AI crawler policy › domain suffix

AI crawler blocking on .chat sites

Latest sealed date 2026-08-07.

In the latest sealed reading for .chat on 2026-08-07, 22 tracked crawler tokens were named by at least one observed file: Amazonbot: 10 named, 3 blocked site-wide; ClaudeBot: 10 named, 3 blocked site-wide; CCBot: 5 named, 4 blocked site-wide; Applebot-Extended: 4 named, 3 blocked site-wide; Bytespider: 4 named, 3 blocked site-wide; Google-Extended: 4 named, 3 blocked site-wide; GPTBot: 4 named, 3 blocked site-wide; meta-externalagent: 4 named, 3 blocked site-wide; anthropic-ai: 1 named, 0 blocked site-wide; ChatGPT-User: 1 named, 0 blocked site-wide; Claude-SearchBot: 1 named, 0 blocked site-wide; Claude-User: 1 named, 0 blocked site-wide; Claude-Web: 1 named, 0 blocked site-wide; cohere-ai: 1 named, 0 blocked site-wide; cohere-training-data-crawler: 1 named, 0 blocked site-wide; DuckAssistBot: 1 named, 0 blocked site-wide; FacebookBot: 1 named, 0 blocked site-wide; meta-externalfetcher: 1 named, 0 blocked site-wide; MistralAI-User: 1 named, 0 blocked site-wide; OAI-SearchBot: 1 named, 0 blocked site-wide; Perplexity-User: 1 named, 0 blocked site-wide; PerplexityBot: 1 named, 0 blocked site-wide. Each entry keeps the named count before the fully-blocked count. A tracked token absent from the list had zero observed mentions; the counts describe file declarations, not crawler behavior.

Every tracked crawler on .chat

Blocks outright — the file has a group for that exact product token containing Disallow: / with no overriding allow. Partly restricts — a group exists with at least one non-empty Disallow, but not the whole site. Names & allows — a group exists with no effective Disallow. Everything else is not mentioned: the wildcard * group governs it, and wildcard effects are counted separately rather than folded in, because a site that blocks everything has not made a decision about any AI crawler in particular. Matching is exact and case-insensitive per RFC 9309 — never substring.

14 newest sealed readings for .chat

For each date, the table keeps the panel count beside the parseable robots.txt denominator, the sites naming or blocking a tracked AI crawler, and wildcard blocks. Unreachable or unusable responses are excluded rather than treated as permissive. These are aggregate observations of published files, not crawler behavior or the intent behind a declaration.

Across the 14 sealed readings shown for .chat, the newest recorded 29 domains publishing a parseable robots.txt, against an observed low of 28 on 2026-07-29 and a high of 29 on 2026-08-07. In the newest reading, 11 named at least one tracked AI crawler (range 11 on 2026-07-25 to 11 on 2026-08-07), while 4 blocked at least one (range 4 on 2026-07-25 to 4 on 2026-08-07). These are aggregate observations of published files. A movement can reflect changed fetch coverage as well as a changed file; it does not describe motive or crawler behavior.

Observed AI-crawler verdict changes on .chat

Across the sealed archive, 18 crawler-verdict transitions were attributed to .chat domains on 4 sealed change dates. The newest aggregate movements are listed below; no domain name or fetched file body is published.