/ llmtxt.info

llms.txt checker

The llmtxt.info checker validates your llms.txt file against the official spec. Paste your content or enter your live URL to get instant, line-by-line diagnostics.

Last updated:

What the checker validates

The llmtxt.info checker runs every rule defined in the official spec against your file. Here is what it checks, in order of severity:

Missing H1 (required, must be first element)

The spec requires the file to start with exactly one H1 heading (# Site Name). The H1 must be the very first element — no blank lines, no paragraphs, no comments before it. The checker flags:

  • No H1 in the file at all.
  • H1 that is not the first non-blank element.
  • Multiple H1 headings in the file.

Missing or misplaced blockquote

A blockquote summary (> Description.) is required immediately after the H1. The checker flags:

  • No blockquote in the file.
  • Blockquote that does not immediately follow the H1 (content in between).
  • Blockquote that contains HTML or non-plain-text formatting.

Relative URLs

All links in llms.txt must use absolute URLs beginning with https://. Relative paths (/docs/start) break when the file is fetched by an AI crawler that does not know your base URL. The checker flags every link with a relative or protocol-relative URL.

Malformed link syntax

Each link must follow the Markdown list format: - [Title](https://url): short description. The checker flags:

  • Links missing the leading dash (-).
  • Title or URL not wrapped in the correct brackets.
  • Link items with no description after the colon.
  • Nested bullets (not supported by the spec).

Empty sections

An H2 heading with no list items beneath it wastes space and may confuse parsers. The checker flags any section that contains a heading but no links.

File size

The spec recommends keeping llms.txt under 20 KB (the checker warns at 20 KB and errors above 100 KB). AI systems have context window limits; a bloated index degrades the signal. Put exhaustive content in llms-full.txt instead.

Common errors and how to fix them

Error: "No H1 heading found"

Before (broken):

> My site description.

## Docs

- [Guide](https://example.com/guide): Overview.

After (fixed):

# My Site

> My site description.

## Docs

- [Guide](https://example.com/guide): Overview.

Error: "Relative URL in link"

Before (broken):

- [Getting Started](/docs/start): Install guide.

After (fixed):

- [Getting Started](https://yoursite.com/docs/start): Install guide.

Error: "Blockquote not immediately after H1"

Before (broken):

# My Site

Welcome to My Site. This is a description paragraph.

> My site description.

After (fixed):

# My Site

> My site description.

Welcome to My Site. This is an optional context paragraph.

Warning: "Empty section"

Before (broken):

## Integrations

## Optional

- [Changelog](https://yoursite.com/changelog): Release history.

After (fixed): Either add links to the Integrations section or remove the heading entirely.

FAQ

Is the checker free?

Yes. The llmtxt.info checker is completely free. No account, no API key, no usage limits. Paste your file or enter a live URL and get results instantly.

Can it check a live URL?

Yes. The validator accepts a URL (e.g. https://yoursite.com/llms.txt) and fetches the file, then runs the full spec check — including verifying the Content-Type header.

What format does it expect?

Plain-text Markdown following the llms.txt spec: one H1 at the top, a blockquote summary after, optional H2 sections with unordered lists of absolute-URL Markdown links.

Is there an API?

The validator runs entirely in the browser — there is no public REST API at this time. You can run the same validation rules locally using the open-source llmstxt package on npm, or by integrating the spec rules into your own CI pipeline.

Related

Sources