MBy what you built

Your vibe-coded app works for you. Does it work for anyone else?

AI tools get you from idea to deployed app in a weekend. They also build exactly the path you described and little else. Makers on the wall go through your app with their own data, their own devices and none of your assumptions, and write down where it falls apart.

Get your first testersSee what’s on the wall
  1. 1 · Post itPaste your link, say what testers should look at. It goes on the wall right away, free.
  2. 2 · Test oneClaim another maker's project and write down what broke. Each report earns 20 credits.
  3. 3 · Read yoursMakers test your project. Each report arrives locked with an excerpt; 20 credits unlock it in full.

What AI-built apps tend to miss

The happy path is usually solid, because it's the one you kept prompting about. The trouble is everything around it: the second user, the empty account, the form submitted twice, the phone in landscape, the session that expires mid-task.

Security is the quiet one. Generated back ends often ship with database rules wide open or API keys in the browser bundle, because nobody asked for anything else. Testers aren't a security audit, but they do notice when they can see someone else's data.

Prompting can't replace a stranger

You can ask the model to test its own code, and you should. But it tests what it thinks the app does. A person who has never seen it tests what it actually does, and tells you where they got lost, which no unit test covers.

What to put in your brief

  • The tool you built it with, so testers know where to look (Supabase rules, env variables, generated forms).
  • Two accounts or a way to create them, so testers can check one user can't see another's data.
  • The one task a new user must be able to finish.
  • What you already know is broken, so nobody reports it twice.

Projects open for testing right now

Real projects from the wall. Test one while you wait for your own reports.

See your listing before you sign up

Paste your link. We read the name, the icon and the cover, and show you the card other makers will see on the wall. You only create an account when you publish it.

Free. No card, no account yet.

Questions

Will testers judge me for using AI to build my app?
No. Plenty of projects on the wall are built that way, and testers care whether the app works, not how it was written.
Can testers check my app's security?
Not properly: they're makers, not penetration testers. They will notice obvious problems like seeing another user's data or secrets in the page source, and say so. For anything handling payments or personal data, get a real audit too.
Which AI coding tools are welcome?
All of them: Cursor, Claude Code, Windsurf, Replit, v0, Lovable, Bolt.new, or plain chat. What matters is that testers can open the app from a link.
Is Makers Wall really free?
Yes. There is nothing to pay and no card to add. You pay in tests instead: each report you write for another maker earns 20 credits, and 20 credits unlock a full report on your project.
Who tests my project?
Other makers with projects on the wall. They are good at spotting broken flows, confusing copy and bugs. They are probably not your target customers, so treat their reports as a sharp first pass before launch, not as market research.
What if a report is lazy?
You see an excerpt before you unlock it. Once it's unlocked you can accept and rate it, or request changes once with a message. Ratings follow testers around: a report that says "nice app!" earns a low one.

Put yours on the wall.

Free. Test another maker’s project, get honest reports on yours.

Get your first testers