Citare Tools · Free
AI Citation Score
How citable is your site to ChatGPT, Claude, Perplexity, and Gemini? We run 6 readiness checks (robots.txt, sitemap, llms.txt, JSON-LD, schema, brand entity) and score a composite 0-100 + per-category breakdown + a ranked list of fixes.
Free. No signup. Cached 24h per (URL, brand). Takes 5-10 seconds.
Frequently asked
What does the AI Citation Score actually measure?
Six categories of AI-citation readiness, each scored 0-100 and weighted into a composite. (1) Bot access — does robots.txt let AI crawlers in? (2) Sitemap — is sitemap.xml present + complete + has lastmod coverage? (3) llms.txt — is there a curated AI-ingestion index? (4) Structured data — is there JSON-LD on the page + are the schemas valid + are they AI-citation-relevant types? (5) On-page semantics — does the HTML use semantic tags and clear hierarchy? (6) Brand entity — is the brand in Wikidata with proper sameAs links? The composite is 0-100 plus a letter grade A-F.
Is a low score causing my AI citations to fail?
Indirectly, yes — a low score means AI tools have fewer signals to confidently ground citations on your content. AI grounding systems prefer pages with: clean robots.txt (they're not blocked), valid JSON-LD (they can parse entities), a Wikidata entity (canonical identity), recent sitemap with lastmod (content freshness signals), and ideally an llms.txt (curated start-here index). Fixing each issue surfaced in your score improves the upstream signals; AI citations typically take 1-4 weeks to reflect the changes as the indexes (Bing, Google, Brave) refresh their crawls.
How does this differ from Google's Lighthouse score or the Rich Results Test?
Lighthouse measures performance + accessibility + best practices + traditional SEO. The Rich Results Test validates schema for Google's SERP UI specifically. Citare's AI Citation Score is purpose-built for AI-search readiness — it weighs JSON-LD validity for AI-citation use cases (favoring Article + FAQPage + Organization), checks llms.txt (no Google tool does), checks robots.txt against the AI-crawler universe (GPTBot, ClaudeBot, PerplexityBot, OAI-SearchBot — not just Googlebot), and runs brand-entity disambiguation against Wikidata. Run all three for full coverage.
Why does providing a brand name change the score?
Without a brand name, the entity-disambiguation category is skipped and its weight redistributes to the other categories. Providing the brand name lets the tool check Wikidata + Wikipedia for namesake collisions and verify that your site's URL maps to a Wikidata entity with proper sameAs links. For new brands or common-word brands, this category often surfaces the biggest score gap (and the highest-leverage fix).
How long does the score take to compute?
5-10 seconds. The tool runs five outbound fetches in parallel (your page HTML, /robots.txt, /llms.txt, sitemap discovery, Wikidata search) plus deterministic local checks. We cache results for 24 hours per (URL, brand) so subsequent checks of the same site return instantly.