Always-on monitoring of Amazon's three-bot fleet. Amazonbot trains AI models. Amzn-SearchBot powers Alexa and Rufus search eligibility - and per Amazon's docs, does NOT train. Amzn-User handles live consumer requests. Sentry catches the splits most publishers miss. Cortex handles the fix.
Continuous audits of Amazon's three-bot fleet against the 9 things that determine whether you're trained into Amazon AI, eligible for Alexa, surfaced by Rufus, and reached by live consumer queries. Amazon explicitly states Amzn-SearchBot does not train - so the bots are separately controllable. Sentry catches the gaps. Cortex fixes it.
User-agent: Amazonbot is not Disallowed. Per Amazon's docs: Amazonbot is 'used to improve our products and services' and 'may be used to train Amazon AI models.' Block here if you want to opt out of training.
User-agent: Amzn-SearchBot is not Disallowed. Per Amazon's verbatim statement: 'Amzn-SearchBot does not crawl content for generative AI model training.' Blocking it costs Alexa/Rufus search eligibility without preventing training.
User-agent: Amzn-User present in robots.txt. The third Amazon bot - live consumer-query crawler with its own IP range. Independent from both training (Amazonbot) and indexing (Amzn-SearchBot).
Recent Amazonbot visitor IPs match developer.amazon.com/amazonbot/ip-addresses. Filters spoofed UAs from non-Amazon ASNs.
Recent Amzn-SearchBot visitor IPs match developer.amazon.com/amazonbot/searchbot-ip-addresses. Separate range from the training crawler.
Per Amazon's docs, page-level `noarchive` meta tag is the documented opt-out for AI training specifically (separate from noindex). Sites wanting indexed-but-not-trained set noarchive without noindex.
Recent Amzn-User visitor IPs match developer.amazon.com/amazonbot/live-ip-addresses. Third Amazon IP list, distinct from training and search.
If the site sells products, schema.org/Product markup parses cleanly with name, image, offers, and price. Powers Rufus shopping summaries and ranked answer comparisons.
Per Amazon's docs: 'may take ~24 hours for our systems to reflect changes.' Re-tests of recently changed robots.txt rules are gated on that window to avoid false failures.
Paste your homepage URL. Sentry verifies robots.txt rules for all three Amazon bots, triple IP-range fidelity, noarchive posture, and Product schema validity for Rufus eligibility, then ships a per-rule report. No signup, instant results, always free.
Sentry fetches your site, runs every AMAZONBOT rule, and renders the full result page before your next sip of coffee.
Each failed rule ships with a prescription paragraph. Hand it to engineering and the gap is closed before lunch.
Add your site to the daily Sentry sweep with one click. New regressions get caught the next morning.
9 rules in the AMAZONBOT Sentry. Daily 3:30 AM ET sweep.
One brain. Thirty-six pairs of eyes. Sentries monitor every visibility signal that decides whether search engines, AI engines, and ad platforms show you. Cortex reads what they see, weighs it against a unified corpus of platform documentation, and acts. Every move follows a defined decision protocol: action stated, reason given, impact named.