For apps built with Cursor · Lovable · Bolt · Replit

You shipped it with AI.
We check it didn't ship your secrets.

LaunchMedic checks your live app for leaked API keys, publicly readable databases, exposed .env files and broken signup or checkout pages — then gives you a copy-paste prompt to fix each issue in your AI coding tool.

Run your free scan →

No card needed. Verify you own the domain, get a full report in about a minute.

What we check

Leaked API keys

We read your public JS bundles the way an attacker would and flag Stripe, OpenAI, Anthropic, AWS, Supabase service_role and 15+ other secrets.

Open Supabase tables

Using only the anon key your site already ships, we count which tables anyone can read. Row counts only — we never read your data.

Exposed .env & .git

Checks whether .env, .git/config and other config files are downloadable from your domain.

SSL & security headers

Certificate expiry, HTTPS redirect, HSTS, CSP, clickjacking and more.

Uptime

Is your homepage up and fast? Know before your users tell you.

“Still works” check

Your signup, login and checkout pages must load without errors — catch AI-edit regressions early.

How it works

1. Paste your URL

Add the app you deployed on Vercel, Netlify, Lovable, Replit or anywhere else.

2. Prove it's yours

Add a DNS TXT record, a meta tag, or a file in /.well-known. We only ever scan verified domains.

3. Fix with one paste

Every finding comes with a prompt you paste into Cursor, Lovable, Bolt or Replit to fix it.

Pricing

Free first scan

$0 once
  • One full check of one verified app
  • All 6 checks + fix prompts
Start free scan

1 check

$9 one-time
  • One full check of one verified app.
  • All 6 checks + fix prompts
  • Emailed report
Buy 1 check

Pay per check — no subscription. Checks never expire. Prices in USD, plus sales tax where applicable.

Safe by design

Verified domains only

No verification, no scan. We can't be pointed at someone else's site.

Polite & read-only

~40 requests per scan, rate-limited, clearly identified user agent. We never write, log in, or exploit anything.

We don't keep your secrets

Found keys are masked before they're stored. File contents and database rows are never saved.