You built it with an AI tool, so you didn't write the code and nobody has checked it. The deep scan reads up to 150 of your files, names every risky package, shows who can see your data if you use Supabase, finds similar projects you can learn from or partner with, checks how easy your app is to change (Code fitness), and gives you fixes you can paste straight into Lovable, Replit, Cursor or Bolt.
plus applicable tax, calculated at checkout for your location. One payment, no subscription, and re-scans of the same app for 30 days are included.
The problems that matter most come first. Each one has a plain-English explanation of why it matters and a ready-to-paste prompt for your AI builder, with a Copy button. You don't need to understand the code to act on it.
Your app is built on dozens or hundreds of packages written by other people. The report names each one that has a known weakness, with its exact version and vulnerability ID, plus one prompt per package and one prompt to update them all.
Verilay reads your database rules and backend functions and lists, in plain English, where anyone on the internet could read or change something they shouldn't, with a fix for each. The free scan only shows a count. (Settings changed only in the Supabase dashboard can't be seen from your files.)
Up to 6 free, open-source projects that do something like your app. Each one comes with why it's similar, what it's best at, its licence and when it was last updated, so you can see what already exists, borrow good ideas, build on it, or get in touch with the people behind it about working together. (Check a project's licence before reusing any of its code.) These are free GitHub projects only, not commercial apps, so no match never means nothing similar exists.
Counted from every file in your app, not just a sample. It starts with things you’ve probably noticed (“fixed a bug and it came back?”, “the AI said done but nothing changed?”) and shows the reason in your code: copy-pasted code, giant files, a price or limit typed in many places, leftover test screens. Then whether it needs tidying at all. Verilay checks your code; it doesn’t change it.
150 files instead of 25, so less of your app is left unchecked. If Verilay couldn't read part of your app, the report says so instead of showing a green tick. (If your app has fewer than 25 files, the free scan already reads all of it.)
After you fix something, scan the same app again as often as you like for 30 days and watch the problem disappear. Every report is saved in your account.
| Free | Pro free during the launch (until 30 Nov), then A$4.99 a scan | Deep scan | |
|---|---|---|---|
| Files read | Up to 25, the ones most likely to matter | Up to 50 | Up to 150 |
| The six areas (login, settings, data, API, pages, packages) | Short plain-English notes | Full explanations | Full explanations, every file read |
| Exposed keys, risky packages, who can see your data | Checked | Checked | Checked, with each package and its fix |
| Fix prompts for your AI builder | One, for the most serious problem | One for every finding | One for every finding, plus one prompt to update every package |
| Similar apps | The closest open-source match | The closest open-source match | Up to 6 to learn from or partner with |
| Code fitness (is your app getting hard to change?) | Your app’s size, for big GitHub apps | – | The full check, and whether it needs tidying at all |
| Checking a fix worked | 3 free scans a day | One free Pro scan per email during the launch; then pay per scan | Re-scans of the same app for 30 days |
| Best for | A quick health check | Fixing what the free scan found | Before launch, or a full review |
The free analysis stays free: every file is still checked for exposed keys, risky packages and who can see your data, and you get a fix prompt for the most serious problem. Pro adds a ready-to-paste fix prompt for every finding. During the launch, Pro is free: answer 4 quick questions after a free scan to unlock one Pro scan.
Public repositories only for now.
Stripe handles payment on its own page. Verilay never sees your card.
Most scans take a few minutes; bigger apps take longer. Close the tab if you like: the report is saved to your account and we email you when it's ready.
Honest by design. We build apps with AI tools too, and we run Verilay on our own apps first. If part of your app couldn't be read, the report says so. And if a deep scan fails to run or gives no usable report, you get a full refund.
Payment is handled by Stripe on their own page — Verilay never sees your card. Already bought one? Sign in.
Want your app checked every week? Register interest in monitoring.
No. It only reads your public repository and never writes to it. You decide which fixes to apply, using your own AI builder.
No. It's a thorough automated first pass written for non-developers. AI findings can vary between runs and it can miss things. See the terms and the AI disclaimer.
No. Every finding is explained in plain English, and the fixes are written as prompts you paste into your AI builder.
A deep scan is tied to one repository, including its re-scans. Another app needs its own deep scan.
If a deep scan fails to run or produces no usable report, you get a full refund. See the terms.
Not a penetration test. See the terms and the AI disclaimer for what this is and is not.