With desktop apps this good, when is the terminal still worth using?

Desktop apps handle most AI work. Still, there are situations where seeing every step matters, and knowing how to spot them helps you avoid both too much and too little oversight.

For example, you might move automated tests from GitHub Actions (GitHub's service for running tasks in the cloud, which charges for usage time) to computers in your own office. With the terminal (using Warp, for example, which can access other machines on the network), you can see every incoming request and make sure each task runs on the right computer.

In a typical build, you don't need to follow every line of code. It's enough for the result to work and the right tests to pass.

No one needs to copy someone else's setup. Everyone will have their own mix of tools; what matters is that they work together.

What the terminal shows

The terminal (the text-based command screen) shows what’s happening in real time: each incoming request, each error, and which computer is running each task. A desktop app usually summarizes this and gives you only the result.

When the problem is in the process rather than the result, that detail helps you find the cause.

Signs the terminal is useful

It’s useful when the work involves several computers, something fails without a clear message, there’s a cost based on usage time, or you need to confirm where each task is running.

For example, moving GitHub Actions automated tests to your office’s own computers: you need to see whether each task reaches the right computer and whether anything keeps running—and costing money—in the cloud.

When the app is enough

For most builds, following every line doesn’t help. The real test is whether the result works and the right tests pass. The right tests check what matters to the user, not just whether the code compiles.

If you trust the tests, you can trust the result without watching the process.

Choose the combination that works for you

There’s no ideal setup to copy. One person uses apps for everything and the terminal only for debugging; another does the opposite. What matters is that the tools exchange information without you copying and pasting between them.

1. Ask whether the problem is with the result or the process that led to it.
2. Use the app when checking the result and tests is enough.
3. Open the terminal when there are several computers, time-based costs, or unexplained failures.
4. Confirm that the tests cover what matters before relying only on the result.
5. Go back to the app once the process is stable.

• Following every line in every project: it takes attention without improving the result.
• Trusting only the app’s “it worked” message for a task with usage costs: something may still be running and charging you.
• Copying someone else’s setup: it solves their problems, not yours.

• Terminal: when you need to see every step.
• App: when the result and tests are enough.
• Tools that work together matter more than the “right” interface.

• CourseDo chat ao agente, degrau 1 — o terminalFolders, rules, secrets, Git, and VPS basics for people who’ve never used the terminal.
• CourseCLI-x — o terminal como interface dos agentesWhy the terminal became the entry point for agents.
• CourseCodex BásicoCodex in the terminal and graphical interface, from beginner to advanced.

Some of these materials are in Portuguese.