Anthropic Claude & ClaudeBot IP Address List

Live IP ranges for Anthropic's crawler infrastructure, covering; ClaudeBot (training), Claude-User (user-initiated fetches), and Claude-SearchBot (search indexing). This page fetches directly from Anthropic's official JSON feed hourly and updates automatically.

Anthropic Crawler IP Ranges

Anthropic operates three distinct bots that share the same published IP ranges. Use these to verify incoming requests are genuine Anthropic crawlers, or to configure firewall and CDN rules.

ClaudeBot

Training data collection

Block effect: Excludes your site from future Claude training datasets

Claude-User

User-initiated fetches

Block effect: Prevents Claude from retrieving your content when a user shares your URL

Claude-SearchBot

Search result indexing

Block effect: Reduces your visibility in Claude's search responses

IP Geolocation Finder

Enter any IPv4 or IPv6 address to find its geographic location and network details.

Common Misconceptions

✗There is no published list of every Claude server IP — only crawler and bot ranges are disclosed.

✗Claude API traffic does not originate from a fixed, publicly available IP list you should hardcode.

✓Anthropic operates three bots: ClaudeBot (training), Claude-User (user fetches), and Claude-SearchBot (search indexing).

✓Anthropic publishes a common set of IP ranges for its crawler infrastructure; the individual bots are distinguished by their user-agent strings.

✓Ranges are fetched live from claude.com/crawling/bots.json — this page refreshes automatically every hour.

✓You can block training collection while still allowing user-initiated fetches and search indexing independently.

How to Use These IP Ranges

Verifying genuine Anthropic crawler requests

Malicious scrapers may spoof Anthropic user-agent strings to bypass security rules. If a crawler has a source IP address on this list, it indicates that the crawler is coming from Anthropic.

Controlling access via robots.txt

Each bot has its own user-agent string. You can control them independently — for example, allowing user-initiated fetches while blocking training collection:

# Block training data collection only
User-agent: ClaudeBot
Disallow: /

# Allow user-initiated fetches (Claude answering questions about your site)
User-agent: Claude-User
Allow: /

# Allow search indexing
User-agent: Claude-SearchBot
Allow: /

Blocking all three is also valid — use Disallow: / under each agent.

Firewall and CDN rules

To block or allowlist Anthropic at the network level, import the IP ranges above into your CDN or WAF rules. Use the export button above to download as CSV, JSON, or TXT. Because Anthropic uses GCP and Azure infrastructure, IP-based blocking is designed as a network level verification / control mechanism, not as Anthropic's recommended method for opting out of crawling. Setting robots.txt file rules for crawler control is Anthropic's recommendation for maintaining reliable control over individual bot types. Relying solely on a blanket block of IPs for Anthropic bot blocking can make your robots.txt unreachable, so fine grain control is recommended here.

About the data

Anthropic publishes Anthropic crawler IP ranges as a machine readable JSON feed at claude.com/crawling/bots.json. This page fetches Anthropic's official JSON feed hourly, so the IP ranges display here are automatically refreshed from Anthropic's source. So the ranges shown are always current without you needing to check manually.

IP range and user-agent strings serve different purposes. The user-agent identifies which Anthropic crawler is making the request, while the published IP ranges are used to verify that the request is authentic and is made by Anthropic's infrastucture. The most robust identification methods will check both.