# ClaudeBot vs GPTBot

ClaudeBot and GPTBot compared on 13 documented fields, from 20 claims, last verified 2026-09-15.

| Field | ClaudeBot | GPTBot | Comparison |
| --- | --- | --- | --- |
| anti_circumvention_stance | documented | not documented | different |
| honours_crawl_delay | yes | not_documented | different |
| observed_fetches_json | yes | not documented | different |
| observed_fetches_markdown | yes | not documented | different |
| observed_first_request_path | /sitemap.xml | not documented | different |
| observed_format_pass_order | html_json_md | not documented | different |
| publishes_ip_list | yes | yes | same |
| purpose_documented | training | training | same |
| respects_robots_txt | yes | yes | same |
| robots_txt_fetch_marker | not documented | yes | different |
| separate_search_and_training_tokens | yes | yes | same |
| user_agent_string_full | not_documented | Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko); compatible; GPTBot/1.4; +https://openai.com/gptbot | different |
| verifiable_by_rdns | not_documented | not_documented | same |

## Differences

| Field | ClaudeBot statement | GPTBot statement |
| --- | --- | --- |
| anti_circumvention_stance | ClaudeBot respects anti-circumvention technologies and does not bypass CAPTCHAs on crawled sites. | not documented |
| honours_crawl_delay | Anthropic supports the non-standard robots.txt Crawl-delay extension for ClaudeBot. | OpenAI does not publish whether GPTBot honours the robots.txt Crawl-delay directive. |
| observed_fetches_json | ClaudeBot made 177 JSON requests to Rattlesnakes By Mail between 2026-09-13T20:00Z and 2026-09-15T09:00Z, covering the .json twin of 170 base paths. The .json URLs of Rattlesnakes By Mail appear in no entry of the 171 URL sitemap. | not documented |
| observed_fetches_markdown | ClaudeBot fetched the .md twin of 170 base paths on Rattlesnakes By Mail between 2026-09-13T20:00Z and 2026-09-15T09:00Z, 170 Markdown requests of 542 ClaudeBot requests in that window. The .md URLs of Rattlesnakes By Mail appear in no entry of the 171 URL sitemap. | not documented |
| observed_first_request_path | ClaudeBot's first request to Rattlesnakes By Mail between 2026-09-13T20:00Z and 2026-09-15T09:00Z was /sitemap.xml at 2026-09-14T19:53:45Z. ClaudeBot fetched /robots.txt on the second request to Rattlesnakes By Mail. | not documented |
| observed_format_pass_order | ClaudeBot fetched Rattlesnakes By Mail in three format passes between 2026-09-13T20:00Z and 2026-09-15T09:00Z, HTML from 2026-09-14T20:52Z to 2026-09-14T20:57Z, then JSON from 2026-09-14T21:37Z to 2026-09-14T21:56Z, then Markdown from 2026-09-14T21:37Z to 2026-09-14T21:57Z. The same order holds on 170 of 170 base paths that ClaudeBot fetched in all three formats on Rattlesnakes By Mail. | not documented |
| robots_txt_fetch_marker | not documented | OpenAI may add a robots.txt marker to the GPTBot user agent string when GPTBot fetches robots.txt files. |
| user_agent_string_full | Anthropic does not publish a full user agent string for ClaudeBot. | GPTBot sends the user agent string Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko); compatible; GPTBot/1.4; +https://openai.com/gptbot. |

## Sources

- https://claude.com/crawling/bots.json
- https://developers.openai.com/api/docs/bots
- https://openai.com/gptbot.json
- https://rattlesnakesbymail.com/observed/claudebot
- https://support.claude.com/en/articles/8896518-does-anthropic-crawl-content-from-the-web-and-how-can-site-owners-block-the-crawler