Is llms.txt worth adding to your website?

For most organization websites, not yet. llms.txt is a proposed file meant to help AI tools read a site, but none of the major AI assistants has said it uses the file, Google says you do not need it, and in the largest study so far 97% of llms.txt files received no requests at all in a month. It does little harm and takes little time, so if your site publishes technical documentation it may be worth a small experiment. For everyone else, there are better uses of the hour.

What is llms.txt?

It is a plain-text file, written in Markdown, placed at the root of a website (yoursite.ca/llms.txt). It summarizes the site and links to the pages an AI tool should read first. It was proposed by Jeremy Howard of Answer.AI in September 2024 (llmstxt.org). It is a proposal, not a standard, and no AI company is obliged to follow it.

The idea is sensible. The question is whether anything reads it.

Do AI assistants read llms.txt?

On the evidence so far, barely.

Google says you do not need it. Its guidance on AI Overviews and AI Mode states: “You don’t need to create new machine readable files, AI text files, or markup to appear in these features” (Google Search Central). Google’s John Mueller has compared llms.txt to the old keywords meta tag, which search engines stopped using long ago (Search Engine Journal, April 2025).
OpenAI and Anthropic document their crawlers and how to control them through robots.txt; neither page mentions llms.txt (OpenAI, Anthropic).

What does the data show?

In June 2026, Ahrefs published an analysis of one month of server logs (May 2026) across 137,210 websites that use its analytics (Ahrefs, llms.txt study):

About 28% of the sites had an llms.txt file.
97% of those files received zero requests during the month.
Most requests that did arrive came from automated bots, not named AI assistants. The crawlers behind ChatGPT search, Perplexity and Claude’s search together made only a couple of hundred fetches.
Ahrefs’ conclusion: the file may help coding tools that read developer documentation, but it does little for visibility in AI search.

The sample leans towards sites run by people who pay attention to search, so if anything it overstates how much llms.txt is used.

When might it be worth adding?

You publish technical documentation that developers read through AI coding tools. This is where the Ahrefs data found some use.
It costs you almost nothing. If your site can generate one automatically and keep it accurate, there is little downside.

If you add one, keep it accurate. An out-of-date summary is worse than none.

What should you do instead?

The things the platforms say they use:

Let AI search crawlers in. Sites that block OpenAI’s OAI-SearchBot are not shown in ChatGPT search answers (OpenAI). Check robots.txt and your host or CDN settings.
Get the basics of search right. Clear page titles, one topic per page, fast pages, and content that answers the question in the first paragraph.
Say plainly who you are and what you do on your home and About pages, the same way everywhere.
Earn mentions from directories, associations, partners and press.

We go through each of these in Can you get ChatGPT to recommend your business?

Will this change?

Possibly. If a major assistant announces it reads llms.txt, the answer changes and so will this article. Until then, treat it as an experiment, not a requirement, and be wary of anyone selling it as a way to be recommended by AI.

If you would like us to check how AI search tools see your site, schedule a discussion.

Sources (checked 30 September 2026)