Skip to content
  • Agents

    AgentsEvery AI agent, namedReplaysWhat each agent didIssuesWhat stops agents, ranked

    Results

    ImpactThe visitors AI sends youTestsAsk agents your buyers' questionsReportsEvery report, and a weekly one

    Beyond agents

    ReachWhere your content travelsCampaignsUTM links and real clicksPlatform overviewEverything in one place
  • Pricing
  • Security
  • Learn

    AI agent directoryEvery agent, and how to spot itInstall guidesWebflow, WordPress, Next.js and moreChangelogWhat's newFAQCommon questions

    Company

    Privacy policyHow we handle dataTerms of serviceThe rules for using AirglowStatusIs everything working?Supportsupport@airglow.so
Sign in
Start free

Get started

Start freeOne website, no cardStart with ProUp to 6 websites

Account

Sign inEmail link, no password

Platform

Agents and replaysImpactTestsIssuesReach and campaigns

Resources

AI agent directoryInstall guidesChangelogFAQ

Company

PricingSecurityPrivacyTerms
Start freeSign in

AI agent directory / Amzn-User

Amzn-User

AI fetcher from Amazon

Fetches a page when an Alexa customer's question needs up-to-date information from the web.

Amzn-User is the user agent Amazon uses when a customer's request needs fresh information from the web. Amazon's example is an Alexa question that requires up-to-date facts: Amzn-User fetches the page and the answer is built from it on the user's behalf.

It is one of three Amazon web agents, next to Amazonbot and Amzn-SearchBot. Amazon says each one is controlled separately, and that Amzn-User does not crawl content for generative AI model training.

Run byAmazon
TypeAI fetcher
PurposeActs for a person
Acts forA person, in the moment
robots.txt tokenAmzn-User
IP rangesPublished by Amazon
DocumentationAmazon's documentation

Amzn-User user-agent string

Amzn-User (Amazon's example; W.X.Y.Z is the Chrome version) (source):

Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; Amzn-User/0.1) Chrome/W.X.Y.Z Safari/537.36

Match on Amzn-User; the rest of the string changes between versions.

Verifying it's really Amzn-User

Amazon publishes the IP addresses Amzn-User uses at developer.amazon.com/amazonbot/live-ip-addresses/. The page shows a JSON document with a prefixes list of single ipv4Prefix addresses. Treat a request that names Amzn-User but comes from an address outside that list as unverified.

How Airglow identifies it

  • DeclaredIt names itself in the User-Agent: Amzn-User.

Every visit in Airglow says which of these applied, so you can decide how much to trust it.

Seeing it on your site

It reads pages without running JavaScript, so Airglow counts it through a server connection (Cloudflare). It's labeled as context, not replayed.

What the script and a server connection each see

Allowing or blocking Amzn-User

Amzn-User reads the Amzn-User group in your robots.txt. To keep it out of the whole site:

User-agent: Amzn-User
Disallow: /

Airglow shows, for every agent that visits, whether your robots.txt lets it in, and helps you write rules by purpose (training, search, acting for a person).

Does Amzn-User follow robots.txt?

Not always. Amazon's documentation says that because Amzn-User's actions can be started by a user, it "may not follow all robots.txt directives." A Disallow rule for Amzn-User states your preference, but Amazon does not promise it will block every fetch.

For its automated crawling, Amazon says its agents read robots.txt per host, may use a cached copy up to 30 days old, honor noindex, noarchive, none and rel=nofollow, and ignore crawl-delay. Changes to any of the three Amazon user agents can take about 24 hours to take effect.

Amzn-User, Amazonbot and Amzn-SearchBot

  • Amazonbot crawls to improve Amazon products and services, and its content may be used to train Amazon AI models.
  • Amzn-SearchBot crawls for search experiences such as Alexa and does not collect content for model training. If robots.txt does not name it but allows other search bots, it follows the rules given to those bots.
  • Amzn-User fetches pages in response to a customer request and does not collect content for model training.

Blocking Amazonbot does not block Amzn-User. Each token needs its own robots.txt group.

Questions

What is Amzn-User?

It is Amazon's user agent for fetching web pages in response to a customer action, such as an Alexa question that needs current information.

Is Amzn-User used to train AI models?

Amazon says Amzn-User does not crawl content for generative AI model training. Amazonbot is the Amazon agent whose content may be used for training.

Can I block Amzn-User with robots.txt?

You can add a group for the Amzn-User token, but Amazon says Amzn-User may not follow all robots.txt directives because its fetches can be started by a user.

Does blocking Amazonbot also block Amzn-User?

No. Amazon says each of its user agent settings is independent, so Amazonbot, Amzn-SearchBot and Amzn-User each need their own rules.

Who do I contact about Amzn-User?

Amazon asks content owners to email amazonbot@amazon.com and to include the relevant domain names in the message.

Other agents from Amazon

  • AmazonbotCrawler, training
  • Amzn-SearchBotCrawler, search

Sources

  • Amazon Developer: Amazon Web Search crawlers
  • Amazon Developer: Amzn-User IP addresses

Last reviewed November 10, 2026.

See when Amzn-User visits your site

Airglow names every AI agent on your website, replays what it did and links the visitors it sends you. Free for one website.

Start freeAll agents

Analytics for the traffic you can't see. See what AI agents do on your site.

Product

  • Platform
  • Pricing
  • Security
  • Start free

Company

  • Privacy
  • Terms
  • Status
  • Support

Resources

  • AI agent directory
  • Install guides
  • Changelog
  • FAQ

© 2026 Miscreants