SECURITY FOR WHAT YOU SHIPPED / EST. FOR THE AI BUILD ERA

Your AI-built app is live.
Is your database open?

We audit apps made with Lovable, Cursor, v0 and Bolt. Every finding is verified by a second independent pass before it reaches you.

Checks: TLS · security headers · CORS · exposed files (.env, .git) · secrets in the JS bundle · public database probes · email records.
No logins, no data extraction, nothing stored from the target site.

45%

of AI-generated code samples introduced a security flaw.

VERACODE · 2025 GENAI CODE SECURITY REPORT
2,000+

high-impact vulnerabilities found in 5,600 live vibe-coded apps.

ESCAPE.TECH RESEARCH · 2025
10.3%

of 1,645 Lovable apps in one study had their database access wide open.

LOVABLE APPS STUDY · 1,645 APPS

01 — SCOPE

What we check.

Start at the surface. Go deeper when you need to know what happens after someone signs in.

FREE SURFACE SCAN 01FULL AUDIT 02
Security headers & TLSAuthenticated testing with two test accounts
CORS configurationIDOR & access control
Exposed filesPrivilege escalation
Secrets in the JavaScript bundlePayment webhooks
Public database probesDependency CVEs & git-history secrets
Email recordsPrompt injection
NO ACCOUNT ACCESS REQUIREDINCLUDES THE SURFACE SCAN

02 — METHOD

Find it.
Verify it.
Fix it.

A scanner can make a claim. A second independent pass checks whether it holds up.

01

Surface scan

Inspect what is exposed before signing in.

02

Authenticated testing

Test what two accounts can see and change.

03

Independent verification

Check each finding in a second pass before reporting it.

04

Report with proof & fixes

Get evidence, impact, and prompts you can use to fix each issue.

05

Retest within 30 days

For Full and Deep audits, we check the reported fixes again.

03 — THE DELIVERABLE

Proof you can
act on.

See what a finding looks like: the exposure, the evidence, and a fix prompt written for your builder.

Read the sample report ↗
SECURITY AUDIT / SAMPLEVF—001
VERIFIED

AUDIT REPORT

mealprep-app

APPLICATION SECURITY REVIEW
FICTIONAL SAMPLE · 7 FINDINGS

01Open database tableCRITICAL
02Service-role key in bundleCRITICAL
03Login rate limit missingHIGH
VIEW FULL REPORT ↗

04 — PRICING

Pick the depth
you need.

One-time audits for a clear picture. Monitoring for a recurring look at the surface.

01 / START HERE

Free scan

For a first look at your public surface.

Public surface checks
Public database probes
$0
02 / SURFACE

Quick Audit

For founders who need a verified surface report.

Surface report
Human-verified
3 days
$299 ONE-TIME
04 / HANDS-ON

Deep Audit

For teams who want help supervising the fixes.

Everything in Full Audit
Fix supervision
$2,999 ONE-TIME
05 / ONGOING

Monitoring

For founders who want to revisit the public surface.

Weekly rescan
Alerts
$79 / MONTH

05 — QUESTIONS

The details,
plainly.

Is it legal to scan an app?

Only scan a site you own or have explicit permission to test. You must confirm permission before starting.

What do you need from me?

Your app URL and, for authenticated testing, two test accounts with appropriate roles. We’ll agree on the scope before testing.

How long does an audit take?

A Quick Audit takes 3 days. Timing for Full and Deep audits depends on the scope and is agreed in the brief.

What is in the report?

Each finding includes severity, business impact, evidence, and a practical fix. Findings receive a second independent verification pass.

How does the retest work?

Full and Deep audits include a retest of the reported fixes within 30 days.

What will you never do?

We do not touch production data. The free surface scan does not log in or extract data.

AUDIT INQUIRY / 01

Tell us what
you built.

Share the basics. We’ll reply within one business day.

REPLY TO: