For Lovable · Bolt · Replit · Cursor · v0 apps

Your AI built it.
Latch it before you ship it.

Free security scan for vibe-coded apps. Find leaked API keys, exposed .env files and risky Supabase setups, then get copy-paste fix prompts for your AI builder. Fix it and prove it.

Passive and read-only: we load your page like a normal browser. No signup. Deeper database tests run only after you verify ownership.
LovableBolt.newReplitCursorv0SupabaseFirebase

What the free scan checks

🔑 Leaked secrets

OpenAI, Anthropic, Stripe, AWS, GitHub and more in your HTML and JS bundles, plus Supabase service_role keys that bypass all security.

📂 Exposed files

Is your /.env or /.git folder downloadable by anyone?

🗄️ Backend exposure

Detects Supabase and Firebase so you know what to lock down. Live RLS testing comes after ownership verification.

🛡️ Security headers

CSP, HSTS, clickjacking, nosniff, referrer policy, CORS and source maps.

Why it matters

170 / 1,645

Lovable projects found with exposed data in one 2025 scan (CVE-2025-48757).

400+

secrets exposed across 5,600 vibe-coded apps (Escape.tech).

45%

of AI-generated code samples had security flaws (Veracode).

Fix it and prove it

01 · SCAN

Paste your URL

Grade and issue counts in seconds.

02 · FIX

Copy-paste prompts

Plain-English fixes and prompts written for your AI builder.

03 · PROVE

Re-scan and badge

Re-scan to confirm each fix, and earn a "Latched" badge to show your users.

Pricing

Free scan

$0
  • Passive scan of any URL
  • Grade and full findings by email
  • Fix prompts

Fix Pack coming soon

$29
  • Ownership-verified deep scan
  • Live Supabase RLS and Firebase rules tests
  • 2 re-scans and Latched badge

Stay Latched coming soon

$15/mo
  • Weekly monitoring
  • Alerts when the AI re-breaks something
  • Badge stays verified