llms.txt endpoint
January 08, 2026
Today after the big Tailwind layoff news, which went viral from this GitHub thread (a PR asking Tailwind to add /llms.txt), I ended up learning something important about how AI interacts with documentation.
The viral thread -> the Tailwind CSS docs PR about adding /llms.txt
So… what even is llms.txt?
What is llms.txt ?
-
It is a plain text file written in markdown placed in the website's root directory, the main purpose is to help the LLMs to understand the site (its structure, its content), making it easy for the LLMs to read , which intrun makes it easy for the LLM to answer the user's questions.
-
It is just like robots.txt or sitemap.xml, only difference is that
- robots.txt is generally used to let automated tools know what access to a site is considered acceptable, such as for search indexing bots.
- sitemap.xml lists all files for search engines to scrape.
-
It looks some what like this
There is also something called as llms-full.txt
- It is basically to provide the entire website content into a single file as content for LLMs. so it might be a huge markdown file.
- For images and all, just a link to the images should be enough, though llms are text first, they would read and answer in texts.
Example -> llms-full.txt
Some pattern or syntax to follow
# Site Full Context
> This file contains full text for selected pages.
---
## /til/ai/llmstxt — llms.txt endpoint
<full content>
---
## /blogs/message-queues — Message Queues
<full content>
To read more about it head over here :