Meta-ExternalAgent
In one line
Meta-ExternalAgent is the user agent Meta uses to crawl the web for its AI products and models — manageable separately via robots.txt.
Going deeper
Meta-ExternalAgent is Meta's crawler for training and running its own AI products — the Llama family and the Meta AI assistant. It is a separate identifier from the bots that crawl for Facebook and Instagram surfaces, and robots.txt rules can target it independently.
For marketers the unusual thing about Meta AI is the placement: it answers users directly inside WhatsApp, Instagram and Messenger. That makes it part of the answer pool for in-app questions, which is hard to ignore in global consumer categories.
On the policy side, the Meta bots were carved out and documented relatively recently. If your site's bot policy has not been touched in a while, it is worth a quick review to confirm Meta-ExternalAgent and other newer identifiers are actually represented.
Related terms
GPTBot
GPTBot is OpenAI's official web crawler used for ChatGPT training and search indexing — controllable via robots.txt.
GEO·AEOClaudeBot
ClaudeBot is Anthropic's web crawler used for training Claude and grounding its answers — manageable via robots.txt.
GEO·AEOGoogle-Extended
Google-Extended is the separate user agent Google uses for training Gemini and Vertex AI, letting site owners control AI training access independently from regular search indexing.
GEO·AEOBytespider
Bytespider is the web crawler operated by ByteDance, TikTok's parent — feeding its in-house AI models, search and recommendation systems.
GEO·AEOllms.txt
llms.txt is a proposed text file placed at the site root that tells large language models where the most important content lives — think 'sitemap, but written for LLMs'.
How does your brand show up in AI answers?
Villion measures how your brand appears across ChatGPT, Perplexity and AI Overviews, then automates the work that lifts citation rate and share of voice.
Get a free audit