What is llms.txt?
llms-full.txt, LLMs.txt file
llms.txt is a proposed standard for a Markdown file placed at the root of a website, such as example.com/llms.txt, that gives language models a short summary of the site and links to its most useful pages. Jeremy Howard proposed it in September 2024. It is cheap to add, but no major AI engine has confirmed using it.
llms.txt is a Markdown file that summarizes a site for language models.
No major AI engine has confirmed that it reads llms.txt when answering.
It costs little to add, but crawlability and content matter far more.
How llms.txt works
llms.txt works as a curated map of a website written for language models rather than for browsers or search crawlers. The file lives at the site root and uses plain Markdown, which models read easily.
The proposed format has a simple structure:
An H1 line with the name of the site or project.
A short blockquote summarizing what the site is about.
Optional paragraphs with extra context.
Sections of links to key pages, each with a short note on what the page covers.
The proposal also suggests offering Markdown versions of important pages, and many sites add an llms-full.txt file containing the full text of their documentation in one place. The idea is that a model, or a developer tool, can load the most relevant content without parsing navigation, ads and scripts. It sits alongside XML sitemaps, which list URLs for crawlers but say little about them.
llms.txt example
An llms.txt example for a hypothetical software company could look like this, with each item on its own line in the real file:
# Acme Analytics> Acme Analytics is a product analytics tool for B2B SaaS teams.## Product- [Pricing](https://example.com/pricing): plans and what each includes- [Integrations](https://example.com/integrations): supported tools and setup## Docs- [Quickstart](https://example.com/docs/start): install and send your first event
Keep descriptions short and factual, link only to pages you want models to read, and update the file whenever key pages move or change. Make sure every linked page is also reachable by AI crawlers, since the file only points to content and does not deliver it.
llms.txt vs robots.txt
llms.txt and robots.txt do different jobs. robots.txt tells crawlers which URLs they may or may not fetch, and major search and AI crawlers follow it. llms.txt does not grant or block access. It only suggests which content is most useful and describes it.
The other difference is adoption. robots.txt has been a web convention for decades and is formally documented as a standard. llms.txt is a community proposal, and no major AI engine has confirmed that it uses the file to choose sources or shape answers.
For AI search, robots.txt for AI is the control that actually matters. If robots.txt or a firewall blocks the bots that power ChatGPT search or Perplexity, no llms.txt file will help, because the engines cannot fetch the pages it points to.
Should you add an llms.txt file?
Adding an llms.txt file is reasonable if it takes little effort, but it should not come before basics that are proven to matter. The file is low cost and low risk. It can help developer tools and AI coding assistants that do read it, which is why many documentation sites publish one.
For marketing visibility in ChatGPT, Perplexity or Google AI Overviews, focus first on:
letting search and AI crawlers reach your pages
serving key content in HTML, not only through JavaScript
clear, answer-first pages for the questions buyers ask
accurate structured data
Once those are in place, an llms.txt file is a tidy extra. Treat it as a hint for machines, not a ranking lever.
How KIME fits llms.txt into AI web health
KIME AI Web Health looks at how accessible and readable your site is for AI engines, covering the technical factors that affect whether pages get retrieved and cited. Rather than treating llms.txt as a fix on its own, KIME helps you prioritize the issues with the clearest impact first, such as crawler access and rendering, and turns them into technical tasks in Actions that your team or AWX agents can complete.
Frequently asked questions
Does ChatGPT read llms.txt?
OpenAI has not confirmed that ChatGPT uses llms.txt to find sources or write answers. ChatGPT search relies on its own crawlers and search index, which follow robots.txt. ChatGPT could load the file if a user asked it to, but there is no evidence that having one makes ChatGPT cite a site more often.
Where do I put an llms.txt file?
Place llms.txt at the root of your domain so it loads at a URL like example.com/llms.txt. Serve it as plain text with Markdown formatting, not as an HTML page. If you also publish llms-full.txt or Markdown versions of key pages, link to them from the main file and keep all of them updated when content changes.
Can llms.txt block AI from using my content?
No. llms.txt is not an access control. It only suggests which content is useful and describes it. To stop specific AI crawlers from fetching your pages, use robots.txt rules for each bot's user agent, and back them up with firewall or CDN settings if you need a hard block rather than a request.