What this is
A free, deterministic GEO page-readiness linter: paste a URL and it checks the structural, technical, and rendering signals that decide whether AI answer engines can reach, read, and cite your page — no signup, no AI tokens, no gated score. Updated 2026-06-14.
What we check
Rendering — is your content in the HTML, or only in JavaScript?
The major AI answer-engine crawlers (GPTBot, ClaudeBot, PerplexityBot) do not execute JavaScript. We fetch your page the way they do and measure how much readable text the raw server response actually contains. If your content only appears after client-side rendering, those crawlers see an empty shell — and an empty shell is never cited.
Key check: JS-shell detection: raw server HTML readable-text volume + empty SPA-mount signature.
Crawler access — can the answer engines reach you at all?
Allowing AI crawlers is step zero. We confirm the page responds to an automated request without a hard block, then parse robots.txt for each answer-engine token — OAI-SearchBot, Claude-SearchBot, PerplexityBot, Googlebot, Bingbot. A single Disallow on a citation bot removes you from that engine entirely, no matter how good the page is.
Key check: Per-bot robots.txt allow/block for the five answer-engine crawlers, plus a reachability probe.
Metadata — title, description, and canonical
AI engines use the title and meta description to decide what a page is about before they read it, and the canonical to avoid treating duplicate URLs as separate pages. We check that each is present and within the length ranges that survive truncation and ambiguity.
Key check: Title 15–60 chars, description 50–160 chars, self-referential canonical present.
Content structure — headings and outline
AI parsers infer a page’s structure from its heading outline: one H1 names the topic, and descending H2/H3s mark self-contained sections an engine can lift as a standalone answer. Multiple H1s or skipped levels (H2 → H4) blur that map and make passages harder to extract.
Key check: Exactly one H1, and a heading outline with no skipped levels.
Structured data — JSON-LD
Schema.org JSON-LD is the clearest machine-readable way to state your entities, prices, and Q&A to AI. We confirm it exists, that every block parses as valid JSON (invalid schema is silently ignored), and that your entity declares @id or sameAs so engines can disambiguate it from similarly named ones.
Key check: JSON-LD present, every block valid JSON, and entity identifiers (@id / sameAs) declared.
Extractability — tables, sitemap, llms.txt
LLM fetch pipelines parse semantic <table> markup reliably but strip charts and SVG, so comparable data belongs in a table with a one-line text takeaway. We also check for a discoverable sitemap.xml and, as an early-adopter nice-to-have, an /llms.txt.
Key check: Comparable data in <table> markup; discoverable sitemap.xml; optional llms.txt.
Frequently asked questions
Is the GEO Page Readiness Linter really free, with no signup?
Yes. Paste any URL and get the full deterministic checklist and a 0–100 readiness score instantly, with no email or account required. It runs entirely on verifiable, rule-based checks — no AI tokens are spent — so there is nothing to gate.
How is this different from the main FindingYou.io audit?
The linter is a fast, deterministic pass/fail check of your page’s structure, schema, crawler access, and rendering — the things that are objectively right or wrong. The full audit adds the expensive part: live citation testing across ChatGPT, Perplexity, Gemini, and Claude to measure how often AI answers actually cite you, plus a prioritized action plan.
What is JS-shell detection, and why does it matter so much?
Independent measurement of hundreds of millions of fetches found that GPTBot, ClaudeBot, and PerplexityBot do not render JavaScript. If your content is injected client-side, the raw HTML these crawlers receive is an empty shell, so you are effectively invisible to AI search. We fetch your page the same way they do and measure the readable text in that raw response — a check almost no free SEO tool runs.
Does a perfect score guarantee AI citations?
No. A high readiness score means nothing structural is blocking you — crawlers can reach, read, and parse the page. Whether AI engines actually cite you also depends on relevance, evidence, and corroboration across sources, which is what the full audit measures by sampling live AI answers.