Knowledge · IT · Data · DX

Terminal · CLI

Terminal / Command Line Interface

📂IT·Data·DX⏱2-min read📊Beginner🔗5 related

What is a terminal?

In one line
The black window (terminal) and method (CLI, Command Line Interface) for controlling a computer by 'typing commands' instead of clicking with a mouse.
Bring this topic to your team📚 Related Courses💬 Ask About Training
⚡ Summary in 3 lines
  • The black window for controlling a computer with commands
  • Text commands instead of clicks
  • The foundation of automation and development

Once you start using the terminal, work becomes text instead of hand movements. A setting you configured with ten mouse clicks has to be explained again, out loud, to the colleague next to you, but a single command can be copied as is, pasted into a chat, saved in a document, and run again tomorrow. Being able to reproduce the same result identically, any number of times, is what sets it apart from clicking, and it is why install, build, and deployment tools all run on the command line. Mistakes don’t vanish with the screen either; they stay on record, which makes them easy to trace.

Separating three things clears up most confusion: the terminal is the window, the shell is the interpreter, and commands are what you actually tell it to do. Macs default to zsh and Windows to PowerShell, so the wording for the same task differs slightly, and pasting a Linux command straight into Windows produces an 'unrecognized command' error. That said, the commands non-developers actually use in vibe coding come to about ten: checking your location (pwd, cd), installing (npm install), running (npm run dev), and version control (git), and there is no need to use the command line for tools that have their own installer.

The most common accidents come not from fear of the black window but from pasting without reading. People run a command from an AI or a blog as is, and a sudo privileged command changes system settings, or files deleted with rm are gone for good, with no trash to recover them from. The second is getting your location wrong. Someone types an install command on the desktop and believes it installed into the project folder; half of the 'I definitely installed it, but it’s not there' questions come from this. The third is the habit of skipping past a red error message to the next step. Terminal errors usually state the cause almost verbatim, so it’s much faster to first ask 'What does this command change?' before running it, and when an error appears, copy the whole message and ask about it.

Courses that cover this term · AI & DX
Recommended courses · Recommended instructors · Quick quote
🧮 Mini quote · Based on instructor fees
About KRW 1.2 million (≈ US$900)2-hour talk · up to 20 people · Korean
Get a quote for these requirements →
💡 In plain English
Instead of clicking with a mouse, you give instructions in words (commands). Once you’re used to it, it’s much faster and easier to automate.
Why it matters in practice
Most development tools run in the terminal. Even the basics will keep you from getting stuck as often when vibe coding.
Origin and sources

Before graphical interfaces (GUIs), the command line was the original way to operate a computer. It is still widely used in development and server administration as a precise, automatable tool.

An example

Consider a hypothetical case. Company A, an industrial parts distributor with 210 employees, decided to turn the inventory lookup its sales support team ran in Excel every day into an internal web page. An AI code editor produced the screen in 30 minutes, but then three days were lost on simply 'running the server'. None of the 6 people on the team had any development experience.

  1. On day one, a junior manager pasted a command from a blog as is, and 40-some files spilled onto the desktop. When the team lead said, 'Type pwd first,' it showed that the current location was outside the project folder.
  2. On day two, the team cut the commands they use down to 6 and pinned them in Notion: cd, pwd, npm install, npm run dev, git status, git push. They agreed to run any other command only after asking an AI what it does.
  3. On day three, when the installation kept failing, a senior manager copied all 12 lines of red error text, pasted them into an AI, and got the answer, 'Your Node version is below 18,' within 3 minutes, solving it on the spot.
  4. They set a rule: no command containing sudo runs without the team lead’s sign-off. It paid off when a database reset command was suggested once; the review step caught it and prevented the loss of some 30,000 production records.
  5. Four weeks later, the sales support team was even handling deployment on their own, without developer help. Consolidating inventory, which used to take 50 minutes a day, dropped to 5 minutes, and terminal-related questions fell from 27 in the first week to 2 in the fourth.
Related terms
Training linked to this term
📚Digital TransformationNanum Consulting corporate training · In-person / OnlineView Course →
Need in-house training on this topic?
Get a curriculum and quote tailored to your needs right away.
Ask About Training
Last updated 13 Sep 2026 · Reviewed by John Kim

FAQ

No. You only need to know a handful of frequently used ones, and you can ask an AI about the rest and run them.
Commands like delete call for caution. If you don’t know what a command does, it’s safest to have an AI explain it before you run it.
A handful of frequently used commands is enough to get started. These days people also ask an AI for the commands.
Be careful with commands such as delete. It is important to build the habit of never running a command you do not understand.