ffontstand.com

ChatGPT-User is blocked while OAI-SearchBot may crawl.

• ChatGPT-UserOpenAIL80

  A ChatGPT user opening this page

  "Disallow: /" blocks ChatGPT-User.

If that is the intention. If your intention is to block AI entirely, this configuration does not achieve it: the content is still being collected.

Consequence. This is the opposite of what either intent would produce. Blocking bulk collection while allowing people through is a common position; allowing bulk collection while turning away someone who asked for the page is not one anyone sets out to take. It usually means a rule aimed at AI crawlers named the wrong user agent.

If that is not intended. Decide which way round it should be. To keep people out but content in, no change is needed — but that is rarely the intent. To block collection instead, disallow the collecting agents named here and allow the user-triggered ones.

Crawl-delay was never part of the standard. Google has never supported it; Bing and Yandex do.

Consequence. Any rate limit intended for Googlebot is not being applied. If the goal was to reduce load from Google specifically, it is not working.

If that is not intended. Set the crawl rate for Google in Search Console instead. Keep the directive if the limit is meant for Bing or Yandex, which do honour it.

• Crawl-delay: 15L6

A broader rule of the same type already covers these paths.

Consequence. These lines change nothing. They are harmless, but they add length to a file that has a size limit, and they make the real policy harder to read.

If that is not intended. Remove the redundant rules, or narrow the broader rule if it is catching more than intended.

• allow: /framework/thirdparty/L10Already covered by "allow: /" on line 9
• allow: /framework/javascript/L11Already covered by "allow: /" on line 9
• allow: /framework/css/L12Already covered by "allow: /" on line 9

Blocked: Applebot.

• ApplebotAppleL56

  Apple Search and Siri suggestions

  "Disallow: /" blocks Applebot.

If that is the intention. If your intention is to keep the site out of the major search indexes, the current configuration achieves that for these crawlers — the rest of the group is still allowed.

Consequence. These search engines cannot crawl the site, so its pages will not appear in their results. Note that robots.txt does not remove URLs that are already indexed — it stops them being re-crawled.

If that is not intended. Remove the matching Disallow rule, or give these crawlers their own group with an Allow rule. Path matching is longest-wins, so a specific Allow overrides a broader Disallow.

Blocked: PerplexityBot, Google-Extended.

• PerplexityBotPerplexityL134

  Perplexity answers

  "Disallow: /" blocks PerplexityBot.

• Google-ExtendedGoogleL116

  Google AI Overviews and Gemini grounding

  "Disallow: /" blocks Google-Extended.

If that is the intention. If your intention is to keep the site out of AI-generated answers, the current configuration achieves that for these crawlers — the rest of the group is still allowed.

Consequence. The site cannot be cited as a source in AI-generated answers. Unlike training crawlers these send referral traffic, so blocking them forfeits those visits.

If that is not intended. Remove the matching Disallow rule, or allow these agents on the sections you want cited.

Blocked: ChatGPT-User.

• ChatGPT-UserOpenAIL80

  A ChatGPT user opening this page

  "Disallow: /" blocks ChatGPT-User.

If that is the intention. If your intention is to prevent assistants from fetching pages when a person asks for them, the current configuration achieves that for these crawlers — the rest of the group is still allowed.

Consequence. When someone asks an assistant to open a page on this site, it cannot be retrieved. These requests are made because a person asked for that specific page, not to collect content in bulk.

If that is not intended. Allow these agents globally, or only on the sections you want retrievable.

Blocked: GPTBot, ClaudeBot, Bytespider, meta-externalagent, CCBot, Amazonbot.

• GPTBotOpenAIL122

  OpenAI model training

  "Disallow: /" blocks GPTBot.

• ClaudeBotAnthropicL92

  Anthropic model training

  "Disallow: /" blocks ClaudeBot.

• BytespiderByteDanceL68

  ByteDance model training

  "Disallow: /" blocks Bytespider.

• meta-externalagentMetaL23

  Meta model training

  "Disallow: /" blocks meta-externalagent.

• CCBotCommon CrawlL74

  Common Crawl, widely redistributed as training data

  "Disallow: /" blocks CCBot.

• AmazonbotAmazonL44

  Amazon model training and Alexa

  "Disallow: /" blocks Amazonbot.

If that is the intention. If your intention is to prevent this content being used as model training data, the current configuration achieves that.

Consequence. These crawlers cannot retrieve content from the site, so it will not be collected as training data. They send no traffic back, and blocking them does not affect search rankings.

If that is not intended. Remove the site-wide block, or replace it with narrower path restrictions.

All 5 checked social link previews are allowed at the site root.

The request for the page failed (Not requested: the site asks for a Crawl-delay of 15 seconds, and waiting for it would take this analysis past 10 seconds), so its meta robots tags and X-Robots-Tag headers were not read.

Consequence. Nothing is known about whether this page asks to be kept out of search results. Its absence of directives here proves nothing either way.

If that is not intended. Analyse the page again once it answers normally to our bot.

These sitemaps were not seen: each was blocked for our bot, left out by a limit, or did not answer in time. Each line says which.

Consequence. Nothing is known about whether they work. They are not counted against the file.

• https://www.fontstand.com/sitemap.xmlL142Not fetched: it is on another origin, whose robots.txt could not be read

Every line parses as a directive crawlers understand.

Comfortably within the 500 KB that crawlers will read.