Sites that block Amzn-User
208 of the 29,991 domains we have measured, 0.7%, disallow amzn-user at the site root in robots.txt. Amzn-User fetches a page live when a user asks the assistant for it.
Every domain here has removed itself from Amazon's answers. Blocking a retrieval agent withholds nothing from model training, because training uses a different token.
208 of 29,991 measured domains (0.7%) disallow amzn-user at the site root in robots.txt. Every domain named on this page was fetched directly and its robots.txt parsed against RFC 9309; none of it is inferred from a third-party index.
Measured 2026-11-21 by direct request to every domain in the corpus. Method: how this is measured. Reuse under CC BY 4.0 with attribution.
Cite as: Crawl Census, "Sites that block Amzn-User", measured 2026-11-21. https://crawlcensus.com/blocked/amzn-user
The list 208 domains
| Rank | Domain | TLD | AI access score |
|---|---|---|---|
| 49,126 | scientific.net | .net | 52 |
| 49,271 | eltern.de | .de | 72 |
| 49,501 | sj-r.com | .com | 64 |
| 49,574 | montgomeryadvertiser.com | .com | 62 |
| 50,278 | burlingtonfreepress.com | .com | 62 |
| 50,312 | mujerhoy.com | .com | 80 |
| 50,933 | nwaonline.com | .com | 73 |
| 51,018 | everydaycalculation.com | .com | 52 |
Rank is the domain's position in the popularity list used to seed the corpus; a dash means it was scanned on request rather than seeded. Score is the site's overall AI access score, which blocking depresses but does not by itself determine.
Check a specific domain
See also the Amzn-User reference, the full census, and the JSON endpoint.