Free AI citation readiness checker
Can AI engines read your page? Find out in about 30 seconds.
Paste a URL, or a list of them. We fetch each page the way ChatGPT, Perplexity, Claude and Google do, then score whether they can reach it, read it and lift an answer out of it, and which fixes would raise the score most.
Free, no account. Every page is scored for four engines:
- ChatGPT
- Perplexity
- Claude
- Google AI Overviews and AI Mode
What you get
One page, read the way AI engines read it.
Every result quotes the line, heading or tag it came from, so nothing in the report has to be taken on trust.
A score out of 100, and a verdict
The mean of the four engines, banded Ready, Mostly ready, Gaps or At risk, with one sentence on where the points went.
Who can fetch the page
Every AI crawler’s robots.txt verdict, quoting the line that decided it, and a live request with each crawler’s user agent.
What survives without JavaScript
The raw HTML beside the rendered page, and the share of the text a crawler that never runs JavaScript actually receives.
How answerable it is
Whether the page answers early, in sections an engine can lift whole, with claims that are specific, sourced and dated.
Fixes, worth the most first
Every fix ordered by the points it recovers, quoting the line on your page it is about, with the effort it takes.
Ready to send to your team
The fixes grouped for the platform owner, the developer and the writer, as messages to copy. Export as PDF, CSV or JSON.
How it works
Four steps. Usually under 30 seconds.
- 1
Paste a URL
One page, or several at once. No account. A page checked in the last day opens its stored report straight away.
- 2
We fetch it as the crawlers do
The page, its redirects and its robots.txt, then a request with each AI crawler’s user agent to see what comes back.
- 3
We render it and compare
A headless browser runs the JavaScript, and the rendered page is set against the raw HTML most AI crawlers stop at.
- 4
Scored, with the fixes in order
26 checks for each of the four engines, then every fix ordered by the points it would recover. The report has its own link.
How the score is built
26 checks in five categories, out of 100.
The same page content always gets the same score. A page that is blocked, excluded from search or empty without JavaScript is capped, whatever else it does well.
Access
Whether an AI engine can reach the page at all.
25 points
Machine readability
Whether the content is in the HTML, or only appears after JavaScript runs.
20 points
Answer structure
Whether the page answers its own question in a form an engine can lift.
25 points
Evidence and credibility
Whether claims are specific, sourced and dated.
20 points
Entity and structured data
Whether the page tells an engine what it is about in machine-readable form.
10 points
What it cannot tell you
Whether the page is actually cited. That depends on search ranking and off-site authority, which this check does not measure. Readiness is what you control; citation is what you earn.
Questions
What people ask before they check a page
What the score covers, and what it does not.
What does “citation ready” mean?
That an AI engine can reach the page, read it without running JavaScript where it does not, and lift a clear, sourced answer out of it. It is the part of being cited that is in your hands.
Which engines does it check for?
ChatGPT, Perplexity, Claude, and Google’s AI Overviews and AI Mode. Each has its own crawlers and its own rules, so the page is scored for each one and the headline score is their mean.
Does a high score mean the page will be cited?
No. Citation also depends on search ranking and off-site authority, which this check does not measure. A low score, though, usually means something on the page is stopping it.
Why does it fetch my page more than once?
Once as it is served, once rendered in a browser, and once with each AI crawler’s user agent, because each of those can get a different page. The difference is often the finding.
How is the score worked out?
26 checks across access, machine readability, answer structure, evidence and credibility, and entity and structured data, weighted 25, 20, 25, 20 and 10.
A page that blocks an engine, is excluded from search, suppresses snippets or is empty without JavaScript is capped, because nothing else on it can reach that engine.
Can I check several pages at once?
Yes. Choose “Several pages” and paste a list. They are checked one at a time, and the run has its own page that fills in as each one finishes, with a CSV or JSON download.
How is this different from Wellows?
This checks whether one page is ready to be cited, once. Wellows tracks whether your pages actually are cited across AI answers, which competitors are named instead, and what changed week to week.