Gemini: Google's Retired Crawler
Gemini is no longer in use. A robots.txt rule naming Gemini has no effect today. It was replaced by Google-Extended, which is the token to address instead.
What is Gemini?
There is no Gemini crawler. Google documents no standalone Gemini user agent or robots.txt token: Gemini is served by existing Google crawlers, and access is controlled with Google-Extended. Gemini is a user-agent fragment on some Google fetches rather than a token you can address, so Google-Extended is what actually controls it.
Key facts
- Gemini is operated by Google, and its purpose is live fetch: fetches a page in real time because a person asked.
- Google publishes no full user-agent string for Gemini, so Kymo matches the token Gemini observed in real logs.
- Gemini cannot be addressed in robots.txt, so a rule naming it has no effect.
- Gemini can send a visitor to your site, and Gemini attaches a linked citation when it uses your content.
Specification
| Operator | |
|---|---|
| Purpose | Live fetchFetches a page in real time because a person asked. Traffic here tracks human interest, not a crawl schedule. |
| Agent kind | FetcherFetches on a person's request. Appears in your logs. Request volume follows human activity. |
| User agent | Not documented by Google. Kymo matches this agent on the token Gemini, observed in real logs rather than published in operator documentation. |
| robots.txt token | None. This agent cannot be addressed directly in robots.txt. |
| robots.txt compliance | No operator statement located |
| Verification | No published method, user agent is the only signal |
| Sends referral traffic | Yes |
| Attaches a citation | Yes |
| Status | Deprecated |
| Legacy Kymo category | ai_answer |
| Source | Observed in real logs, no operator documentation locatedGoogle publishes no crawler documentation that Kymo could locate. Everything on this page comes from observed traffic, not from a vendor page. |
| Verified on |
What this looks like in your logs
grep -i 'Gemini' access.log
Google publishes no full user-agent string for Gemini, so the rest of the line varies and only the token Gemini is dependable. Kymo matches on that token, case-insensitively, for the same reason.
What Gemini does not do
Gemini does not guarantee a visit, because a read only becomes a visit if a person clicks the citation that came out of it.
Gemini does not crawl on a schedule, because it only requests a page when a person has already asked something that needs it.
Google publishes no documentation for Gemini at all, so no crawl frequency and no compliance statement exists to quote.
How Kymo classifies it
Kymo files Gemini under Google as ai_answer, classified from the HTTP request on the server rather than from anything running in a browser.
A hit from Gemini counts on the AI Visibility page, which is the only page that counts crawls at all. An answer crawl is the only kind a person can click through from, so it is the only kind that belongs beside a visitor number.
Allow it or block it
A robots.txt rule naming Gemini has no effect, because no crawler answers to that token. Use Google-Extended to opt out of Gemini training and grounding. Kymo previously matched the bare string Gemini in user agents; that token was removed after it matched zero stored requests and because it is short enough to collide with unrelated strings.
| Allow it if | Block it if |
|---|---|
| Nothing. Gemini is retired and makes no requests. | Nothing. A robots.txt rule naming Gemini has no effect. Address the token that replaced it instead. |
Questions
How do I know a request claiming to be Gemini is genuine?
Google publishes no verification method for Gemini, so the user-agent string is the only signal available, and anyone can send one. Treat a claimed Gemini request as unproven.
Does blocking Gemini also block Google-Extended?
No, and blocking Gemini does nothing at all. Gemini is retired and makes no requests, so a robots.txt rule naming it has no effect on anything. Google-Extended is the active token, and it has to be named in a rule of its own.
Will I see Gemini in Google Analytics?
No. Google Analytics runs a JavaScript tag in a visitor's browser, and Gemini reads your HTML and leaves without running any script. Server-side logging is the only way to record it.
Does Gemini send traffic back to my site?
No. Gemini is retired and makes no requests.
Related crawlers
All AI crawlers and control tokens
See this bot in your own logs
Gemini arrives because a person just asked Google something, and a browser tag never runs in time to record the moment.
Kymo reads the HTTP request on your server, so a crawler that never runs JavaScript is still recorded. You see which bots reached the site, which pages they took and how often, next to your human traffic.
Start free → 14-day free trial. No card required.
No account yet? Run a free AI visibility check on your own site.