AEO Auditor

ReadyHow it works

Can AI answer engines read your page?

Paste a URL to see what ChatGPT, Claude, Perplexity, Gemini and Google's AI Overviews can reach and read on it: robots.txt rules per AI agent, how each agent's request is answered, the text in the raw HTML, structured data and indexability.

AEO Auditor: a free AI search visibility check

AEO Auditor checks whether AI answer engines (ChatGPT, Claude, Perplexity, Gemini and Google's AI Overviews) can reach, read and understand a page. It reads your robots.txt the way each AI crawler does, requests the page as those crawlers, looks at the HTML they actually receive, and checks structured data and indexability, with a short classic SEO section at the end. It's built for founders and marketing leads who want to know what's in the way before they spend on content.

How it works

  1. Paste a URL

    Enter any public page and press Enter. The audit checks that one page; it never crawls the rest of your site.

  2. We fetch it like the crawlers do

    We read the page's HTML, robots.txt, sitemap and llms.txt, then request the page again as 6 AI search and assistant agents to compare the answers.

  3. Fix what's in the way

    Blockers come first. Each finding shows what we saw, why it matters and the fix, often as robots.txt lines or JSON-LD you can copy.

What it checks

  • AI crawler rules

    robots.txt is read per RFC 9309 for 16 AI agents, grouped into search, user-triggered and training crawlers, with the exact line that decides each one.

  • Live access by user agent

    Status, redirects, size and challenge pages for a browser and each AI agent, so a firewall rule that only hits AI crawlers stands out. Marked approximate, because our requests don't come from the operators' IP ranges.

  • What an AI crawler sees

    Word count, h1 and main content measured in the raw HTML, a text preview of what a crawler that doesn't run JavaScript gets, and detection of empty single-page-app shells.

  • Structured data

    JSON-LD and microdata, checked for Organization, WebSite, SoftwareApplication, Product, FAQPage and BreadcrumbList, with parse errors and missing required properties.

  • Indexability

    Status and redirects, meta robots and X-Robots-Tag (including noindex, nosnippet and noai), canonical, lang, and whether the URL is in your sitemap.

  • Classic SEO and Core Web Vitals

    Title and description length, heading outline, image alt text and Open Graph, plus real-user Core Web Vitals from PageSpeed Insights.

Questions

Is AEO Auditor free?
Yes. There's no account or sign-up. Each network address can run 10 audits a minute, and a repeat run of the same URL within an hour is served from a cache.
Will fixing these get my site into ChatGPT or AI Overviews?
Nobody can promise that. The audit checks whether answer engines can reach and read the page. Whether they cite it depends on the question, your content and their own ranking, which no outside tool can see.
What do Verified and Best practice mean?
Verified means we observed it on this request, such as a Disallow line that matches GPTBot or a 403 for PerplexityBot. Best practice is a recommendation with no confirmed effect on AI answers, such as llms.txt. The score counts verified findings only.
Why is the live access check marked approximate?
Our requests come from Cloudflare Workers, not from OpenAI's, Anthropic's or Perplexity's published IP ranges. Bot management that verifies crawlers by IP may treat the real crawler differently from us, in either direction. Cloudflare zones can also block AI crawlers at the edge; the toggle is under Security, Settings, Bot traffic, AI Crawler and Agent access policies.
Should I block AI training crawlers?
That's your call, and blocking them is a reasonable choice. Training crawlers like GPTBot and ClaudeBot are separate from search agents like OAI-SearchBot and Claude-SearchBot, so you can block training and stay visible in AI search.
Do you store my pages?
Fetched HTML isn't stored. The finished report is cached for an hour so repeat runs are fast, and its findings are kept for 7 days behind a share link, without the page text. Share pages are marked noindex.
Does it check llms.txt?
Yes, as an informational best practice. llms.txt is a proposal, and no major AI provider has confirmed that it uses the file.
Can it audit a staging site or a page behind a login?
No. It only fetches public http and https pages, refuses private networks, never sends cookies or credentials, and follows at most five redirects.