llms.txt Explained

llms.txt is a proposed file that gives AI systems a short, plain-text guide to your site. It is easy to make and costs nothing. It is also unproven: no major AI company has confirmed that it uses the file for search or ranking. Here is what it is, and how to decide whether to bother.

Where it came from

The proposal was published in 2024 by Jeremy Howard of Answer.AI. It suggests putting a Markdown file at /llms.txt that summarizes the site and links to the key pages, so that a language model with limited context can find the important material quickly (llmstxt.org). It is a community proposal, not an internet standard, and not a product of any one AI company.

The format

According to the proposal, the file contains, in order:

1. An H1 with the name of the project or site (required).
2. A blockquote with a short summary.
3. Optional paragraphs or lists with more detail, without headings.
4. Sections under H2 headings, each a list of links in the form [title](url): optional notes.
5. An optional section named "Optional" for links that can be skipped when context is short.

A template for a local business

# Example Roofing Co.

> Example Roofing Co. is a licensed roofing contractor serving Winter Park,
> Maitland, and Orlando, Florida. We repair and replace residential roofs.

Founded in [year]. License: [your real license number]. Phone: [your number].

## Services
- [Roof repair](https://example.com/roof-repair): leaks, storm damage, flashing
- [Roof replacement](https://example.com/roof-replacement): shingle, tile, metal
- [Free inspections](https://example.com/inspection): what is checked and how long it takes

## Pricing and process
- [How pricing works](https://example.com/pricing): factors that change the cost
- [Our process](https://example.com/process): steps from quote to cleanup

## Company
- [About us](https://example.com/about)
- [Service area](https://example.com/service-area)
- [Contact](https://example.com/contact)

## Optional
- [Blog](https://example.com/blog)

Replace every bracket and example URL with real facts. Do not claim awards, license numbers, or years in business that are not true.

Does it work?

Be honest with yourself here. At the time of writing we know of no confirmation from OpenAI, Anthropic, Google, or Perplexity that their search products read /llms.txt to decide what to cite. Some developer tools and documentation sites use it, and some AI coding assistants can read it when pointed to it. For a local business, treat it as a harmless extra, not a lever.

If the major vendors adopt it, you will be ready. If they do not, you spent twenty minutes. The cost-benefit is fine, but do not pay a consultant a lot for it and do not expect a jump in mentions.

What matters more

Everything in an llms.txt file should already be on your site in normal pages. If it is not, fix the site first. The summary file only helps if the pages it links are clear and complete. The things that more plausibly affect AI visibility are:

• Allowing the right crawlers in robots.txt (guide).
• Pages that state facts plainly.
• Structured data that matches the page (guide).
• Consistent, accurate listings on third-party sites.

How to publish it

1. Write the file in a plain text editor and save it as llms.txt, UTF-8.
2. Upload it to the root of your site so it loads at https://yourdomain.com/llms.txt.
3. Make sure it returns a normal 200 status and plain text, and that robots.txt does not block it.
4. Update it when your services or pages change.

On WordPress, a few plugins generate the file. Check that the output is accurate before leaving it on, since generated summaries can include pages you do not want promoted.

llms-full.txt

The proposal also mentions a larger variant that includes the full text of key pages. This is more useful for documentation sites than for a local service business. If you publish one, keep it accurate and in sync with your site.

Our own file

We publish an llms.txt for this site as an example. It follows the format above.

Check the rest

SearchSignal's full report audits whether you have an llms.txt and whether it is well formed, alongside robots.txt rules, structured data, and content inventory. It reports these as findings, and does not claim any of them will cause a mention. Start with the free check to see where you stand, and read the methodology for how we score.

Common questions

Is llms.txt the same as robots.txt?

No. robots.txt tells crawlers what they may fetch. llms.txt is a content guide for language models. One does not replace the other.

Can llms.txt hurt me?

Unlikely, as long as it is accurate. A wrong or outdated file could mislead a system that reads it, so keep it in sync with your site.

Do I need to include every page?

No. Link the pages that best describe what you do, and put the rest under Optional.

Is there an official validator?

Not that we know of. Check that the file loads, is plain Markdown, and follows the structure on llmstxt.org.

Published 2026-10-30. We cite vendor documentation where we state facts about it; vendors change products often, so check the linked pages for current details. This guide describes what you can influence and does not promise any result.