getAdvantage
Log inBook a callBook
Anonymized · built from real scans

Shipped at machine speed. Checked by no one.

AI lets anyone ship at machine speed. Here is what that work looks like once someone checks it: how many public apps skip basic security protections, how many ship with private files exposed, and how many an AI engine cannot even read.

308
public apps measured
44%
skip basic security protections
0%
ship a private file to the public web (.env / .git)
63%
give AI no facts to read (no JSON-LD)

Ship Verified

221 apps · median 84/100
Missing basic security protections (headers)
44%
The good news, found in none of the measured apps: data open to other websites (over-broad CORS) · a server password or key visible in the public code · no HTTPS encryption · source code readable by anyone (public source maps) · a private config file served publicly (.env / .git) · backend settings exposed to visitors.

And AI cannot even read most of it.

Get Found

308 apps · median 48/100
No AI intro file (llms.txt)
65%
Page that reads nearly blank to AI
64%
No AI-readable fact sheet (JSON-LD)
63%
Messy heading structure
47%
No link-preview tags (Open Graph)
21%
AI crawlers blocked from reading the site
10%
No page description for AI and search
10%
No clear page title
1%
The good news, found in none of the measured apps: no canonical URL.

Convert

223 apps · median 53/100
No clear next step for visitors (CTA)
79%
No sign anyone else uses it
75%
No clear first-screen explanation of what it does
14%
Slow, heavy first load
10%
Unclear headline
2%
The good news, found in none of the measured apps: a page not set up for phones · a sign-up that asks too much too soon.

What they're built with

Which builder each app was made with, detected from public clues in the app itself, never guessed.

66Lovable
36Base44
36Next.js
33WordPress
23Replit
15v0
14Vite
12Bolt
11Webflow
7Framer
3Shopify
1Squarespace

of 308 measured apps · only fingerprinted stacks shown

Generated July 27, 2026 from 308 measured apps, all vibe-coded, the population every number above refers to. For contrast: the 13 known mainstream brands we measure separately score a median of 65/100 on AI-findability, tracked apart, never mixed in. Each percentage is the share of apps where we observed that specific issue at scan time, a measurement, not a verdict or a guarantee. Not every engine completes on every app, so per-section sample sizes vary. Apps are discovered from public launch lists and public web records; nothing here identifies an individual app.

Is your app in the majority?

We run the same three engines on your own app, then walk the findings with you. Book a call and find out.

Book a call →