AI
llms.txt
Machine-readable entry points for agents and language models.
Four routes make the documentation legible to tools, not just browsers.
The index
/llms.txt follows llmstxt.org: a title, a summary, and
link lists grouped by sidebar section. Each link points at the page's markdown
rather than its rendered HTML.
llms.txt
# Docora
> A documentation framework for Next.js.
## Getting Started
- [Introduction](https://example.com/raw/docs/getting-started/introduction.md): What the theme gives you.With i18n on, headings carry the locale so an agent can stay in one language.
The whole corpus
/llms-full.txt concatenates every page — title, URL, description and body —
separated by rules. It suits pasting into a context window in one go.
Both routes are force-static, so they are generated at build time and served
as files.
Raw markdown
Every page is also available as its source at /raw/{path}.md:
Page
curl https://example.com/raw/docs/getting-started/installation.mdWiring it up
app/llms.txt/route.ts
import { createLlmsTxtRoute } from 'docora'
import docsConfig from '../../docs.config'
import { source } from '../../lib/source'
export const { GET, dynamic } = createLlmsTxtRoute(source, docsConfig)Set site.url so the links are absolute; without it they stay relative.