What is FacebookBot?
FacebookBot is a Meta crawler that collects public web content. Meta may use that material to improve language models and other AI products, which is why the bot is categorized for both AI crawling and AI training.
A request means FacebookBot reached a public URL. It does not prove that Meta retained the page, added it to a training dataset, or will reproduce its contents in an AI response. The practical choice for a publisher is whether Meta may fetch the page for this stated development purpose.
FacebookBot is not facebookexternalhit. The latter fetches titles, descriptions, and images when people share links on Facebook, Instagram, or Messenger. Blocking FacebookBot therefore does not express a preference for the link-preview crawler, which needs its own robots.txt group.
It is also separate from the crawlers in Meta's current web crawler documentation. Meta-WebIndexer supports Meta AI search, while Meta-ExternalAgent is described as crawling for model training and product improvement. FacebookBot itself has no documented search-index or citation role, so allowing it should not be presented as a direct way to appear in Meta AI answers.
The stable token in this bot is FacebookBot. No verified IP ranges or Web Bot Auth key directory are attached to the entry, and IP verification is false. A matching user-agent header is useful for log analysis and policy, but another client can copy it.
this bot says FacebookBot respects robots.txt. Sites can allow public material, exclude selected sections, or opt out of the crawl entirely. Policies for Meta's other bots should be written separately because each crawler has its own purpose and token.
