Real measurements,
openly scored.
Every number on this site is an observed measurement at scan time — what AI engines can actually read on a page, what we checked, what we found. Never a verdict, never a guarantee. Here is exactly how the scores work.
Get Found — the AI-readability score
We fetch one page the way an AI crawler does and score 9 signals, weighted by how much they decide whether ChatGPT, Claude or Perplexity can read, cite and recommend the app. The weights sum to exactly 100. The hardest, rarest signals carry the most points — a pretty page with none of them scores low, honestly.
The curve: scores up to 70 are taken as measured; above 70 we compress the top end, so high-80s and 90s are rare and earned — only a page passing everything scores 100. Grades: A ≥ 90 · B ≥ 72 · C ≥ 55 · D ≥ 35 · F below.
Convert and Ship Verified
Convertscores what decides a signup — a clear first-screen explanation, one obvious next step, signs of life, signup friction, mobile readiness, page weight — and runs a 5-second first-impression test: a ChatGPT-style AI plays a brand-new visitor and reports what it thinks the app does and whether it would sign up. That read is always labelled as a model's observed first impression, never a prediction. Ship Verified checks a live app for the leaks vibe-coded apps ship most — exposed passwords and keys, public private-config files, missing security protections, over-broad cross-site access — and reports exactly what was checked and what was found. A clean result is a record of checks at scan time, not a safety verdict.
The benchmark dataset
Every day we scan a fresh batch of public apps discovered from public launch lists and public web records, and record the results. Medians and percentiles always use the newest score per app — an app re-scanned fifty times still counts once. Below a minimum sample we say “still building” instead of showing a number. Vibe-coded apps and recognizable mainstream brands are tracked as separate categories and never mixed into each other's medians. We run the same scan on ourselves and publish the result on the homepage — including the points we lose.
Live outputs: the leaderboard and the state-of-vibe-coding study, both rebuilt from the dataset every day.
Run it on your app.
All 9 signals, scored live — free, no signup.