AI Readiness & Discovery / Free validation tool

llms.txt Checker

Check whether your llms.txt is structured for AI agents. Validate the current proposal, review file lists and get straightforward suggestions to make your site easier to navigate.

Aligned with the llmstxt.org v2 proposal, updated August 2026. This is an optional discovery format, not an AI ranking or citation guarantee.

Validate your llms.txt

A domain checks /llms.txt. You may also enter /docs/llms.txt directly.

What is llms.txt and why check it?

llms.txt is a proposed Markdown guide for AI agents and other compatible clients. It introduces a site or documentation area and can point to useful pages with short descriptions. Instead of making an agent sift through navigation and page chrome, the file provides an editorial map of the most relevant material.

The proposal was introduced in 2024 and updated to v2 in August 2026. The current format permits site-wide files at /llms.txt as well as path-specific files such as /docs/llms.txt. The only required section is a single H1 name. A summary, context and H2 topic link lists are optional, although useful in practice.

Adoption is evolving. Publishing the file does not guarantee automatic discovery, citations or improved rankings. It supplements, rather than replaces, search engine sitemaps, robots.txt and helpful HTML pages.

How to read your report

The required-structure result answers whether the file follows the minimum H1 rule and is recognizably Markdown. The guidance score evaluates helpful optional details such as a summary, file lists and clear link formatting. An optional item marked Tip is a recommendation, not a specification failure. Warnings identify structural or formatting issues worth reviewing.

Where to create or update llms.txt

Add a UTF-8 text file at your web root, for example public_html/llms.txt, public/llms.txt or your CMS's public asset directory. If your site uses routes or a CDN, ensure requests to /llms.txt return the file directly with a suitable text content type. For a documentation subsection, use a path-specific file when appropriate.

Include a short summary and a curated selection of durable, publicly accessible pages. Keep it current when key product, documentation or service URLs change. Do not include private URLs or assume that it controls AI crawler access.

Example llms.txt

# Example Website

> Practical resources and documentation for customers.

Find the key guides and documentation below.

## Getting started

- [Overview](https://example.com/overview): Start here.
- [Documentation](https://example.com/docs): Setup guides.

## Optional

- [Blog](https://example.com/blog): Background articles.

This example demonstrates a useful structure. Only the H1 is mandatory under the v2 proposal.

Understand related files

robots.txt communicates crawler access rules.

sitemap.xml lists discoverable URLs.

llms.txt provides curated context and links.

llms-full.txt is an optional ecosystem convention, not a requirement of v2.

Primary sources and further reading

llmstxt.org, proposal v2
The primary source for format, purpose and revisions.
AnswerDotAI, proposal repository
Source files and evolving implementation discussion.
IETF RFC 9309, robots.txt
Authoritative reference for crawler access rules.
Google Search Central, sitemaps
How sitemaps support URL discovery.
This checker assesses file structure and extracted links. It does not confirm which AI systems retrieve the file, test every linked URL, or certify impact on AI-generated answers.

Frequently asked questions

llms.txt is a proposed Markdown index that helps AI agents and compatible software understand a site or documentation area, and find important resources. It does not grant crawler permission.

Choose Check website and enter a domain or exact llms.txt URL, or choose Paste content and submit your Markdown. Review the required structure check first, then recommendations and extracted links.

Required structure assesses the essential H1 and Markdown response. Recommendations cover helpful but optional conventions such as a summary, curated H2 sections, link formatting and file size. The guidance score is not an official compliance certification.

No. The llmstxt.org document is a community proposal. Its v2 revision was published in August 2026. Adoption and tooling are evolving.

No. In v2 the H1 site or project name is the only required section. A blockquote summary, explanatory preamble and H2 link sections are optional but often useful.

Publish a UTF-8 Markdown text file at https://example.com/llms.txt. The v2 proposal also permits path-scoped files such as /docs/llms.txt. Ensure the URL returns the file, not an HTML page.

The H2 heading Optional conventionally groups less important links. It is not a required section, and the proposal does not require every agent to skip it.

No. An llms-full.txt file is an ecosystem convention, not a requirement of llms.txt proposal v2. The proposal also discusses offering Markdown alternatives for individual content pages.

robots.txt communicates crawl access rules to cooperative bots under RFC 9309. llms.txt provides human-readable and machine-readable context and curated resource links. It is not an access control or indexing directive.

A sitemap is mainly for discovering URLs. llms.txt is an editorial guide that highlights important resources with contextual labels and descriptions. Both can be useful for different purposes.

No. Publishing llms.txt does not guarantee that a particular model fetches it, trusts its content, cites your site or ranks it higher. Measure actual crawl traffic and outcomes separately.

No. It validates the downloaded or pasted llms.txt structure and extracts links. It does not fetch every linked URL, which avoids slow scans and unintended traffic.

Warnings flag opportunities to make the file more useful, not necessarily proposal violations. For example, a minimal file with only an H1 can satisfy the sole required heading requirement but still offer little context.

A cooperative crawler still follows applicable robots.txt rules. llms.txt does not override robots.txt, authentication requirements or WAF restrictions.

For a broader website assessment, use our Site Agent Readiness Checker. You can also check AI crawler access or explore other AEO tools.