Does Anthropic publish crawler IP ranges?

Definition: Yes. As of 17 October 2026, Anthropic's crawler privacy documentation links to a public IP range list for identifying requests from Anthropic. It treats a matching source IP address as an indication of Anthropic origin.

Key facts

• Anthropic's crawler privacy documentation provides the operator's source-address verification guidance.
• The linked crawler feed contains a creationTime field and a prefixes array of ipv4Prefix entries.
• Anthropic's IP addresses documentation separately describes network addresses for access to the Claude API and Console.

How can a website check a crawler's source address?

The published crawler feed is available at this address:

https://claude.com/crawling/bots.json

Its prefixes entries describe address ranges. For example, the downloaded list includes this entry:

{ "ipv4Prefix": "216.73.216.0/22" }

A website operator compares the incoming request's source IP address with the ranges in the feed. Range membership is the relevant comparison: the address must fall within a listed prefix, rather than merely share some characters with its text representation.

The downloaded file has creationTime set to 2026-10-07T00:00:00Z. Keeping that field alongside a saved copy makes the version used for a check explicit. The file supplies a shared address list without labels assigning individual prefixes to ClaudeBot, Claude-User or Claude-SearchBot, so a match alone identifies no particular bot role.

Are the API address ranges interchangeable with the crawler list?

Anthropic's IP addresses documentation describes fixed inbound and outbound addresses for firewall configuration around the Claude API and Console. Inbound addresses receive connections to Anthropic services; outbound addresses cover requests those services send.

For a crawler-origin check, the appropriate starting point is the feed linked from the crawler privacy article. Keeping the source documents attached to an address rule preserves its intended scope: API connectivity and recognition of incoming website crawlers are separate configuration tasks.

How does verification relate to blocking?

A robots.txt file contains rules that crawlers are requested to honor. RFC 9309, the Robots Exclusion Protocol specification, defines this mechanism and explains that its rules provide no access authorization. Its security section calls for content security measures when access requires protection.

Anthropic's crawler privacy article directs website owners to change robots.txt for crawl opt-outs. Its ClaudeBot example is:

User-agent: ClaudeBot
Disallow: /

The article places this file in the website's top-level directory and requires the configuration on each subdomain being excluded. It warns that IP blocking can obstruct the bots' ability to read robots.txt. Source-address verification therefore serves a different operational purpose from communicating crawl preferences.

Related terms

• ClaudeBot
• Claude-SearchBot
• Claude-User
• robots.txt

FAQ

Where should I obtain Anthropic's crawler ranges?

Use the crawler JSON feed linked by Anthropic's privacy documentation. Preserve its creationTime when recording which list you used.

Can robots.txt protect private content?

RFC 9309 distinguishes crawler rules from access authorization. Use an appropriate security measure to protect content that requires restricted access.

Sources

1. Anthropic, "Does Anthropic crawl data from the web, and how can site owners block the crawler?", 7 April 2026
2. Anthropic, crawler IP range JSON feed, creationTime 7 October 2026
3. Anthropic, "IP addresses", accessed 17 October 2026
4. IETF, "RFC 9309: Robots Exclusion Protocol", September 2022

---

Check how a page reads to answer engines. Open the GEO score analyzer