FREE · EVERY RESULT SHOWN · NOTHING UPLOADED

AI readability check

What a machine can lift off your page, and what it cannot. Eight checks, red–amber–green, with the fix for each. Paste your page source below — the whole check runs in your browser and nothing you paste leaves it.

aicitationtools.com · readability check · runs locally

Your page URL — optional. We do not fetch it; it only names the report.

Page source — open your page, right-click, View page source, then copy everything

robots.txt (optional)

llms.txt (optional)

runs in your browser · nothing is uploaded

The report

not run yet

–Run the check to see results—

Generate the fixes the generator writes llms.txt, the robots block and JSON-LD from what you paste here

What each of the eight checks actually tests

thresholds included

Every check is a rule you could apply by hand, so here are the rules. Two of them have a citation-rate comparison behind them — structured data at 36% vs 10%, crawler access at 32% vs 7%. The right column is the rank correlation with being cited across 48 sources; at that sample size anything under 0.285 is not distinguishable from noise, which is why six of these eight are hygiene rather than findings. The tick marks the two that have a rate comparison.

✓Crawler access — any of the six retrieval agents blocked at / in robots.txt is a red.ρ 0.162

–llms.txt is real — a 200 that returns <!DOCTYPE html is a red, and so is a file that does not start with #. No > summary line is an amber.ρ 0.168

✓Structured data — any application/ld+json on the page. Zero is a red.ρ 0.391

–Word count — under 300 is an amber, 300 to 599 a different amber, 600 and up a green. Chinese, Japanese and Korean text is counted per character, not per space.ρ 0.149

–Question-shaped block — a Question or FAQPage type in the JSON-LD, or the words FAQ or frequently asked in the text.ρ 0.095

–Summary block — a blockquote on the page. Without one the answer stays buried in the first paragraph.ρ 0.070

–Meta description — missing is a red, longer than 160 characters an amber.ρ 0.071

–Heading structure — exactly one h1 and at least three h2 to h4. Zero h1 is an amber; several h1 is an amber too.ρ −0.056

Why paste instead of a URL box

read this

A URL box would mean our server fetching your page, which means a copy of your page passing through our infrastructure every time you press the button. We would rather not hold that, and it costs you nothing to paste. The check is a few regular expressions and a word count — it does not need to be a server job.

Two of the eight checks — crawler access and llms.txt — need files that sit outside the page, which is why there are two extra boxes. Leave them empty and those two rows will say so rather than guessing.

The two checks we have actually measured against citation rates are structured data and crawler access. Pages with JSON-LD were cited 36% of the time against 10% without. Pages that allowed retrieval crawlers were cited 32% against 7%. The other six are hygiene: they make the page easier for a machine to lift, and we have not measured whether they move citations.

Questions about this check

eight checks, no signup

What are the eight checks?

Crawler access in robots.txt, whether llms.txt is a real file or a fake 200, structured data, word count, a question-shaped block, a summary block, the meta description, and heading structure. Six come from the page source; the other two need robots.txt and llms.txt, which is why there are separate boxes for them.

Why paste the source instead of a URL?

A URL box means our server fetching your page, which puts a copy of it through our infrastructure every time you press the button. The check is a few regular expressions and a word count, so it does not need to be a server job.

Is anything I paste stored?

No. The script reads the textarea, runs the checks and writes the result into the page. There is no request, so there is nothing to store.

Does a green report mean I will be cited?

No. Only structured data and crawler access are checks we have measured against citation rates. The other six are hygiene: they make the page easier to lift, and we have not measured whether they move citations. We would rather say that than sell you a score.

What if I do not have a robots.txt?

Leave that box empty and the row will say not checked rather than guess. Most sites have one at /robots.txt; if yours returns a 404 the site is open to every crawler by default, which is a different finding from being blocked.