llms.txt: what it is, whether your site needs one and how to make it

llms.txt is a proposal (llmstxt.org): a Markdown file at the site root with a summary and links for AI agents. Google says such files are not needed to appear in its AI search features. According to llmstxt.org, the file can complement robots.txt; in our view it does not replace good pages.

Key points

  • llms.txt is a proposal by Jeremy Howard, published on llmstxt.org on 3 September 2024: a Markdown file at the site root that helps AI agents make sense of a site.
  • Only an H1 with the site's name is required; after it come a short summary in a blockquote, optional detail and H2 sections holding lists of links.
  • Google Search Central (read 9 November 2026): to appear in AI Overviews and AI Mode you "don't need to create new machine readable files, AI text files, or markup".
  • OpenAI's crawler page (read 9 November 2026) does not say ChatGPT reads llms.txt; robots.txt controls OAI-SearchBot's access to a site.

llms.txt is a plain Markdown file placed at the root of a website, at yoursite/llms.txt, that tells an AI agent what the site is and which pages matter. It is a proposal, not a standard that every engine follows. Below is what the proposal says, what four AI vendors say on their own pages (read on 9 November 2026), and what we found when we checked our own file against the format.

What is llms.txt?

The llmstxt.org page, written by Jeremy Howard and published on 3 September 2024 (last modified 10 August 2026), calls it "a proposal to standardise on using an /llms.txt file to provide information to help agents use a website". The file is Markdown: a page of text that a person and a language model can both read without parsing HTML menus and scripts.

What goes in the file?

By the proposal, in this order:

  1. An H1 with the name of the project or site. It is the only required part.
  2. A blockquote with a short summary.
  3. Optional paragraphs or lists with more detail.
  4. H2 sections that hold lists of links. Each link is written [name](url), optionally followed by a note.

An H2 section called "Optional" holds secondary links that an agent can skip when it needs less context. The file can sit at the site root or at any path inside it, covering the pages under that path.

Do the AI engines read it?

The proposal itself says agents "are expected to view or search" the file. It names no AI assistant or search engine that reads it; the tools it lists generate the file, audit for it or let an agent fetch it. What the vendors say, as of 9 November 2026:

Vendor What its own page says about llms.txt
Google AI features and your website does not name llms.txt. It says: "You don't need to create new machine readable files, AI text files, or markup to appear in these features", and that there are no additional requirements beyond being indexed and eligible to show in Search with a snippet.
OpenAI The crawler overview does not say ChatGPT reads llms.txt. It lists OAI-SearchBot (ChatGPT search), GPTBot (training), ChatGPT-User (user actions) and OAI-AdsBot, and says sites opted out of OAI-SearchBot will not be shown in ChatGPT search answers (though they can still appear as navigational links). OpenAI links its own llms.txt as the index of its documentation.
Anthropic The help article on crawling does not mention llms.txt. It names ClaudeBot, Claude-User and Claude-SearchBot and says they honour robots.txt.
Perplexity The bots page links Perplexity's own documentation index at llms.txt. It describes PerplexityBot and Perplexity-User and recommends allowing PerplexityBot in robots.txt. It does not say the bots read a site's llms.txt.

So publishing an llms.txt for your own documentation is common, but none of these four pages says its product reads yours. Google says outright that you do not need such a file. We have not tested whether any engine reads ours, and we do not claim it.

What does that mean for a business in Uzbekistan?

Treat llms.txt as a cheap extra, not as the way into AI answers. Order of work:

  1. Make sure the pages can be crawled and indexed: robots.txt must not block the bots you want (for ChatGPT search, OAI-SearchBot), and Google needs the page indexed.
  2. Put the direct answer in the first lines of each page, in the language your buyers use (Russian, Uzbek).
  3. Only then add llms.txt.

For a small site the file costs little: a short summary and the pages you would send a stranger to.

What we found checking our own file

On 9 November 2026 we compared nusrat.tech/llms.txt with the format above. It has the H1 ("Nusrat Tech") and six H2 sections: Services, Work, People, Pages, Contact and Journal. It lists the newest posts of this journal with their key points. The main differences from the proposal: the summary is a plain paragraph, not a blockquote; the links are bare URLs, not [name](url); the Contact section is plain lines rather than a list; and the links point to HTML pages, not the markdown versions the proposal recommends. So our own file follows the spirit of the format but not all of its letter. All of these are small and fixable.

How to make one

  1. Write the H1 with your business name.
  2. Add a two or three sentence blockquote: who you are, where, what you do.
  3. Add H2 sections (for example Services, Contact) with links as [name](url) and one line of note each.
  4. Put the file at the site root and open yoursite/llms.txt in a browser to confirm it loads as plain text.
  5. Regenerate it when pages change; ours rebuilds its journal section from our posts on every build; the rest we update by hand.

Questions and answers

What is llms.txt in plain words?
It is an ordinary text file in Markdown at yoursite/llms.txt. It says briefly what the site is and links the main pages, so an AI agent does not have to work through the whole menu and page layout.
Does ChatGPT read llms.txt?
OpenAI's crawler page, which we opened on 9 November 2026, does not say so. It says robots.txt controls OAI-SearchBot's access, and OpenAI publishes its own llms.txt for its documentation.
Do I need llms.txt to appear in Google's AI answers?
Not according to Google. Its page on AI features says there are no additional requirements and that you need not create new files for AI: the page has to be indexed and eligible to be shown in Search with a snippet. Google does not guarantee it will be shown.
How do I make an llms.txt?
Write a text file with an H1 (the site's name), a blockquote summary and H2 sections of links written as [name](url), put it at the site root and check that it opens at yoursite/llms.txt.

Sources

Prepared by the Nusrat Tech team with AI assistance, from the sources below.

  1. The /llms.txt file (proposal by Jeremy Howard, published 3 September 2024, modified 10 August 2026) — llmstxt.org, 2026-11-09
  2. AI features and your website — Google Search Central, 2026-11-09
  3. Overview of OpenAI crawlers — OpenAI, 2026-11-09
  4. Does Anthropic crawl data from the web, and how can site owners block the crawler? — Claude Help Center, 2026-11-09
  5. Perplexity crawlers — Perplexity, 2026-11-09
  6. Nusrat Tech llms.txt (our own live file) — Nusrat Tech, 2026-11-09

More notes

  1. How do you add a business to Yandex Maps in Uzbekistan?

    You add a business through Yandex Business or directly on the map: the profile goes to review, then appears in Maps. Yandex's help does not name Uzbekistan on yandex.ru or on yandex.uz, so check the outcome yourself: after review, search for the company in Tashkent Maps.

  2. Why a site loads slowly on a phone: how we took our home page's Lighthouse score from 40 to 80

    A hero video, two WebGL star skies and an animation far down the page were slowing nusrat.tech's home page. We moved their start-up later and in our test (local build, Lighthouse phone profile) the score rose from 40 to 80; this is a lab run, not real-visitor data.

  3. Publishing an iOS app on the Uzbekistan App Store: lessons from two releases

    An App Store app is more than code: the account type sets the seller name, and App Review can send a build back. We went through it twice; here is what to decide before ordering.

All notes

Want this working for your business?

Tell us what you sell and to whom. We answer in Telegram the same day.

Reach usDiscuss your project on Telegram

  • We reply the same working day, Tashkent time.
  • You write to Nodirkhon; he brings in whichever of us your problem needs.