What is Meta-ExternalAgent?

Meta's AI training crawler. It is separate from facebookexternalhit, which builds link previews. Intent: training. Downloads your content to train AI models.

What is the Meta-ExternalAgent user agent?

Requests from Meta-ExternalAgent carry "meta-externalagent" in the User-Agent header (matching is case-insensitive). The full string includes a version number that changes over time; Meta's documentation lists the current one. In robots.txt, the token is Meta-ExternalAgent.

How do I block Meta-ExternalAgent in robots.txt?

Add "User-agent: Meta-ExternalAgent" followed by "Disallow: /" to your robots.txt. robots.txt is a request, not a lock: a scraper using Meta-ExternalAgent's name ignores it, which is why verification matters.

Can I see when Meta-ExternalAgent visits my site?

Yes. HadoSEO logs every Meta-ExternalAgent request by page, and labels it declared, since Meta publishes nothing to verify it against. For a one-off look, the free log file analyzer reads a week of your server logs in the browser.

[ AI training crawler · Meta ]

Meta-ExternalAgent: Meta's AI training crawler.

Meta's AI training crawler. It is separate from facebookexternalhit, which builds link previews. Its intent is training: downloads your content to train AI models.

traininguser agent only

IP ranges last updated Oct 25, 2026

Check a Meta-ExternalAgent IP All AI crawlers

[ technical facts ]

operator: Meta
intent: training
user agent contains: meta-externalagent
robots.txt token: Meta-ExternalAgent
verification: user agent onlydeclared
official IP list: not published
operator docs: developers.facebook.com/docs/sharing/webmasters/web-crawlers/

[ verify ]

How to verify a real Meta-ExternalAgent request.

Meta publishes no IP list or reverse DNS pattern for Meta-ExternalAgent, so a request using its name can't be confirmed. HadoSEO labels it declared: recognized by user agent only.

01Check the User-Agent header contains "meta-externalagent" (case-insensitive). That is only a claim: anyone can send it.
02Meta publishes no IP list or reverse DNS pattern for Meta-ExternalAgent that HadoSEO can check, so there is no way to prove a request is real.
03Treat it as declared: it says it is Meta-ExternalAgent, and nothing confirms or contradicts that.

Skip the lookups: the free bot verifier checks one IP against the same registry, and the log file analyzer checks every Meta-ExternalAgent request in a week of access logs. Requests that fail are flagged as spoofed, not hidden.

[ robots.txt ]

Block or allow Meta-ExternalAgent in robots.txt.

Add one of these groups to the robots.txt at your domain root. The token matches case-insensitively.

block Meta-ExternalAgent

User-agent: Meta-ExternalAgent
Disallow: /

allow Meta-ExternalAgent

User-agent: Meta-ExternalAgent
Allow: /

robots.txt only binds crawlers that choose to follow it. A scraper sending Meta-ExternalAgent's user agent from its own servers won't, which is why HadoSEO checks the IP behind every request.

[ questions ]

Meta-ExternalAgent, answered.

[ free scan ]