Privacy

Esta página solo está disponible en inglés. La versión en inglés es la que prevalece.

Last updated: 2026-07-27

Who runs this

AgentFit is operated by Gumeniuk Stanislav, an individual, in a personal capacity. Contact: [email protected].

What we collect

Minimal data. There is no registration, no user accounts, no login on the public surface. The admin panel at /admin uses strictly-necessary cookies for the operator's session — see /cookies for the full list.

When you submit a documentation URL for audit, we record:

Audit reports are public by default. Anyone can browse them at /browse or via a direct /r/{id} link. Submitting a URL is consent to that.

Not available in Russia

As described above, this service processes some personal data: it logs your IP address and request timestamps, and the URL you submit, and audit reports are public by default. The service is not made available in the Russian Federation.

By using the service you confirm that you are not located in the Russian Federation and are not accessing it from the Russian Federation.

If you cannot confirm this, do not use the service and leave the site now.

Note: this notice is a condition of use, not a technical control. Leaving the site does not erase data already processed when the page loaded — for example the IP address your browser sent in the request that delivered this page, which is logged like any other request (see "Logs" below). To have such data removed, email [email protected].

Cookies

See the dedicated cookie policy for the full per-category list, lifetimes, and how to change or withdraw your consent.

Logs

Server logs include IP addresses and request timestamps for operational purposes (debugging, abuse prevention, rate limiting). Logs are retained for up to 90 days, then deleted automatically.

Removing your audit

If a public audit references a URL you control and you want it removed from the public index:

  1. Email [email protected]
  2. Include the report link (e.g. https://agentfit.dev/r/<id>)
  3. Within 30 calendar days the report will be hidden from /browse and the direct link will return 404

Aggregate, anonymised statistics derived from the audit (e.g. score distributions across all sites) may be retained even after takedown — these never include the URL or any identifying detail.

The bulk open-data export at /dataset is not covered by that retention clause: it lists the entry URL, so it follows the takedown. It catches up within a further 24 hours, because it is regenerated daily from the same data /browse reads. Copies other people already downloaded cannot be recalled — see /dataset.

Crawling third-party sites

To produce an audit, the service issues HTTP requests against the submitted URL and its public sub-pages. The User-Agent strings, rate limits, and request budget are documented at /bot.

Future

A paid tier with private audits and user accounts is planned but not yet available. When it ships, this notice will be updated and the "last updated" date above will change.

Changes

Changes to this notice will be reflected in the "last updated" date. Continued use of the service after a change indicates acceptance.