Trust QA for AI-built websites

Find launch trust issues before your AI-built site goes live.

TrustDebt checks AI-built launch pages for consent, privacy, accessibility, legal-link, tracker, and security-header risk before founders buy ads or agencies hand off client work.

Risk detection, not legal advice AI fix prompts Saved reports

trustdebt.dev/report

25Trust score

5Critical

3Warnings

8Fixes

url your-ai-built-site.com

tracker or consent risk flagged

privacy and launch proof incomplete

AI remediation prompt ready

Browser evidenceTracker and consent behavior

5 pillarsPrivacy, a11y, legal, trackers, security

Monthly quotaUsage tied to verified email

Fix promptReady for Claude Code or Cursor

Browser evidence scanner

Run a launch trust audit.

TrustDebt opens the live page, captures tracker and consent evidence, checks HTML and headers, then saves reports to your verified workspace.

01Target page

Website URL

Optional pasted homepage HTML

Launch report

https://your-ai-built-site.com

Generated after scan

25Grade F

Accessibility2

Privacy1

Trackers2

Security1

Launch proof2

Trackerscritical

Cookie banner may be decorative

You show consent UI, but analytics or ad pixels may still load before a visitor accepts.

Gate all non-essential scripts behind an explicit consent.accepted state before injection.

Trackerscritical

Advertising pixel increases privacy risk

Meta, TikTok, Bing, and retargeting pixels are common sources of pre-consent tracking mistakes.

Load ad pixels only after consent, document them in the privacy policy, and test reject-all behavior.

Privacycritical

Collects email without privacy policy

Email capture, waitlists, analytics, and forms need a visible privacy policy before launch.

Publish a privacy policy that explains data collected, purpose, processors, retention, and contact path.

Launch proofwarning

Terms link missing

For SaaS, checkout, waitlists, or paid audits, missing terms creates avoidable launch risk.

Add terms of service and link them in the footer and checkout flow.

Accessibilitycritical

Image alt text not verified

AI-generated landing pages often look polished while missing basic non-text content alternatives.

Audit every image. Use descriptive alt text for informative images and empty alt for decoration.

Accessibilitycritical

Email/form labels not verified

Placeholder-only fields are not sufficient labels and can fail accessibility checks.

Use visible labels or properly connected aria-label/aria-labelledby values for each input.

Securitywarning

Security headers not verified

Launch pages should verify baseline headers before collecting leads or payments.

Set Content-Security-Policy, X-Frame-Options or frame-ancestors, Referrer-Policy, and Permissions-Policy.

Launch proofwarning

AI-generated copy not reviewed

AI site builders can hallucinate guarantees, legal claims, testimonials, or compliance promises.

Review claims manually. Remove unverifiable guarantees and any legal/compliance promise you cannot prove.

AI remediation prompt

You are fixing launch trust issues for https://your-ai-built-site.com. Do not redesign the site. Patch only trust, accessibility, privacy, tracker, legal-link, and security-header problems.

Issues to fix:
1. [CRITICAL] Cookie banner may be decorative: Gate all non-essential scripts behind an explicit consent.accepted state before injection.
2. [CRITICAL] Advertising pixel increases privacy risk: Load ad pixels only after consent, document them in the privacy policy, and test reject-all behavior.
3. [CRITICAL] Collects email without privacy policy: Publish a privacy policy that explains data collected, purpose, processors, retention, and contact path.
4. [WARNING] Terms link missing: Add terms of service and link them in the footer and checkout flow.
5. [CRITICAL] Image alt text not verified: Audit every image. Use descriptive alt text for informative images and empty alt for decoration.
6. [CRITICAL] Email/form labels not verified: Use visible labels or properly connected aria-label/aria-labelledby values for each input.
7. [WARNING] Security headers not verified: Set Content-Security-Policy, X-Frame-Options or frame-ancestors, Referrer-Policy, and Permissions-Policy.
8. [WARNING] AI-generated copy not reviewed: Review claims manually. Remove unverifiable guarantees and any legal/compliance promise you cannot prove.

After patching, summarize changed files and what to manually verify before deploy.

Download report See paid report options

How it works

From AI-generated page to deploy-ready proof.

browser evidence

openChromium loads the public URL

watchEvery request is classified by host and type

rejectConsent controls are tested when visible

scoreFindings become a saved launch report

Scan

Open a live URL in Chromium, intercept network requests, and inspect returned HTML, links, forms, tracker hints, and response headers.

Fix

Get issue-level instructions and an agent prompt for Claude Code, Cursor, or OpenCode.

Prove

Save the report to your dashboard, export the result, and share a client-ready summary from the paid tiers.

Productized wedge

Built for builders, not compliance departments.

Fake banner detection

Catch sites where the banner exists but trackers still fire before consent.

Accessibility triage

Find missing alt text, unlabeled forms, placeholder-only fields, and launch-blocking a11y debt.

Tracker inventory

Flag analytics, ad pixels, fingerprinting hints, and undocumented third-party scripts.

Security header checklist

Verify baseline launch headers before collecting leads or payments.

Founder PDF brief

Turn the latest scan into a polished audit packet with checklist and score snapshot.

Coming next: monitoring

Scheduled scans, regression alerts, and deploy checks are labeled as upcoming Studio capabilities.

Verified scan history

Your last scans stay in this browser.

Scans are tied to a verified email so free, founder, and studio usage can be limited cleanly each month.

Verified email Monthly reset Saved report

No saved scans yet. Run your first audit above.

Pricing

Start with verified scans. Upgrade for client-ready proof.

Credible v1 packaging for founders validating one launch and agencies managing repeat client work.

Free

3 scans/month

For validating a single launch page

Signed-in browser evidence scan
Header and HTML checks
AI fix prompt
Text report download

Start free scan

Best first offer

Founder Audit

$29

25 scans/month

For turning a launch audit into a proof packet

PDF audit brief
Manual review checklist
Priority fix prompt
Badge-ready score snapshot

Request Founder Audit

Studio Preview

$49/mo

250 scans/month

For agencies and repeat launches

Multi-site dashboard
Client-ready CSV export
Workspace history
Coming next: scheduled scans

Preview Studio workspace

FAQ

Trust monitoring without fake legal promises.

No. TrustDebt detects risk and produces remediation guidance. It is not legal advice and does not certify full compliance.

AI site builders make launch pages faster, but they also make it easy to ship fake banners, missing labels, hallucinated legal claims, and forgotten trackers.

The v1 scanner captures browser evidence when available, checks HTML and headers, saves reports, and exports text, PDF, and CSV outputs from the dashboard.

Founders before launch, micro-SaaS builders, Webflow and Framer agencies, AI site builders, and dev shops that want client-ready trust reports.

Launch safer

Before your AI agent ships the site, run a trust agent against it.

Use the signed-in scanner today. Founder reports and Studio exports turn the evidence into proof you can act on.

Request access