// Anthropic
anthropic-ai
// Overview
What it is
anthropic-ai is Anthropic's legacy training-data crawler token, separate from Claude's user-triggered fetcher. Aiola classifies it as training because it performs automated collection rather than a user-triggered fetch.
In Aiola's taxonomy, Training means scraped for model training or dataset development.
// Headers
User agent
Look for the match token inside the complete HTTP User-Agent header. Tokens can be spoofed, so use the network checks below for authentication.
anthropic-ai/1.0; +https://www.anthropic.comanthropic-ai// Network
Official IP ranges
Anthropic has not published official IP ranges that Aiola successfully fetched for anthropic-ai. Aiola does not list cloud-provider ranges or community guesses as if they authenticated this crawler.
// Verification
Verify authenticity
Start with the source IP recorded by your trusted edge or server, not an untrusted forwarded header.
host REQUEST_IP
# Confirm the hostname ends in an official suffix
host RETURNED_HOSTNAME
# The forward lookup must return REQUEST_IPFor production checks, test against every current prefix in the vendor feed. Re-fetch feeds regularly: a July 2026 snapshot is evidence of publication, not a permanent firewall list.
// Robots.txt
Control it with robots.txt
Anthropic documents crawler controls for this bot family. Robots.txt can direct cooperative crawling, but it cannot prove that a request is authentic.
User-agent: anthropic-ai
Allow: /User-agent: anthropic-ai
Disallow: /// Analytics
Track it
Aiola tracks this crawler on your site — crawls, pages, trends. See when anthropic-ai arrives, which URLs it requests, and how activity changes over time.
// FAQ
FAQ
What is anthropic-ai?
anthropic-ai is Anthropic's legacy training-data crawler token, separate from Claude's user-triggered fetcher. Aiola classifies it as training because it performs automated collection rather than a user-triggered fetch.
What user-agent token identifies anthropic-ai?
Match the case-insensitive token “anthropic-ai” in the User-Agent header. A matching header alone does not authenticate the sender.
Can I verify anthropic-ai by IP address?
No official anthropic-ai CIDR snapshot was successfully fetched for this verification date. Do not treat an arbitrary cloud IP as proof of identity.
Can robots.txt block anthropic-ai?
Anthropic documents crawler controls for this bot family. Robots.txt can direct cooperative crawling, but it cannot prove that a request is authentic. Use the specific “anthropic-ai” group when you want a rule for this identity without changing rules for every crawler.
// Related