Nothing to hand? Load the — a sourced, answer-first explainer — or the , a launch post with bare numbers, a blocked crawler and nothing an engine could lift. Both replay a saved run for free.
SEO optimises for a ranking. GEO optimises for a citation.
An answer engine does not rank your page — it splits it into passages, embeds them, retrieves a few for whatever a user asked, and writes an answer crediting the sources it can stand behind. So the question is never “where does this rank” but “can one passage of this, pulled out alone, answer something and be credited?” The five factors that decide it — semantic relevance at 40, keyword match at 20, authority at 15, freshness at 10, source diversity at 15 — are the ones this app scores, and they are scored on the same scale twice: once by your browser, once by the model, so a disagreement is visible rather than hidden.
The free read is the deterministic half, and it is most of it
The moment you paste, your browser chunks the article the way a retriever would and flags every chunk that opens on this, it or as described above, or that names nothing at all — retrieved alone, those chunks say nothing. It finds every statistic and classifies each one as sourced, your own original data, softly credited (“studies show”), or a bare number an engine will not repeat. It extracts definitions, named quotes, question headings, FAQ pairs and comparison tables, reads freshness and authorship markers, pulls the JSON-LD @type values out of pasted HTML, and matches your robots.txt against the real crawler user-agents engine by engine. Then it generates the Article, Person and FAQPage schema from what it found. All of that costs nothing and never leaves the page.
The metered pass writes the part a static page cannot
The AI pass supplies the judgement: the 40-to-70-word direct answer an engine can lift, an FAQ phrased as questions users actually type, a severity-ranked finding with a concrete fix for every checklist miss, a rewritten self-contained opening for each dangling passage, and a verdict on every statistic your browser found. It is held to your article: it must return exactly one entry per statistic id, and the page reports any it skipped or invented. Nothing in the answer block or the FAQ may introduce a fact your article does not contain — you would be publishing it under your own byline. Pricing is honest: a worst-case amount is reserved before the run and only what the run uses is charged. Past audits are saved to your account, each scored against the run before, so fixing the article and re-running shows you exactly what the fixes bought.
A derived work of @sickn33/geo-fundamentals, the Generative Engine Optimization skill: its five weighted retrieval factors, its citation-worthy element types and its thirteen-item content and technical checklist are the rubric this app scores against. Perplexity, ChatGPT, Claude and Gemini are trademarks of their respective owners; this app is not affiliated with any of them.
What is Generative Engine Optimization?
Generative Engine Optimization is the practice of structuring a page so that an AI answer engine will retrieve one of its passages and quote it inside a generated answer, with a citation. The engine does not rank the page: it splits it into passages of roughly a hundred words, embeds them, retrieves a few that match whatever a user asked, and writes an answer crediting the sources it can stand behind.
How is GEO different from SEO?
SEO optimises for a ranking position on a results page. GEO optimises for being quoted inside the answer itself. The practical difference is the unit of work: SEO tunes a whole page against a keyword, while GEO has to make each individual passage able to stand on its own, because a passage pulled out alone is all the engine sees. A paragraph that opens on “this” or “as described above” ranks fine and cites terribly.
Which factors decide whether an answer engine cites a page?
Cite Ready scores the five weighted retrieval factors it inherits from the geo-fundamentals skill: semantic relevance at 40 points, keyword match at 20, authority at 15, freshness at 10 and source diversity at 15. The whole hundred is what the readiness score at the top of every report adds up to.
Does auditing my article cost anything?
The browser report is free and needs no sign-in: the five-factor score, the thirteen-item checklist, the statistic list, the retriever passage read, the crawler policy and the generated JSON-LD are all computed on your own machine and nothing is uploaded. Only the AI pass — the answer block, the FAQ, a fix per finding and a verdict per statistic — is metered, and the page reserves a worst-case amount before the run and charges only what the run actually uses.
Is my article uploaded anywhere?
Not until you press the audit button. Pasting, dropping a file and the entire browser report happen locally. When you do run the audit, the article is sent to the model for that run; past audits are saved to your own account so they follow you across devices, and the Clear history button deletes them.
What should I fix first?
Put a direct answer in the opening. An answer engine lifts a lead paragraph that answers the question outright, and skips one that warms up for three paragraphs — it is the single change with the largest effect on citation rate. After that, give every number a traceable source, and rewrite any passage that opens on a pronoun.
Can a robots.txt stop my page being cited?
Yes, and it is the one GEO problem nothing on the page can compensate for: an engine that cannot fetch the page cannot cite it. Paste your robots.txt into the form and each engine’s real user-agents — PerplexityBot, GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, Google-Extended and the rest — are matched against it in your browser, free, and the app hands back a corrected file.