Free tool

Can answer engines read your site?

Paste an address. Pemberley checks the surfaces ChatGPT, Claude, and Perplexity actually read: crawler access, feeds, sitemaps, llms.txt, and the metadata they quote.

What the check looks at

Nine checks in three groups. Whether machines can read you: HTTPS, robots.txt access for the AI crawlers (GPTBot, ClaudeBot, PerplexityBot, Google-Extended), and whether your words arrive in the raw HTML rather than rendering client-side. Whether they can find your work: a sitemap, a working RSS or Atom feed, and llms.txt. Whether they can quote you: the title and description, OpenGraph tags, and JSON-LD structured data.

Everything is checked from the outside, the way a crawler sees it. No account. Pemberley keeps a daily count per site to prevent abuse, and stores nothing else about you or your pages.

What to do with the results

Most fixes here are an afternoon of work: a robots.txt edit, a feed plugin, a meta description that says what you actually do. For the longer game, being the source answer engines cite when your field comes up, start with how ChatGPT decides which newsletters to recommend.