Sona

GoogleAgent-URLContext

Fetches pages that Gemini API developers pass as URL context. A user-triggered fetcher, so robots.txt rules generally aren't honored.

Reference reviewed by Sona

OperatorGoogle
PowersURL-context grounding for Gemini API apps
PurposeUser-triggered fetch
User-agent tokenGoogleAgent-URLContext
Respects robots.txtNo (user-triggered)

GoogleAgent-URLContext retrieves web content when a Gemini API application uses the URL-context tool - developers supply URLs and the fetcher pulls a live copy so the model can ground its answer in the page and cite it.

Like Google's other user-triggered fetchers, Google documents that it generally ignores robots.txt because each fetch is requested by a person or their application. Server- or firewall-level user-agent filtering is the only way to enforce a block.

Allow GoogleAgent-URLContext

Apps built on the Gemini API can ground answers in your real, current pages - your content reaches AI products beyond Google's own.

User-agent: GoogleAgent-URLContext
Allow: /

Block GoogleAgent-URLContext

You don't want third-party Gemini-powered apps ingesting your pages - enforce it in your WAF, since robots.txt won't be honored.

User-agent: GoogleAgent-URLContext
Disallow: /

Heads up: GoogleAgent-URLContext does not reliably honor robots.txt. To enforce a block, combine the rule above with server- or firewall-level filtering of the user-agent.

Can GoogleAgent-URLContext read your page right now?

Test any URL and see exactly what AI crawlers receive.

Check my site