Agent Ready Crawlers

The field guide to every AI crawler and agent at the door, and the vendors who decide who gets in.

Meta: 7 rows

In short

Meta has 7 rows in the Index: Meta-ExternalAgent, Meta-ExternalFetcher, Meta-WebIndexer, Meta-ExternalAds, FacebookExternalHit, FacebookBot and Muse. They cover 5 purposes: training, live answer fetch, search index, other, shopping agent. 6 of them have a robots.txt token and 7 publish nothing a site can verify. The most refused is Meta-ExternalAgent, disallowed at the root by 16.0 percent of top-10,000 sites.

Meta: 7 rows.

Operator Token Purpose Respects robots.txt Verification Block rate 2026-11-04 Profile
Meta meta-externalagent Training Yes None published 16.0% Profile
Meta meta-externalfetcher Live answer fetch May not apply None published 8.5% Profile
Meta meta-webindexer Search index Yes None published not measured Profile
Meta meta-externalads Other Yes None published not measured Profile
Meta facebookexternalhit Other May not apply None published not measured Profile
Meta FacebookBot Other Legacy None published 10.2% Profile
Meta Muse (personal agent) no token Shopping agent No docs None published no token Profile

Vendor profile: Meta, with its products, standards, our probes and what it does not do.

Verification Ledger entry

Summary
User agent strings only; the page recommends IP allowlisting but links no list. Publishes nothing for Muse.
IP lists
publishes nothing
Reverse DNS
publishes nothing
Key directory
publishes nothing
Cloudflare
not checked
Visa and Mastercard
Visa TAP: no public operator-level registration found (checked 2026-11-04). Mastercard Agent Pay: no public operator-level registration found (checked 2026-11-04).

Block rates for Meta tokens

TokenRefused, top 10,000NamedVia *All crawled
Meta-ExternalAgent16.0%12.3%3.7%12.0%
FacebookBot10.2%6.4%3.7%7.3%
Meta-ExternalFetcher8.5%4.7%3.8%6.0%

Based on 5,938 top-10,000 sites with a parsed robots.txt; crawl of 2026-11-04.

In the news

Sources

Last reviewed 2026-11-04