Skip to content

Glossary

What Is llms.txt

llms.txt is a community proposal placing a Markdown content index for large models at the site root.

Author
Xingge
Published
2026-11-09
Updated
2026-11-09
Level
L2 Reproducible
Version
1.0

Term details

Name
What Is llms.txt
Type
Glossary
ID
llms-txt-en

What Is llms.txt

llms.txt is a community proposal that places a Markdown file at the site root to give large models a content index. It is not a W3C or IETF standard, its crawling effect requires measurement, and no one should claim it is certain to be indexed.

Definition

This site's working definition: llms.txt is a site description file generated by hand or at build time, listing core pages, the evidence index, and citation suggestions in a fixed structure to lower the model's comprehension cost.

This site generates it from the content manifest during the build, without manual maintenance, so that it cannot drift from what is actually published.

Boundaries: what it is easily confused with

Easily confused conceptKey difference
robots.txtrobots governs whether crawling is allowed; llms.txt is only a guide
sitemap.xmlA sitemap is a crawl entry point; llms.txt is a semantic summary
Official standardllms.txt is a community proposal with no standards status

Common misconceptions

  • Treating llms.txt as an indexing switch. It is a guide file and does not change crawl permissions.
  • Maintaining it by hand until links expire. It should be generated by the build.
  • Claiming indexing is certain once it is provided. The effect requires measurement and depends on model and platform policy.

Basis and verification

This definition is this site's working definition, marked L2. The actual file structure and generation rules are on the machine-readable layer method page in the method index, and experimental status is in the research index.

Review cycle: 180 days, moved earlier when the community proposal status changes.

Related sections: glossary index, method index, metric index, editorial policy.

Evidence

Author

Xingge

GEOA method research, focused on how Chinese LLMs cite and describe brands accurately.

Entity https://shingle.com.cn/#person

Author profile