Blog · Answer Checkup team · Last checked

llms.txt explained: what it is and how to write one

llms.txt is a short markdown file at the root of your site, at /llms.txt, that tells AI agents what your site is and which pages are worth reading. Think of it as a curated table of contents written for a language model instead of a person. It's cheap to add. It is also not a ranking switch.

Where llms.txt comes from

The format is a proposal by Jeremy Howard, first published on llmstxt.org in September 2024 and revised as version 2 in August 2026. The reasoning is practical. Web pages wrap their content in navigation, scripts and layout, which makes them hard for a model to read cleanly, and a model's context window can't hold a whole website. A small file that summarizes the site and points to clean pages saves the agent work.

The proposal says thousands of sites now publish one, that some documentation platforms and CMSs generate it automatically, and that Chrome's Lighthouse checks for one as part of its agentic browsing audits. OpenAI and Perplexity both publish llms.txt files for their own developer documentation: OpenAI's and Perplexity's.

What goes in the file

The spec is short. In order:

  1. An H1 with the name of your site or business. This is the only required part.
  2. A blockquote with a one or two sentence summary.
  3. Optional paragraphs or lists with more detail, but no headings.
  4. Optional H2 sections, each a list of links in the form [name](url): note.

By convention, a section called ## Optional holds links an agent can skip when it's short on space.

Here's a template for a small business. Replace everything in brackets and keep every statement true and visible on your site.

# [Business name]

> [Business name] is a [what you are] in [city or service area]. We [what you do] for [who you serve].

- Hours: [hours, as shown on your contact page]
- Contact: [email or phone shown on your site]

## Main pages

- [Services](https://example.com/services): what we offer and how it works
- [Prices](https://example.com/prices): current prices
- [FAQ](https://example.com/faq): answers to common questions
- [Contact](https://example.com/contact): how to reach us

## Optional

- [About](https://example.com/about): who runs the business

The proposal also suggests offering clean markdown versions of key pages at the same URL with .md added, and linking them with rel="alternate" type="text/markdown". That's worth doing if your platform makes it easy. If it doesn't, skip it.

What llms.txt does and doesn't do

Here is what the sources say.

It helps agents that look for it. The proposal describes llms.txt as information "used on demand, when an agent needs information about a topic while assisting a user." A coding agent or a research agent that checks /llms.txt gets a fast, accurate overview.

Google says it doesn't affect Google Search. Google's guide to optimizing for generative AI features lists creating "unnecessary AI text files (like llms.txt)" among tactics you can ignore for Google Search. Its June 15, 2026 documentation update says these files aren't needed for Google Search and won't positively or negatively affect visibility or rankings there, but that it's fine to keep them for other systems.

No engine has said it uses llms.txt to choose citations. OpenAI and Perplexity document which crawlers they use and how to allow them (OpenAI, Perplexity). Neither page says llms.txt changes whether you get mentioned or cited. Publishing their own docs as llms.txt is not the same as reading yours when they answer a question about plumbers.

What happened when we had one

Our previous site, showsupin.com, was a one-page "launching soon" placeholder. Its llms.txt had a name, a short summary and a single link, to the homepage. Its robots.txt allowed every crawler:

User-agent: *
Allow: /

Both were in place when we ran our report. Keep that in mind: this was a tiny placeholder, not a full site test.

On October 15, 2026 at 12:47 AM ET we ran our own report on that site. It asked 25 buyer questions, 3 times on each engine, through the OpenAI API with web search and the Perplexity API, for 150 answers. The site scored 0 out of 100. None of the 150 answers mentioned it, and none cited it.

The answers did cite plenty of pages: 510 citations to 294 different pages on 199 domains. Those were ordinary articles, review pages and comparison posts, and only 19 of 510 citations went to the competitors' own sites. For a new site with no coverage elsewhere, having one wasn't enough on its own: the site wasn't mentioned or cited even once. We didn't test the same site without the file, so this doesn't show llms.txt does nothing. It fits what the sources above say. It's a helpful file for agents, not a reason for an engine to recommend you.

We still keep one, and we'd tell you to add one. It takes minutes, it makes your site easier for agents to understand, and it gives you a single place to state your facts plainly. Just don't expect it to fix visibility on its own. Our AI SEO guide covers the steps that matter more, starting with crawler access and coverage on other sites.

How to write yours in about 15 minutes

  1. List your five to ten most useful pages. Services, prices, FAQ, contact, about. Skip blog archives and tag pages.
  2. Write the summary first. One or two sentences: what you are, where, for whom. Use the same business name, spelled the same way, as everywhere else. Inconsistent names confuse engines. In our own report, 28 of the 66 answers to questions that used our old name brought up a different product with a similar name.
  3. Add facts only if they're on your site. Hours, service area, contact. If a fact changes, update the page and the file together.
  4. Use full URLs. https:// links, not relative paths.
  5. Upload it to the root so it loads at https://example.com/llms.txt as plain text.
  6. Test it. The proposal suggests giving an agent only your llms.txt and asking it questions about your business. If it gets things wrong, the file is unclear.
  7. Make sure robots.txt doesn't block it for the bots you want to read it. Our robots.txt guide for AI crawlers shows how.

Common mistakes

  • Promises the site doesn't make. If your pages don't say it, the file shouldn't either.
  • A sitemap dump. Hundreds of links defeat the purpose. The proposal contrasts llms.txt with sitemap.xml for exactly this reason: a sitemap lists everything, llms.txt picks what matters.
  • Stale facts. An old price or old hours in llms.txt is worse than no file.
  • Treating it as a substitute for readable pages, structured data or real reviews. It isn't one.

FAQ

What is llms.txt?
A markdown file at /llms.txt that summarizes a website for AI agents and links to its most useful pages. It's a community proposal published at llmstxt.org, not a formal web standard.
Does llms.txt help me show up in ChatGPT or Perplexity answers?
Neither OpenAI nor Perplexity has said it does. Our old placeholder site had one and scored 0 out of 100 in our report. Add it for agents, and work on crawler access and outside coverage for visibility.
Does Google use llms.txt?
Google says llms.txt isn't needed for Google Search and won't help or hurt rankings there.
Is llms.txt the same as robots.txt?
No. robots.txt says which bots may crawl which paths. llms.txt describes your content for agents that choose to read it. You want both.

Check your own site

Run the free scan on the Answer Checkup homepage. It checks whether you have an llms.txt file, whether AI crawlers can reach your site, and whether your structured data is in place. For the bigger picture, read our guides to generative engine optimization and checking whether ChatGPT recommends your business.

Answer Checkup is not affiliated with OpenAI, Perplexity or Google.
Contact: [email protected]

Sources

All pages fetched and checked on October 15, 2026.