the right panel, Raka notes the check results: the welcome message and opening hours are showing. All four products appear with their prices. Raka compares them with the menu file, and they all match. It's simple, but this is a web page created purely from everyday-language instructions. Let's sum up the good habits. First, one instruction for one task. Break big jobs into several small instructions. Second, say what to change, where, and the limits. The clearer the instruction, the fewer revisions you'll need. Third, read the summary and changes before pressing Yes. This is the most important habit in the whole Claude Code session. Fourth, press Esc if it heads the wrong way, then give new directions. There's no need to wait for Claude to finish something we don't want. Fifth, use slash clear when switching topics. Try all these habits with your own business's product list.
We've recognized the Claude Code screen, created and changed the Sunset Coffee menu file, read the change lines, stopped Claude with Esc, and used the help and clear commands. In the next episode, Laras sends a folder of website materials. We'll ask Claude to read it, then go over what each option in the permission box means. See you there!
Try these on your own after watching to lock in what you learned.
Type a question or keyword; we find the part of this episode that explains it and jump the video there.
Notes are saved with the current video time and are only visible to you.
Stuck on a step? Ask here; the instructor and other learners can answer.
No questions yet.

Cerdas Inovasi Digital Team
Software developers, BIM & 3D engineers, and AI practitioners
We build software, BIM models, games and AI automations for clients every day, and turn that hands-on work into step-by-step classes: every step is shown on screen and comes with a PDF guide you can keep.
See our clients →