Crawlability
Reads robots.txt for the AI crawlers of OpenAI, Anthropic, Perplexity, Google, Bing, Common Crawl and Apple, then checks sitemap.xml and llms.txt exist, validate and are served as the right content type.
Run a practical AI-readiness audit in seconds and see exactly what to fix first — without creating an account.
3 fixes would move this to “Ready to be cited”.
The extension looks for the technical and content issues that make a page harder for AI assistants and search engines to access or interpret. Rules come from OpenAI, Anthropic, Perplexity and Google documentation and from published studies; the weight of each check follows the strength of its evidence.
Reads robots.txt for the AI crawlers of OpenAI, Anthropic, Perplexity, Google, Bing, Common Crawl and Apple, then checks sitemap.xml and llms.txt exist, validate and are served as the right content type.
Finds JSON-LD and checks it parses, uses the right types for the page (Organization, Service, Product, FAQPage) and doesn't contradict the visible text.
Title and description length, canonical, language, and a complete Open Graph set so the page is indexable and previews correctly.
One H1, a sane heading hierarchy, an opening paragraph that answers “what is this page” in under 60 words, and a Q&A block with direct answers.
Hard for AI to read (0–39) · Partly readable by AI (40–69) · Readable by AI (70–89) · Ready to be cited (90–100). Blockers first, then improvements. The score is a diagnostic guide, not a promise of inclusion or ranking.
Open the extension on any page and start the check. It reads the page and the site's public files from your browser, then builds the report locally.
Website data stays on your device throughout the audit.
Each failed check links to a short explanation with the exact file or tag to change. Most fixes are a few lines in robots.txt, a corrected sitemap, a JSON-LD block or a rewritten opening paragraph.
Send me the address. I go through your pages myself and tell you what is worth fixing — free — then price the ones you choose and verify them with the same check. See the fix service →
The next version asks ChatGPT, Claude, Gemini and Perplexity about your brand and reports mentions, share of voice against competitors and a weekly history. Enter your email and we'll send one message when it's ready.
The form could not be sent. Send your address by email instead: ivan@flowpro.dev.
One launch email only. No newsletter. You can ask for deletion at any time — see the privacy policy.
Whether a page and its site files meet the documented preconditions for AI systems to access and interpret them: crawler permissions, valid files, structured data, complete metadata and readable structure. It does not query any AI assistant.
Yes. The Chrome extension is free to install and use, with no account.
None from the pages you audit. Checks run locally; the only network requests go to the audited site's own origin for robots.txt, sitemap.xml and llms.txt. Joining the waitlist submits your email to flowpro.dev, covered by the site privacy policy.
Whether AI assistants such as ChatGPT, Claude, Perplexity and Gemini can access your pages, understand what your business does and use them as sources. The terms AEO and GEO describe the same work; there is no agreed definition, and Google treats it as part of SEO.
Not by itself. A high score means your side is done: nothing technical stops an assistant from reading you. Mentions also depend on what other sites say about you and on each platform's own selection.
GPTBot (OpenAI), ClaudeBot (Anthropic) and PerplexityBot are the ones whose block is reported as a failure: block them and those assistants cannot read the site at all. OAI-SearchBot, Claude-Web, Google-Extended, Bingbot, CCBot and Applebot-Extended are checked too and reported as a warning, because blocking them narrows where you can appear rather than shutting you out entirely.