Spec-compliant ยท llmstxt.org

Generate llms.txt for any website

Crawl a site, extract its key pages, and get a spec-compliant llms.txt in seconds.

llms.txt โ€” example output
# FastHTML
> FastHTML is a Python library for building server-rendered hypermedia applications with HTMX and Starlette.
## Docs
- [Quick start](https://docs.fasthtml.com/quickstart.md) A brief overview of many FastHTML features
- [HTMX reference](https://github.com/bigskysoftware/htmx/blob/master/www/content/reference.md) All HTMX attributes, CSS classes, and events
## Examples
- [Todo list application](https://github.com/AnswerDotAI/fasthtml/blob/main/examples/adv_app.py) Complete CRUD walkthrough showing idiomatic FastHTML patterns
## Optional
- [Starlette docs](https://gist.githubusercontent.com/jph00/.../starlette-sml.md) Starlette subset useful for FastHTML development

How it works

01

Crawl

We traverse the site via sitemap.xml, robots.txt, and link discovery โ€” respecting robots.txt rules and prioritizing high-value pages like docs and API references.

02

Enrich

An LLM classifies each page, assigns it to a meaningful section, scores its importance, and writes a concise description โ€” all tuned to the site's domain.

03

Generate

A spec-compliant llms.txt is assembled with a generated preamble, importance-ordered sections, and an Optional section for supplementary pages.

Works for any website

The output is always a spec-compliant llms.txt. Learn more about the llms.txt spec here.

Sign in for more features

Page history

Every llms.txt page you request is saved to your dashboard.

Stay up to date

Download a folder with the latest llms.txt versions in one click.