# Google-Agent

> Google · inference

_The AI Crawler Registry · /crawlers/google-agent · [JSON](/api/crawlers/google-agent) · [all The AI Crawler Registry](/crawlers)_

- **name:** Google-Agent
- **operator:** Google
- **purpose:** inference
- **ua_substring:** Google-Agent
- **robots_token:** Google-Agent
- **respects_robots:** no
- **verify:** Google IP ranges (user-triggered-agents.json) + reverse DNS to google.com / googleusercontent.com
- **notes:** User-triggered fetcher used by agents hosted on Google infrastructure to navigate the web and perform actions on a user's request (for example, Project Mariner / Gemini Agent). As a user-triggered fetcher, Google documents that it generally ignores robots.txt rules.
- **canonical_name:** Google-Agent
- **user_agent_token:** Google-Agent
- **ua_full:** — (verify-against-primary-at-build)
- **bot_type:** user-action-fetcher
- **bot_type_extension:** —
- **opt_out_mechanism:** limited — Google documents this user-triggered fetcher as generally ignoring robots.txt; rate-limit / verify at the edge instead
- **published_ip_range_url:** https://developers.google.com/static/crawling/ipranges/user-triggered-agents.json
- **asn:** — (verify-against-primary-at-build)
- **reverse_dns_suffix:** .google.com (source: https://developers.google.com/search/docs/crawling-indexing/verifying-googlebot)
- **supports_web_bot_auth:** — (verify-against-primary-at-build)
- **signature_agent_domain:** — (verify-against-primary-at-build)
- **jwks_url:** — (verify-against-primary-at-build)
- **verification_methods:** published-IP-range, reverse-DNS
- **crawl_traffic_share:** — (verify-against-primary-at-build)
- **targeted_content_type:** HTML, text
- **documentation_url:** https://developers.google.com/search/docs/crawling-indexing/google-user-triggered-fetchers
- **first_seen_date:** — (verify-against-primary-at-build)
- **last_verified_date:** 2026-06-15
- **block_vs_allow_recommendation:** conditional — agent fetcher acting on a user's request (Gemini Agent / Project Mariner lineage); allow for legitimate user-driven actions, but it generally ignores robots.txt, so rate-limit / verify at the edge if load or abuse is a concern.
- **citation_referral_value:** medium (acts on a user's behalf; can surface your page to that user)
- **cloudflare_verified_category:** — (verify-against-primary-at-build)
- **status:** active
- **triples:** ["Google-Agent","operated_by","Google"], ["Google-Agent","has_bot_type","user-action-fetcher"], ["Google-Agent","verified_via","published-IP-range"], ["Google-Agent","verified_via","reverse-DNS"], ["Google-Agent","succeeds","Project Mariner"]
- **attribute_sources:** {"claims":["user_agent_token","robots_token","published_ip_range_url","documentation_url","respects_robots"],"source":"https://developers.google.com/search/docs/crawling-indexing/google-user-triggered-fetchers","last_verified":"2026-06-15"}, {"claims":["reverse_dns_suffix"],"source":"https://developers.google.com/search/docs/crawling-indexing/verifying-googlebot","last_verified":"2026-06-15"}