llms.txt Generator

What is llms.txt?

llms.txt is an emerging convention — a plain markdown file served at /llms.txt on your site's root — proposing a concise, LLM-readable map of your key pages. Instead of an AI assistant or crawler wading through navigation menus, ads, JavaScript widgets, and cookie banners to figure out what a site is about, llms.txt gives it a short curated index: what the site is, and links to the sections and docs worth reading.

It isn't a standard enforced by any crawler yet, and it doesn't replace robots.txt or sitemap.xml — those control crawling and discovery mechanics. llms.txt is closer to a README aimed at language models: a human-curated summary rather than an automatically generated list of every URL.

When to use it

It's most useful for documentation sites, developer tools, and content-heavy blogs that want AI assistants (coding copilots, research agents, chatbots with browsing) to cite them accurately instead of guessing from stale training data or a messy HTML scrape. If you maintain API docs, a knowledge base, or a product with a lot of interlinked pages, an llms.txt gives you a lever to say "these are the pages that matter" explicitly.

It's low-effort to adopt: write it once, drop it in your site's public root as /llms.txt, and update it when your site structure changes meaningfully.

Example

For a docs site, fill in Site Name ("Acme Docs"), a one-line Description ("API and SDK documentation for the Acme platform"), a couple of Topics ("REST API", "Webhooks"), and Sections like Getting Started|/docs/start|Setup and authentication and API Reference|/docs/api|Full endpoint reference. The generator produces a markdown file with an H1 title, blockquote summary, and linked sections ready to save as llms.txt at your domain root.