llms.txt Visible Vocabulary Coverage reports how much of the measured page’s meaningful visible vocabulary also appears in llms.txt. This is lexical overlap — not semantic similarity, factual parity, completeness, or a stuffing target.
| Factor ID | RG-AIX-039 |
|---|---|
| Family | AI access & extractability |
| Measurement | Percentage |
| Measured zone | Validated root llms.txt and visible page text |
What it measures
This factor estimates whether the curated index describes the site in the same language the site actually uses. Low overlap can mean the file is written in generic marketing terms while the pages talk about something specific.
How RankGear measures it
RankGear takes distinct visible-page tokens of at least three characters, removes a small stop-word list, intersects them with the resource tokens, and reports the result as a percentage.
How to optimize it
Describe listed resources with accurate topical language and meaningful labels rather than generic calls to action or keyword lists. The file should read like a map of the site, written in the site’s own vocabulary.
Important considerations
- This is lexical overlap, not semantic similarity, factual parity or completeness.
- It is not a stuffing target — copying page text into the file raises the number and helps no one.
- A page on a topic the file does not cover will legitimately score low.
- This is a measured correlation within a result set, not a guarantee: the value relates to how the pages ranking for your query are built — it does not, on its own, make a page rank.
Related factors
Part of the Factors reference · how RankGear measures · glossary.