llms.txt generator and validator

What is an llms.txt file?

llms.txt is a proposed standard for a plain Markdown file at the root of your site (/llms.txt). It gives AI tools a short summary of your site and links to the pages that matter most. It was proposed in 2024 and is published at llmstxt.org.

The proposed format is simple:

• An H1 with your site or project name, which is the only required part
• A short summary in a blockquote
• Sections with H2 headings, each a list of links with a short note
• An optional section called "Optional", for links that can be skipped when a reader is short of context

Does llms.txt help with Google or AI answers?

Be clear about this before you spend time on it.

• Google Search ignores it. Google says Google Search, including its generative AI features, doesn't use llms.txt, and that the file will neither harm nor help your visibility or rankings in Google Search. The same guide lists creating unnecessary AI text files like llms.txt among things you can ignore for Search.
• Other systems may use it. Google also says it's fine to keep one for other services that use the file. But when we checked on 5 November 2026, we could not find a major AI provider that publicly documents reading llms.txt on other people's sites.

So treat llms.txt as low-cost and optional. Readable pages, open crawler access and a clean sitemap matter more.

How does the generator build the file?

It reads your sitemap and picks up to 60 pages, most important first, then groups them into sections. If it can't find a sitemap, it uses the links on your home page instead. It takes each page's title and description from the page itself. You then edit the site name, the summary, the sections and the pages before downloading.

Legal, policy, dated and event pages go in the Optional section, unticked. Review and prune the page list before you download, because the automatic selection can still include low-value pages, such as tag or archive pages. Keep the pages that best explain what you do.

The download is a Markdown file in the proposed format:

# Example Ltd

> Example Ltd makes oak furniture and sells online across the UK.

## Products

- [Oak chairs](https://www.example.com/chairs): range, sizes and prices

What does the validator test?

You can give it a URL, or paste the file.

• Format. Whether the file starts with one H1, has a blockquote summary and uses H2 sections of links, and which lines don't follow the structure, such as list items that aren't links, malformed links, relative links and deeper headings.
• Links. Whether each link works, up to the first 100. Broken links are listed so you can fix or remove them, and links that redirect or are marked noindex are flagged.
• How it's served. Whether the URL returns a text file rather than a web page, and whether the file is at the root of the site. It also tells you if you have an /llms-full.txt, a companion file some sites publish.
• Size. It warns when the file is over 100 KB, so you can keep it short and useful.

What do the validator results mean?

Worked example

We ran the validator on a file on a test site we set up, shown here as example.com. The file has 3 links, and one page has moved. One list item reads "- Pricing" with no link:

Error:   1 broken link
Warning: "Products" has 1 list item that isn't a link
Info:    No /llms-full.txt
Passed:  Served as text/plain; charset=utf-8
Passed:  Starts with an H1
Passed:  Has a blockquote summary

The fix is to update the moved URL and change the pricing line to:

- [Pricing](https://www.example.com/pricing): plans and what each includes

What should I fix first?

1. Broken links. A dead link sends AI tools and people nowhere.
2. Format issues. Lines that aren't links are hard for any tool to use.
3. The summary. One or two plain sentences on who you are and what you do.
4. Page choice. List your key pages, not every page. Prune anything the generator picked that doesn't help explain what you do. Your sitemap already lists everything.

What can't the tool tell you?

• Whether any AI system reads your file. It checks the file, not who uses it.
• Pages missing from your sitemap. The generator builds from your sitemap, or your home page links if there isn't one.
• Content that needs JavaScript. The tool reads the HTML your server sends and does not run JavaScript.
• Access rules. llms.txt does not control access. To allow or block AI crawlers, use robots.txt.

How does this compare with other llms.txt tools?

This summary uses each provider's own description, reviewed on 5 November 2026.