Nuxt LLMs module

The Nuxt Content module integrates nuxt-llms to prepare your content for Large Language Models (LLMs). When nuxt-llms is detected, Content module automatically extends the LLMs module and inject collections of type page to the LLMs module.🚀

Setup

npm install nuxt-llms

export default defineNuxtConfig({
  modules: ['@nuxt/content', 'nuxt-llms'],
  llms: {
    domain: 'https://your-site.com',
    title: 'Your Site Name',
    description: 'A brief description of your site',
  },
})

That's it 🚀 /llms.txt file is automatically generated and pre-rendered.

Raw markdown access

When nuxt-llms is enabled, Nuxt Content also exposes a raw markdown endpoint so you can stream LLM-ready source files without going through the full rendering pipeline.

• Endpoint: /raw/<content-path>.md (use the same path as the page URL, drop trailing /index, and keep the .md extension), returning text/markdown; charset=utf-8.
• Scope: only page collections are included; exclude specific collections with llms.contentRawMarkdown.excludeCollections. Set llms.contentRawMarkdown = false to disable the endpoint entirely.
• Output: if the requested document is missing a top-level heading or description, the route prepends the title and description to the markdown body before returning it.
• llms.txt links: document links generated in llms.txt are automatically rewritten to the /raw/...md endpoint (unless the collection is excluded or the feature is disabled) so agents fetch compact markdown instead of full HTML, reducing token usage and improving response speed. Control this with llms.contentRawMarkdown.rewriteLLMSTxt (defaults to true).

export default defineNuxtConfig({
  modules: ['@nuxt/content', 'nuxt-llms'],
  llms: {
    contentRawMarkdown: {
      // Optional: prevent specific page collections from being exposed
      excludeCollections: ['blog'],
      // Optional: keep llms.txt links pointing to rendered pages
      rewriteLLMSTxt: false,
    },
  },
})

Sections

When generating content, you can create custom sections to process your content into LLM-friendly formats.

You can create custom sections to the llms.sections array and define the contentCollection and contentFilters option for each section.

contentCollection

This option specifies which content collection to use as source.

export default defineNuxtConfig({
  llms: {
    sections: [
      {
        title: 'Documentation',
        description: 'Technical documentation and guides',
        contentCollection: 'docs',
       },
    ],
  },
})

contentFilters

This options defines filters to select specific content within the collection.

You precisely control which content is included. Each filter consists of:

• field: The content property to check
• operator: Comparison operator (=, <>, >, <, LIKE, IN, NOT IN, IS NULL, IS NOT NULL, etc.)
• value: The value to compare against

export default defineNuxtConfig({
  llms: {
    sections: [
      {
        title: 'Documentation',
        description: 'Technical documentation and guides',
        contentCollection: 'docs',
        contentFilters: [
            // Only include markdown files
            { field: 'extension', operator: '=', value: 'md' },
            // Only include published content
            { field: 'draft', operator: '<>', value: true },
            // Filter by directory
            { field: 'path', operator: 'LIKE', value: '/guide%' },
        ]
      },
    ],
  },
})