Amzn-User
Amazon operates Amzn-User, a fetcher that supports user actions such as answering Alexa queries with live web information; Amazon says it does not crawl content for AI training. Its robots.txt token is Amzn-User, but because a user can initiate its requests it may not follow all directives. Verification uses Amazon's published IP address list only; Amazon documents no reverse-DNS suffix.
Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; Amzn-User/0.1) Chrome/W.X.Y.Z Safari/537.36What Amzn-User does
Log entries from this fetcher carry a browser-style user-agent string with the token Amzn-User, for example "Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; Amzn-User/0.1) Chrome/W.X.Y.Z Safari/537.36". Amazon describes the visits as fetches made on a user's behalf: when a customer asks Alexa a question that needs current information, Amzn-User may retrieve live pages from the web to answer it. Amazon states that Amzn-User does not crawl content for generative AI model training. Amazon does not document whether it executes JavaScript.
Because the requests are tied to individual user actions, Amazon says Amzn-User "may not follow all robots.txt directives." A Disallow rule for the Amzn-User token is therefore not guaranteed to stop every fetch. Blocking it means Alexa answers that would have drawn on your live pages may no longer do so, per Amazon's description of the purpose. Amazon's general robots.txt section says its listed user agents respect the Robots Exclusion Protocol, honor rel=nofollow and the noarchive, noindex and none meta directives, and do not support crawl-delay. Each user agent's settings are independent, and changes may take around 24 hours to apply.
To verify a visit, match the source address against the list at developer.amazon.com/amazonbot/live-ip-addresses/, an HTML page with an embedded JSON blob whose entries are single IPv4 addresses without CIDR suffixes. Amazon documents no reverse-DNS suffix for this fetcher, so a user-agent match alone is not proof of origin, and the IP list is the only published check.
Operator note. Amazon's documentation states: "Because actions taken by Amzn-User can be initiated by a user, it may not follow all robots.txt directives." The page's general 'Our Approach to Robots.txt' section says automated crawling from the listed user agents respects the Robots Exclusion Protocol, honors rel=nofollow and meta noarchive, noindex and none, and does not support crawl-delay. Each user agent setting is independent, and changes may take about 24 hours to take effect. JavaScript execution is not documented.
Controlling Amzn-User with robots.txt
Use the token Amzn-User in robots.txt. Amazon documents limits on how this bot applies robots.txt; see the operator note.
User-agent: Amzn-User
Disallow: /User-agent: Amzn-User
Allow: /Verifying a request is really Amzn-User
Anyone can put Amzn-User in a User-Agent header. Amazon publishes an IP address list for Amzn-User at https://developer.amazon.com/amazonbot/live-ip-addresses/, embedded as JSON within an HTML page. Entries are single IPv4 addresses without a CIDR suffix; no IPv6 entries are listed. No reverse DNS suffix is documented.
Amazon lists addresses on its documentation page rather than in a machine-readable file. Our probe read that page on 2026-09-06 and extracted the entries below.
- Prefixes
- 1024
- IPv4
- 1024
- IPv6
- 0
First entries at fetch time: 84.15.177.17, 100.24.86.21, 100.24.94.158
SHA-256 of the fetched body: 34eeac4eadf2604d1014dd38a0f26cc81834f2cac1fbc286087e5c481d8bd6ee
Common questions
Should I block Amzn-User?
Amazon says Amzn-User fetches live pages to answer Alexa queries and does not collect content for AI model training. Blocking the Amzn-User token in robots.txt stops those user-initiated fetches where the directive is followed, though Amazon notes it may not follow every directive. Whether that trade-off suits you depends on whether you want Alexa answers drawing on your pages.
Does Amzn-User obey robots.txt?
Partially. Amazon states that because Amzn-User's actions can be initiated by a user, it may not follow all robots.txt directives. Its general policy for listed user agents is to respect the Robots Exclusion Protocol, honor rel=nofollow and noindex-style meta tags, and not support crawl-delay. Settings per user agent are independent and can take about 24 hours to take effect.
How do I verify Amzn-User?
Check the source IP against Amazon's live IP address list at developer.amazon.com/amazonbot/live-ip-addresses/. The page embeds JSON with a creationTime and a prefixes array of single IPv4 addresses. Amazon documents no reverse-DNS suffix, so the IP list is the only published verification method.
Sources
Every fact on this page was checked against Amazon's own documentation, listed below, and re-checked by a second reviewer before publication. Reviewed September 2026.
Related crawlers
This registry documents how operators describe their own bots so site owners can identify and control them. It does not publish third-party IP lists or guess at undocumented behaviour. To see how your own site responds to automated visitors, the bot detection scanner reads a URL's live response and names the protection it finds.