From URL to fix list, automatically.
Give Veylant a domain. It visits your funnel the way a prospective patient would, captures what it sees, audits it with four engines and hands back evidence your team can verify.
Crawl like a patient would.
A real headless browser loads your homepage, then follows links to the routes that matter most for compliance: treatment, intake, quiz, checkout and legal pages.
- Renders JavaScript, so it sees what visitors see
- Same-origin only: it never wanders onto other sites
- Polite timeouts and a small page budget per scan (up to 11 pages)
Capture everything.
For every page, Veylant saves a full-page screenshot, the rendered page text and every network request the page makes, including each third-party tracker.
- Full-page screenshots for visual review and evidence
- Rendered text, not just raw HTML
- Complete network log with hosts, parameters and the page that fired them
| GET | connect.facebook.net | /intake/step-2 |
| POST | analytics.tiktok.com | /semaglutide |
| GET | www.googletagmanager.com | / |
Four audit engines.
Each engine looks for a different kind of risk. Deterministic rules handle what can be measured exactly; AI handles language and layout, then gets checked.
Pixel & privacy
Flags Meta, TikTok, Google/GA4, Snap, Reddit, Bing, Pinterest and Google Ads requests on intake and checkout pages, and health keywords or contact details inside tracker payloads.
FDA & peptide claims
Reviews copy for branded-drug equivalence, implied FDA approval, guaranteed outcomes and missing compounding disclosures.
FTC / ROSCA checkout
Reviews checkout and intake screenshots for auto-renewal terms, cancellation language, pre-checked add-ons and billing before consultation.
LegitScript-style disclosures
Checks for pharmacy, clinician-oversight, state-availability, privacy and terms disclosures.
Evidence you can verify.
Every AI finding must quote copy that actually appears on the page. Veylant checks each quote against the rendered page text, and if it isn’t there, the finding is discarded.
- Exact quote, page URL and screenshot for each finding
- Bounding boxes on checkout and intake screenshots
- Rule code and plain-language reference for review
Score, fix, monitor.
A 0–100 risk score summarizes the scan. Findings are ranked by severity with copy-paste remediation. Scheduled monitoring (rolling out now) re-scan automatically, compare against the previous scan and alert you to new critical or high findings.
- 0–100 risk score with Low / Moderate / High / Severe bands
- Remediation text you can hand to your team
- Daily or weekly re-scans with diffs and alerts (coming soon)
What Veylant doesn’t do
Clear limits are part of how we earn trust.
- Doesn’t submit forms or create accounts on your site
- Doesn’t access patient data or anything behind a login
- Doesn’t give legal opinions: findings are automated risk indicators
- Doesn’t guarantee regulatory, platform or legal outcomes
See it on your own funnel.
Run a free scan of your homepage, treatment, intake, checkout and legal pages.