Guides › Free tools

llms.txt Validator: check your llms.txt file free

Enter a domain to fetch its live /llms.txt and check it against the llms.txt proposal: a single H1 title first, an optional blockquote summary, H2 sections containing Markdown link lists, and links that actually resolve. We also look for /llms-full.txt.

What is llms.txt?

llms.txt is a proposed plain-Markdown file served at the root of a site (https://example.com/llms.txt). It gives AI assistants and agents a short, curated map of the pages that matter, so they don't have to crawl and parse your whole HTML site. It does not control crawling. That is still the job of robots.txt.

The required format

The proposal defines the sections in this order. Only the H1 is required.

1. An H1 with the name of the project or site.
2. A blockquote with a short summary.
3. Optional free-form Markdown (no headings) with more detail.
4. Zero or more H2 sections, each holding a list of [name](url): optional notes links.
5. An H2 named Optional for secondary links that can be skipped when context is short.

# Acme Docs

> Acme is an API for sending invoices. These links cover setup, the REST API and pricing.

## Docs
- [Quickstart](https://acme.com/docs/quickstart.md): create a key and send a first invoice
- [API reference](https://acme.com/docs/api.md)

## Optional
- [Changelog](https://acme.com/changelog.md)

Common llms.txt errors

• HTTP 404 or an HTML page. The file must be served as plain text at /llms.txt. Many frameworks return the app's HTML shell for unknown paths, which is not valid.
• No H1 first. The first non-blank line must be # Title.
• Links outside H2 sections or H3 headings used for file lists.
• Broken links. Every listed URL should return 200.

Does ChatGPT or Google read llms.txt?

There is no public confirmation that major AI search products use llms.txt for ranking or retrieval. Some AI coding tools and documentation platforms do fetch it. Treat it as a low-cost, low-risk addition, not a ranking factor, and keep your robots.txt rules for AI crawlers correct, since those are what crawlers actually obey.

Related