AISight

Every scan distills the raw technical findings into a single evidence-based score. Instead of a wall of raw audit output, you get one number that tells you roughly where a page stands for AI visibility. The score is meant to be directional, so use it to track progress across rescans rather than to compare unrelated pages.

This is where a lot of pages quietly fail. AISight checks whether AI and search crawlers can technically reach the page at all. It reads index headers, parses robots.txt, and looks for a source-guidance file such as llms.txt. A single bad Disallow rule aimed at AI bots can keep your best content out of every answer engine, and this check surfaces it fast.

Once a crawler reaches a page, the content still has to be organized well enough to parse. AISight extracts the H1 to H4 heading hierarchy, looks at entities, and reviews structured data. If your meaning is buried in design or split across vague headings, an AI system guesses, and guessing leads to misattributed facts.

Citation readiness measures how easily an answer engine can pull and quote your own factual statements. Pages full of marketing language give an AI nothing to cite, while pages with clear, self-contained claims are easier to attribute. This feature tells you whether a page is quotable or just readable.

The report doesn't stop at what's wrong. It ranks issues by expected impact and pairs each one with copy-paste implementation guidance. You get three things from a scan: what's measurably broken, how much fixing it's expected to help, and what to tackle first. That ordering matters when you can only touch a handful of pages this week.

There's no login, no email, and no signup wall. You get five free scans per hour, which is plenty for spot-checking a few pages in a sitting. The trade-off is that nothing is saved across sessions, so note your scores yourself if you want a history.