This page was translated from Dutch with AI.

Security belongs in your briefing, not at the end

Some vibe coders tell me they built three apps today. Then they must be one hundred percent sure those apps are secure too. I say that in every building training. And then I always see a few people swallow hard. Making an app work has actually become the easy part. You do not get a secure app automatically.

Below this article you will find the checklist for building securely with AI, free for Members.

Cartoon
Veiligheid hoort in je briefing, niet op het einde: veilig bouwen met een AI-agent

Why afterwards does not work

The classic reflex is: build first, and if it works, we look at security. If you build with AI, that is extra risky. The agent makes choices about the architecture at every step, based on what it knows. If it does not know that security is important, it chooses the fastest route. And that fastest route is often difficult to change afterwards.

So security belongs in your briefing. That is the document with the requirements you use to start the build. I would put in it that the app must be resistant to hidden instructions in the input. And that secrets are stored encrypted and personal data is protected. And that the app must pass a vulnerability test. An agent that knows this from the start builds differently.

Free for Members: Checklist, building securely with AI before your app goes online

Frequently asked questions

Why must security be in the briefing when I build with AI?

An agent makes choices at every step based on what it knows. If it does not know that security counts, it chooses the fastest route. That is often difficult to change afterwards.

How do I store passwords and keys securely in an app that AI builds?

Agents often store them in a plain text file in your project by default. I would just ask your agent to store secrets encrypted, never in plain text, and never to put them online. Most agents know perfectly well how to do that.

How do I test if my app with AI is secure enough?

Ask a separate agent in a new session to attack your app and write a report. A second agent solves the problems. Then test again, preferably with a different model, until the report is empty.

Where can I practice building securely with AI hands-on?

In Vibe Coding with AI: building applications without programming knowledge you build with your own idea. If you are already building and want to push on for two days, there is the AI Builder Hackathon. You can find all other courses at Workshops & Events.

Keep reading for free

Register free with your e-mail address and read on. Your account opens the free articles in the learning space; the rest is for subscribers, from Learn Plus up.

Already registered? No account yet?

No comments yetRead the comments

Comments

Members comment here and read each other's comments. Register for free or sign in to join the conversation.

Register for free or sign in

No comments yet. What do you take away?

    QR code of this page

    coachsteff.live/en/veiligheid-vanaf-de-briefing

    Scan with your phone to open the page.