What is llms.txt and should you add one?
A plain-English look at the llms.txt proposal: what it is, what it is not, and whether it earns a place in your site's root today.
llms.txt is a proposed file you place at the root of your site, like robots.txt, but aimed at large language models rather than search crawlers. The idea is simple: give an AI a short, curated map of your most useful pages in clean Markdown, so it can find and quote the right content instead of guessing from a noisy HTML page.
What llms.txt actually is
It is a single Markdown file at /llms.txt. At the top you put your site name and a one-line summary. Below that you list the pages that matter, grouped under headings, each as a Markdown link with a short note on what it covers.
The goal is a curated table of contents an AI can read in one request. Some sites also publish a companion llms-full.txt that inlines the full text of those pages, so a model has the content without crawling each URL.
What it is not
- It is not a standard. No major AI vendor has committed to reading it, so treat it as a bet, not a requirement.
- It is not robots.txt. robots.txt controls access; llms.txt suggests what is worth reading. Neither one enforces anything on a model that ignores it.
- It is not a ranking factor. Publishing one will not move you up in Google or in an AI answer on its own.
Should you add one today
The cost is low and the downside is close to zero, so for a documentation site or a content-heavy product, it is reasonable to ship one. Keep it short, link only to pages you would be happy to see quoted, and keep it in sync when those pages change.
Do not expect traffic from it yet. The honest position in 2026 is that adoption is early and unproven. Add it because it is cheap and tidy, not because it is a growth lever.
How to write a good one
- Start with an H1 that is your site or product name, then a blockquote with a one-sentence summary.
- Group links under clear H2 headings such as Docs, Guides, or API.
- Write each link as a Markdown link followed by a colon and a short description of the page.
- Point only at canonical URLs, and prune anything you would not want an AI to surface.
Treat llms.txt as a courtesy to machines that may read it, not a promise that they will.
If you want a head start, our llms.txt generator drafts one from a list of your key pages, and our guide on robots.txt for AI crawlers covers the access side of the same question.