Terms

What the audit does

A single request is made to the URL you submit, plus requests for that origin's robots.txt and llms.txt. If you enable crawler probing, one additional request is made per crawler user-agent. Nothing is submitted, no forms are filled, and no authenticated content is accessed — the audit sees only what a public crawler would see.

What the score means, and does not

• It measures whether AI engines can reach, read and extract an answer from the page — a technical property of the HTML and HTTP response.
• It does not predict whether any engine will cite the page. Citation depends on content quality, authority and competition, none of which are measured.
• Crawler probe results are labelled with a confidence level. Probes are sent from this server, not from a vendor's verified IP range, so a bot-verification challenge is expected and is reported separately from a genuine block. Treat probe rows as signal, not fact.
• The scoring methodology is versioned. A report records the version that produced it, so past reports stay comparable when the rules change.

Acceptable use

• Audits are rate limited per IP address. Do not attempt to circumvent the limit.
• Do not use the service to scan hosts you have no relationship with at volume, to enumerate infrastructure, or as a proxy for reaching networks you cannot reach directly. Requests to private, loopback, link-local and reserved addresses are refused.
• Reports are public to anyone with the link. Do not audit URLs whose existence is confidential.

No warranty

The service is provided without warranty of any kind. Findings describe the page as fetched at that moment; a page can change, and an engine's behaviour can change independently of anything measured here. Decisions taken on the basis of a report are your own. The service may be rate limited, changed or withdrawn at any time.

Contact

Questions, corrections and removal requests: [email protected]. The full check registry and weights: geotool.site/methodology.