Feedfetcher vs APIs-Google
Feedfetcher-Google and APIs-Google are both special-case Google fetchers that are easy to confuse with each other and with Googlebot. Feedfetcher pulls RSS/Atom feeds for Google products that subscribe to them; APIs-Google delivers API-triggered messages such as push notifications. Neither is the search crawler. Google documents both in its crawlers and fetchers list.
What this means
Google operates several special-case fetchers beyond Googlebot. Feedfetcher-Google retrieves RSS and Atom feeds on behalf of Google products that subscribe to them — it is a pull initiated by a feed subscription. APIs-Google handles API-triggered requests such as push notification deliveries — it is typically a push triggered by an API integration.
Both are documented in Google's crawlers and fetchers list, and neither builds the search index. Confusing them leads to wrong conclusions about what Google is doing on your site.
How to tell them apart
Match on the token: Feedfetcher-Google for feed pulls, APIs-Google for API-triggered deliveries. Look at what is being requested — a feed URL points to Feedfetcher; an API callback endpoint points to APIs-Google.
Blocking either in robots.txt can break the function you subscribed to (feed inclusion or API delivery), so change policy deliberately. For authenticity, verify the source IP against Google's published fetcher ranges, since user agents can be copied.
- Feedfetcher-Google: pulls RSS/Atom feeds (subscription-driven)
- APIs-Google: delivers API-triggered messages (push)
- Neither is Googlebot; both are special-case fetchers
How it appears in analytics and logs
A Feedfetcher token means a Google product is pulling your RSS/Atom feed; an APIs-Google token means a Google API is pushing a message to your endpoint. Both are functional Google traffic, distinct from Googlebot indexing.
Diagnostic use case
Correctly attribute Google feed pulls versus API push deliveries in your logs, so neither is mistaken for search crawling or for the other.
What WebmasterID can help detect
WebmasterID distinguishes Feedfetcher and APIs-Google as separate Google fetchers, so feed pulls and API deliveries are categorised correctly and never folded into search-crawl coverage.
Common mistakes
- Confusing Feedfetcher feed pulls with APIs-Google push deliveries.
- Treating either as Googlebot search-crawl activity.
- Blocking one in robots.txt and breaking a Google feature you rely on.
Privacy and accuracy notes
Both fetchers are identified by user-agent token and verifiable against Google's ranges. They are automation, not visitors; WebmasterID records them as bot events with no human profile.
Related pages
- Feedfetcher-Google — feed fetcher
Feedfetcher-Google is the user agent Google uses to fetch RSS and Atom feeds for Google products. Google documents that Feedfetcher is not used for indexing, and that because feed fetches are user-requested subscriptions, it is handled differently from indexing crawlers.
- APIs-Google fetcher
APIs-Google is a Google fetcher user agent used when Google products send push notifications or other API-driven requests to a developer's server — for example a PubSubHubbub (WebSub) delivery. It is not the search crawler and is not used to build the search index. Google documents it in the list of Google crawlers and fetchers, and it is verifiable against Google's published crawler IP ranges.
- Mediapartners-Google (AdSense crawler)
Mediapartners-Google is the crawler Google AdSense uses to fetch and analyse the content of pages that show AdSense ads, so it can choose relevant ads. It is documented in Google's crawler list as a special-case crawler tied to the AdSense product, distinct from the Googlebot search crawler, and it has its own robots.txt token.
- Bot intelligence
Separates Google's special-case fetchers from Googlebot.
Sources and verification notes
- Google Search Central — Google crawlers and fetchers overviewBoth Feedfetcher-Google and APIs-Google documented as special-case fetchers.
Last reviewed 2026-06-24. Facts are checked against primary/official sources where available; uncertain specifics are marked “Data not yet verified” rather than guessed.