Weak AI visibility with 8 of 22 criteria passing. Biggest gap: llms.txt file.
Verdict
refine.dev currently shows low AEO readiness with an overall score of 35/100, driven by major machine-consumption gaps rather than crawlability issues. Core discoverability signals are missing, including llms.txt (0/10), structured data coverage (0/10), FAQ signals (0/10), RSS/Atom (0/10), and AI permissions/licensing signals (0/10). At the same time, technical foundations are partially strong with clean crawlable HTML (9/10), semantic structure/accessibility (8/10), and canonical strategy (10/10), which means improvements can compound quickly once structured answer-layer signals are added. The biggest constraint to AI citation appears to be weak answer formatting and freshness metadata, not indexability.
Scoreboard
Fix It With AI
Copy-paste these prompts into Claude Code or Cursor to fix each criterion.
These prompts are designed for projects where you have direct access to the codebase (Next.js, React, static HTML, WordPress, etc.). If your site runs on a hosted platform like Webflow, switch to the "Webflow" tab for platform-specific instructions. Using a different hosted platform? Contact us for custom guidance.
Top Opportunities
Improve Your Score
Guides for the criteria with the most room for improvement
Tidio has a 251-line llms.txt. Crisp has zero. The score gap: +29 points. This single file tells AI assistants exactly what your site does -and without it, they're guessing.
Tidio runs 4 JSON-LD schema types. Crisp runs zero. That's not a coincidence -it's the difference between a 63 and a 34. Structured data is the machine-readable layer AI trusts most.
Our site runs 87 FAQ items across 9 categories with FAQPage schema on every one. That's not excessive -it's how we hit 88/100. Each Q&A pair is a citation opportunity AI can extract in seconds.
Sitemaps tell crawlers what exists. RSS feeds tell them what changed. If you don't have one, your new content waits days -or weeks -to be discovered.
Want us to improve your score?
We build citation-ready content that AI engines choose as the answer.