What is Meta-ExternalAgent?

Meta · AI training · class ai

Crawls for Meta AI model training and product improvement.

Verdict: Charge it or block it. It does not pay yet.

On the mugshot wall: No published way to verify a request is really this crawler

Profile

Compiled by Crawler Atlas from the operator's published documentation. Not yet declared by the operator. JSON

Operator	Meta · developers.facebook.com
Purpose	AI training crawlers. Collect pages to train models. Bulk, periodic, and the ones most owners want to charge or block.
User-agent contains	`meta-externalagent · meta-externalfetcher · facebookbot`
robots.txt	`User-agent: Meta-ExternalAgent` `Disallow: /`
How to verify	No published verification method. Treat the name as a claim, not proof.
Should you block it?	Charge it or block it. It does not pay yet.
Does it pay?	No. It does not carry a wallet and cannot answer a 402. A site that charges simply serves it nothing.
Docs	https://developers.facebook.com/docs/sharing/webmasters/web-crawlers

Anonymous daily counts from sites running Pay-per-Crawl that opted in, plus the ForgeMesh fleet. Last 7 days. Small sample: 1 sites reporting.

No sightings reported yet. Meta-ExternalAgent appears here the day an opted-in site sees it.

See it in your own logs, verify it, and charge, allow or block it by name with the free Pay-per-Crawl plugin (WordPress, Python, JavaScript, Go, PHP).
Run this crawler? Publish a crawler-profile.json on your domain and this page shows your declared profile with a verified badge.
Machine-readable twin: /bots/meta-externalagent/profile.json (free in a browser). Machines: POST {"slug":"meta-externalagent"} to x402.forgemesh.io/bot-profiles, $0.005 per call over x402.

Crawler Atlas is maintained by ForgeMesh. Names and marks belong to their operators and are used to identify their crawlers.