What Is llms.txt? A Practical Guide for Small Websites

llms.txt is a simple Markdown file that gives AI systems a curated map of the most useful pages on your website. It usually lives at https://yourdomain.com/llms.txt and points agents toward canonical pages, docs, product pages, contact paths, and other resources you want them to understand first.

For a small website, the value is not magic rankings. The value is clarity. If ChatGPT, Perplexity, Gemini, Claude, or another answer engine tries to understand your site, llms.txt gives it a concise starting point instead of forcing it to infer everything from navigation, templates, JavaScript, and scattered pages.

The original proposal from Jeremy Howard describes /llms.txt as a way to provide information that helps LLMs use a website at inference time. It is still an emerging convention, not a formal search engine ranking standard. That distinction matters: use it as a helpful technical signal, not as a replacement for crawlable pages, useful content, schema, or a sitemap.

What should llms.txt contain?

A useful llms.txt file should be short, readable, and selective. For most small websites, it should include:

  • the site or brand name
  • a short summary of what the site does
  • links to priority pages
  • links to docs, guides, product pages, or service pages
  • a contact or support path
  • brief guidance on how agents should treat the listed resources

Think of it as a curated table of contents for AI systems. It should not be a dump of every URL on your site. That job belongs to your XML sitemap.

A simple llms.txt example

# Example Company

> Example Company sells maintenance software for small manufacturing teams.

Website: https://example.com
Contact: https://example.com/contact

## Priority pages
- [Homepage](https://example.com): Main brand and product overview
- [Product](https://example.com/product): Maintenance management software details
- [Pricing](https://example.com/pricing): Plans and buying information
- [Implementation guide](https://example.com/blog/implementation-guide): Setup guidance for new customers

## Guidance for AI agents
- Prefer canonical pages listed above when summarizing this site.
- Cite the source URL when using product or pricing details.
- Do not treat this file as permission to bypass robots.txt or page-level access rules.

How is llms.txt different from robots.txt and sitemap.xml?

robots.txt is mainly about crawler access guidance. It tells crawlers which paths are allowed or disallowed. sitemap.xml is mainly about discovery. It lists URLs that search engines can crawl and index.

llms.txt has a different job: it explains which resources are most useful for understanding the site. It can complement robots.txt and sitemap.xml, but it does not replace either one.

FileMain purposeUse it for
robots.txtCrawler access guidanceAllowing or disallowing crawler paths
sitemap.xmlURL discoveryListing canonical pages for search engines
llms.txtAI context guidancePointing AI systems to priority resources

How to create llms.txt for a small website

  1. Choose 5 to 15 priority URLs. Start with your homepage, core product or service pages, best guides, pricing page, documentation, and contact page.
  2. Write a one-paragraph summary of what the site does and who it serves.
  3. Group links under clear headings such as Priority pages, Docs and guides, Products, or Support.
  4. Add short descriptions after important links so the file is useful without requiring guesswork.
  5. Upload the file to your domain root as /llms.txt.
  6. Open the URL in your browser to confirm it is publicly accessible.

If you want a faster starting point, use the free llms.txt Generator. It creates a clean starter file from your brand summary, priority pages, docs URL, and contact path.

What not to put in llms.txt

Because llms.txt is public, keep it clean and non-sensitive. Do not include:

  • API keys or passwords
  • private customer documents
  • unpublished pricing or roadmap details
  • internal admin URLs
  • claims that are not supported by public pages

How to test your llms.txt file

After publishing the file, check three things:

  • Does https://yourdomain.com/llms.txt return a normal 200 response?
  • Are the listed URLs live, canonical, and useful?
  • Do robots.txt, sitemap.xml, and homepage metadata support the same story?

You can run a quick technical check with the free AI Crawler Access Checker. It checks whether robots.txt, llms.txt, sitemap.xml, and homepage metadata are reachable. For a deeper content and technical audit, use GEO Optimizer Pro.

Does llms.txt improve SEO?

There is no reliable evidence that llms.txt directly improves traditional Google rankings. Treat it as a GEO and AI visibility support file, not as a ranking shortcut.

The practical benefit is that it forces you to define the most important pages and explain them clearly. That helps your own content strategy, and it may help AI systems orient themselves if they choose to read the file.

Recommended workflow

  1. Use the GEO Starter Kit to choose the page and signals you want to improve.
  2. Create your file with the llms.txt Generator.
  3. Publish it at the root of your domain.
  4. Run the AI Crawler Access Checker.
  5. Use GEO Optimizer Pro when you need a full content and technical audit.

FAQ

Is llms.txt a formal web standard?

No. It is an emerging convention and proposal. It is useful to test, but it should not be treated as a guaranteed ranking or AI citation mechanism.

Where should I place llms.txt?

Place it at the root of your domain, for example https://example.com/llms.txt.

Should every page be listed?

No. Keep it curated. Use your XML sitemap for broad URL discovery and use llms.txt for the pages that best explain your site.

Can llms.txt replace schema markup?

No. Schema markup, crawlable HTML, clear content, internal links, robots.txt, and sitemap.xml still matter. llms.txt is an additional orientation layer.

Reference: The /llms.txt file proposal.