llms.txt and plain-text copies of your pages
llms.txt is a proposed file that lists a website’s pages in plain text for AI tools. A plain-text copy is a page’s words in markdown with the layout taken out. Both are cheap to add on top of pages AI crawlers can already read.
The llms.txt file
llms.txt is a markdown file that sits at /llms.txt on a website. It opens with the site’s name and a one-line description, then lists its main pages with a line about each. The proposal is at llmstxt.org.
Add it once AI crawlers can read your pages. It doesn’t stand in for them.
A page in plain text
A plain-text copy is one page’s words in markdown: the headings, paragraphs, lists, tables and links, without the menus, styles and scripts. An AI agent that reads it gets the content without the layout around it.
How bisica.com does it
Every page on bisica.com and on these docs has a plain-text copy:
- At the page’s address with
index.mdadded. This page’s copy is here. - For any request that sends the header
Accept: text/markdown, at the page’s own address. - Listed in bisica.com/llms.txt and docs.bisica.com/llms.txt.
- These docs also have every article in one file at llms-full.txt.
Each page also names its copy in a Link header for agents that read headers.
Add them to your site
- List your main pages in
/llms.txt, grouped under headings, each with a one-line description. - Give each page a markdown copy, generated from the page when it is built so the two never drift apart.
- Keep the copy’s words the same as the page’s. People and AI tools should read the same thing.
- Link the copy from the page with
<link rel="alternate" type="text/markdown" href="...">.
Our site check reports whether your site has an llms.txt file and a plain-text copy of its homepage.
Was this page helpful?
Thanks for telling us.
Thanks. Tell us what was missing at support@bisica.com and we’ll fix the page.