Grok visibility:
what can actually be verified

Every other major assistant documents its crawlers

OpenAI documents four agents and says what each is for. Google documents its crawler tokens and states explicitly what Google-Extended does and does not affect. Perplexity publishes a bots guide covering both its indexing and user-initiated agents. Against that, the absence of any equivalent from xAI is not a small gap — it means a site owner has no vendor-confirmed way to express a preference about Grok's access, and no authoritative list to check their server logs against.

That has a direct consequence for anyone selling you Grok optimisation: the mechanism they are describing cannot be verified against anything the vendor has published. It may be an accurate description of observed behaviour. It is not documentation, and the difference matters when you are being asked to change your infrastructure on the strength of it.

Observation, and the conditions underneath

Two things remain solid even without vendor documentation. The first is observation: asking Grok the questions your customers ask and recording what comes back — whether your business is named, and how it is described when it is. That is a record of a moment rather than a position, and OG01 reports it as such, because responses vary between runs and presenting them as a stable ranking would be its own kind of invention.

The second is the state of your own domain. Whether your pages can be fetched and parsed, whether your organisation resolves to one clear entity, whether your claims are corroborated somewhere that is not you, and whether stale or contradictory information about you is in circulation. Those conditions are measurable regardless of which assistant is reading, they are the ones you control, and they are why OG01's measurement is deliberately platform-independent.

The source position, recorded

There is no vendor documentation to cite. That absence was checked and recorded rather than papered over, and it is held in the same source file as the other platforms so it can be re-reviewed rather than forgotten. Reviewed 2026-09-08.

• Checked 8 Sep 2026: xAI publishes no crawler or webmaster documentation — no user-agent list, no IP ranges, no stated robots.txt policy.
• Third-party lists circulate several candidate user-agent names, none confirmed by a vendor page. They are therefore NOT stated as fact on the page.
• The page must say the documentation does not exist rather than repeating an unattributable claim, and must confine itself to what OG01 can observe: what Grok returns when prompted.
• If xAI publishes documentation later, add it here and revise the page rather than leaving the absence recorded as permanent.

Grok questions