Amzn-SearchBot
Amazon's search crawler, documented as explicitly not crawling for generative AI training.
Reference reviewed by Sona
| Operator | Amazon |
|---|---|
| Powers | Search experiences in Amazon products such as Alexa |
| Purpose | Search & answers |
| User-agent token | Amzn-SearchBot |
| Respects robots.txt | Yes |
Amzn-SearchBot crawls to improve search experiences across Amazon products and services, making your content eligible to appear in surfaces such as Alexa answers. Amazon documents it separately from Amazonbot, with its own token and its own published IP list.
The explicit exclusion is what makes this token useful. Amazon states that Amzn-SearchBot "does not crawl content for generative AI model training" - a written commitment rather than an inference. Amazonbot, by contrast, is documented as potentially training AI models. So the two tokens give you a clean, vendor-supported split: allow Amzn-SearchBot for Alexa visibility, disallow Amazonbot to stay out of training.
That is a better-specified separation than most operators offer. Where Google asks you to trust that Google-Extended governs training use of a Googlebot crawl, Amazon simply runs a different crawler for the non-training purpose, so the boundary is architectural rather than policy-based.
It respects robots.txt and honors user-agent directives. Note the deliberately similar naming across Amazon's three tokens - Amazonbot, Amzn-SearchBot, Amzn-User - which makes copy-paste errors easy; each needs its own group.
How Amzn-SearchBot behaves
- Documented as explicitly not crawling content for generative AI model training.
- Has its own IP address list, separate from the Amazonbot and Amzn-User lists.
- Respects robots.txt and honors user-agent directives.
- Names are deliberately similar across Amazon's three tokens, making rule mix-ups easy.
Full user-agent string
Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; Amzn-SearchBot/0.1)
Allow Amzn-SearchBot
Alexa and Amazon search visibility with a documented guarantee that the content is not used for generative AI training.
User-agent: Amzn-SearchBot Allow: /
Block Amzn-SearchBot
You don't want your pages surfaced in Amazon's AI search experiences, even though no training use is involved.
User-agent: Amzn-SearchBot Disallow: /
How to verify Amzn-SearchBot
Published IP range feed
Amazon publishes a dedicated IP address list for Amzn-SearchBot, separate from both the Amazonbot and Amzn-User lists - three crawlers, three files. Point your verification at the searchbot list specifically, because a check written against the general Amazonbot list will fail perfectly legitimate traffic. Amazon's crawl.amazonbot.amazon reverse-DNS convention is documented for Amazonbot rather than for this token.
Check an IP against this botCommonly confused with Amzn-SearchBot
The key distinction: Amazonbot may train AI models, this token explicitly does not. Allowing this while blocking Amazonbot is the vendor-supported way to keep Alexa visibility without training exposure.
Amzn-User handles live user-initiated fetches and may not follow robots.txt; this token performs scheduled index crawling and does.
Amzn-SearchBot FAQs
Does Amzn-SearchBot use my content for AI training?
No. Amazon documents that it does not crawl content for generative AI model training. That role belongs to Amazonbot, which Amazon says may be used to train AI models.
How do I keep Alexa visibility but avoid Amazon's AI training?
Allow Amzn-SearchBot and disallow Amazonbot. Amazon runs them as separate crawlers with separate documented purposes, so the split is architectural rather than a policy promise layered over one crawl.
Which IP list should I verify against?
The Amzn-SearchBot list specifically. Amazon publishes three separate files for its three crawlers, so verifying this traffic against the general Amazonbot list will reject legitimate requests.
Why do Amazon's crawler names look so similar?
Amazonbot, Amzn-SearchBot, and Amzn-User are deliberately related, which makes copy-paste mistakes in robots.txt easy. Each needs its own group with its own exact token.
Can Amzn-SearchBot read your page right now?
Test any URL and see exactly what AI crawlers receive.