Skip to main content

A Docusaurus plugin for the llmstxt.org standard

Your Docusaurus docs,
readable by language models

During npm run build, the plugin writes an llms.txt index, an llms-full.txt bundle, and a Markdown copy of every page, so a model reads your docs without parsing HTML.

npm install docusaurus-plugin-llms --save-dev
Get started See this site's llms.txt 
# docusaurus-plugin-llms > llms.txt and Markdown copies of your Docusaurus docs ## Table of Contents - [Overview](https://rachfop.github.io/docusaurus-plugin-llms/docs/overview.md): What the plugin generates and when it runs.- [Installation](https://rachfop.github.io/docusaurus-plugin-llms/docs/installation.md): Install, register, and check the output.- [Configuration options](https://rachfop.github.io/docusaurus-plugin-llms/docs/configuration.md): Every option, type, and default.- [Content cleaning](https://rachfop.github.io/docusaurus-plugin-llms/docs/content-cleaning.md): Strip HTML, imports, and echoes.
build/llms.txtAn index: one link and summary per page.

One line of config

Add the plugin and build. With no options, it reads docs/ and writes both files to the build output. Titles and summaries come from each page's front matter.

All configuration options →
// docusaurus.config.js
module.exports = {
 plugins: ['docusaurus-plugin-llms'],
};

Configure the output

  1. Custom LLM filesOne file per language, product, or audience.
  2. Content cleaningDrop MDX imports, HTML, and repeated headings.
  3. OrderingPut pages in reading order with glob patterns.
  4. Multi-versionA separate llms.txt for each docs version.

This site runs the plugin on its own docs. Read the output: llms.txt · llms-full.txt