Root llms.txt Word Count reports the number of letter-and-number tokens in the validated llms.txt body. Unavailable files return 0 and stored content is capped at 2 MiB. Longer is not automatically better.
| Factor ID | RG-AIX-029 |
|---|---|
| Family | AI access & extractability |
| Measurement | Token count |
| Measured zone | Validated root llms.txt |
What it measures
This factor describes the size of the resource rather than its usefulness. Read with the link counts and the section count, it indicates whether a file is a token placeholder, a working index, or an unwieldy dump.
How RankGear measures it
RankGear tokenizes the bounded text response, counting letter-and-number tokens. Files that failed validation return 0, and stored content is capped at 2 MiB.
How to optimize it
Use enough copy to identify the site and describe useful resource groups, but favour clarity, accurate links and maintenance over length. The file exists to route a machine reader to the right material, which is a job measured in precision rather than volume.
Important considerations
- Longer is not automatically better; a concise accurate index outperforms a padded one.
- Compare the query field and inspect the actual resource before changing it.
- Unavailable files return 0, which is indistinguishable here from an empty file.
- 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.