# ChannelOS — https://channelos.tv # # ONE group, on purpose. Under RFC 9309 a crawler obeys only its most specific matching # group and ignores every other, so the seven per-agent `Allow: /` groups that used to sit # below this one had the effect of hiding the Content-Signal line from exactly the crawlers # it is addressed to: GPTBot would have matched `User-agent: GPTBot`, read one `Allow`, and # never seen the signal. Those groups existed to counter Cloudflare's managed robots.txt, # which prepended `Disallow: /` for the same agents. That setting was turned off for this # zone on 2026-08-04, this file is now the whole document, and nothing here disallows # anything, so naming agents individually buys nothing and costs the signal. # # The named answer engines are still welcome, and every one of them is covered by `*`: # GPTBot, OAI-SearchBot, ChatGPT-User, PerplexityBot, Perplexity-User, ClaudeBot, # Claude-SearchBot, Claude-Web, Google-Extended, CCBot, Amazonbot, Applebot-Extended. # # Content-Signal (contentsignals.org) declares how the content may be used, separately from # who may fetch it. Open on all three standard signals: the job of these 53 posts is to be # found and quoted, and refusing training buys no traffic while shrinking the surface. # `use=reference` is Cloudflare's experimental fourth field and is deliberately NOT `full`: # reference means index, excerpt and link back, and the link back is the entire point. User-agent: * Content-Signal: search=yes, ai-input=yes, ai-train=yes, use=reference Allow: / Sitemap: https://channelos.tv/sitemap.xml # Plain-language summary of the product, pricing and every public URL, for # answer engines. Kept in step with /pricing and the billing catalogue. # https://channelos.tv/llms.txt # # Every blog post also has a markdown twin at the same URL plus `.md`, linked from each # post's HTML with .