Responsible Scanning Policy
Last updated 2026-11-17
This page explains what ScanAnt does when it scans a site, so that site owners know what to expect.
1. What a passive scan does
A passive scan fetches the public page you submit and the scripts, stylesheets and source maps it references, using ordinary GET requests, the same way a browser does. It analyses the responses for exposed secrets, missing security headers and similar issues.
It does not log in, submit forms, send data to your backend, query your database, storage or authentication endpoints, or use any credential it finds.
2. How we stay polite
Each scan has a small request budget, a size limit and a time limit, and runs at low concurrency against a single host. We rate-limit scans per requester and per target, and reuse a recent report instead of rescanning the same site within ten minutes.
3. How to recognise us
Our requests identify themselves with a User-Agent that names ScanAnt and points to this page, so you can find them in your logs.
4. Safeguards
We refuse private, loopback, link-local and cloud-metadata addresses, including after redirects. Secrets found in your code are masked in our reports and are never used or stored in full.
5. How to opt out
Use the opt-out page and tell us your domain and how to reach you. Every request is recorded and reviewed by a person; we may contact you to confirm ownership before adding the domain to our block list. You can also write to [email protected].
6. Reporting abuse
If you believe the Service was used against your site in a way that goes beyond this policy, write to [email protected] with the time and URL, and we will investigate.
7. What a verified full scan does
A full scan runs only after the person who requests it proves control of the site by placing a token on it (a file, a meta tag or a DNS record). We re-check that proof right before every scan. The scan then makes read-only requests: it lists which database tables, storage buckets and authentication settings the app's own public keys expose, reads at most a row or two as proof but discards the values, and looks for files such as .env and .git on the verified host.
It never writes, changes or deletes data, never guesses passwords and never uses a secret key it finds. It contacts only the verified host and the backend (such as Supabase) that the site's own code points to.
Before each full scan the requester must also confirm that the site and the backends it uses are theirs to test, and we limit how often any single backend project can be checked, whoever asks.