Quick answer
llms.txt is a proposed convention: a plain text file at your site's root that summarises the site and links to key pages for language models. It is quick to add and can help, but support varies across platforms, and it does not replace crawlable, clear pages.
Key takeaways
- It is a proposal, not a standard every platform follows.
- It is cheap to create and keep updated.
- It supplements good pages. It does not replace them.
What goes in an llms.txt file?
- The name of the site and a one paragraph summary.
- Links to the most important pages, each with a one line description.
- Optional links to fuller plain text versions of key content.
Should you add one?
It costs little and gives any system that reads it a clean overview. Do not expect a measurable lift on its own, and check your logs and results before drawing conclusions.
This site has one
What the proposal is, precisely
llms.txt is a proposal for a plain-text file, written in Markdown, placed at the root of a site. It gives a short description of the site and a curated list of links to the pages most useful to a language model, often with a one-line note for each. Some sites also publish a longer companion file with fuller text of key pages. It was proposed as a convenience: a clean starting point for a system that would otherwise have to work out which pages matter from a busy site.
What the platforms say
Google is explicit that no such file is needed for its AI features. Its guidance says you do not need to create new machine-readable files, AI text files or markup to appear in AI Overviews or AI Mode. Other providers have not committed to reading the file in a documented, universal way. So the honest position is that llms.txt is a low-cost courtesy that some tools and developers may use, and not a lever with a measured effect.
You don't need to create new machine readable files, AI text files, or markup to appear in these features.
When it is worth doing
- Your site is large or complex and you want to point tools at the pages that matter.
- You publish documentation that developers and coding assistants are likely to read.
- You can generate the file from your existing page inventory, so it stays current without extra work.
What to write
| Section | Content | Tip |
|---|---|---|
| Title and summary | The site name and a two or three sentence description | Use the same wording as your organisation description elsewhere |
| Core pages | Product, pricing, security, support | One line each, in plain language |
| Reference | Documentation, guides, glossary | Prefer stable URLs |
| Optional | Fuller text of key pages | Only if you can keep it current |
What it does not replace
The file does not make a page crawlable, indexable or clear. If a page is blocked, slow or vague, listing it in llms.txt changes nothing. Get the pages right first, then add the file as a tidy index. Keep it in sync with the site, because a file that points to removed pages is worse than none.
Do ChatGPT, Gemini and Claude read llms.txt?
Platforms have not committed to it universally. Treat it as a helpful extra, not a requirement.
Is it the same as robots.txt?
No. robots.txt tells crawlers what they may access. llms.txt summarises content for models.
Can llms.txt hurt me?
It is unlikely to, provided it only lists public pages and stays accurate. Do not put anything in it you would not want read, since it is a public file.
How do I know if anyone reads it?
Check your server logs for requests to the file and note the user agents. Do not expect a measurable lift on its own.
Sources
- 1. llms.txt proposal (Accessed Sep 2026)
- 2. Google Search Central: AI features and your website (Accessed Sep 2026)
Cite this page
BrandWater AI Research. (2026, 9 September 2026). llms.txt explained: what it is, who reads it and whether you need one. https://brandwaterai.in/guides/llms-txt-explained
How we work
Figures are dated and linked to their sources. Where none exist we say so. Read our methodology and AI transparency pages.