llms.txt is a proposed standard for a Markdown file, placed at /llms.txt on your site, that gives large language models a short summary of your site and links to its most important pages. You can build one in minutes with the QCK llms.txt Generator, but treat it as an optional extra: major AI companies have not formally committed to honoring it.
What llms.txt is
The llms.txt proposal was published at llmstxt.org in 2024. The idea is simple. Web pages are full of navigation, scripts and layout that make them hard for language models to read within a limited context window. A single, clean Markdown file can point a model to the content that matters, in a format it reads easily.
The proposed format has a defined order:
- An H1 with the name of the site or project (the only required section).
- A blockquote with a short summary.
- Optional paragraphs or lists with more detail.
- H2 sections containing lists of links, each written as a Markdown link with an optional note after a colon.
- An optional section titled "Optional" for secondary links that can be skipped when context is short.
A short example:
# Example Outdoor Co.> Direct-to-consumer hiking and camping gear, shipping across the US.## Products- [Hiking boots](https://example.com/collections/hiking-boots): Men's and women's, wide fits available- [Tents](https://example.com/collections/tents): 1 to 6 person## Policies- [Shipping and returns](https://example.com/pages/shipping)
The proposal also suggests offering Markdown versions of pages (for example, the page URL with .md appended) and a companion llms-full.txt that contains full content. Those parts are optional.
Does llms.txt help SEO or AI search?
Be realistic about this. llms.txt is a proposal, not a ratified standard. Major AI companies such as OpenAI, Google and Anthropic have not formally committed to reading or honoring it when they crawl the web, and Google has not announced any use of it in Search. Some documentation sites and developer tools use llms.txt files so that coding assistants can pull docs on request, which is where the format has seen the most adoption.
So llms.txt is not a ranking factor and not a substitute for AEO fundamentals. It is low-cost and low-risk, which is why many brands add one anyway: it takes little effort, it may help tools that do read it, and the process of writing it forces you to decide which pages best represent your business.
llms.txt is not robots.txt
The two files are often confused. Robots.txt is an established standard that tells crawlers what they may fetch. llms.txt does not grant or block access to anything. If you want to control which AI crawlers can reach your site, use robots.txt. See the robots.txt and AI crawlers guide.
How to use the QCK llms.txt generator
- Open the llms.txt Generator.
- Enter your site or brand name and a one or two sentence summary of what you do and who you serve.
- Add sections for your main content groups, such as Products, Services, Guides, Policies or Docs.
- Add the most important URLs to each section with a short, factual note describing each one.
- Move secondary links into the Optional section.
- Generate the file, review the Markdown, and download or copy it.
- Upload it to the root of your domain so it resolves at
https://yourdomain.com/llms.txt.
On Shopify, files cannot be placed at the root directly, so you will typically need an app, a proxy or a redirect setup to serve it at that path. Most other platforms and static hosts let you upload it directly.
How to write a good llms.txt file
- Be selective. Link to the pages that best answer what people ask about you: core product or service pages, pricing, key guides, policies and contact. A list of every URL is a sitemap, not an llms.txt.
- Write factual descriptions. Notes like "Sizing guide for all footwear" help a model choose the right link. Marketing slogans do not.
- Use canonical URLs. Link to the final, indexable version of each page, not redirects or parameter URLs.
- Keep it current. Update the file when you launch, rename or retire key pages.
- Match your site. The file should summarize what is on your pages, not make claims your pages do not support.
Common mistakes
- Expecting it to fix AI visibility on its own. Crawler access, clear answers and structured content matter far more. Start with the AEO checklist.
- Using it to block AI. It cannot block anything. Use robots.txt.
- Serving it with the wrong content type or behind a redirect chain. Serve it as plain text at the root, returning 200.
- Letting it go stale. Broken links in llms.txt undercut the point of the file.
If you want to be cited by AI assistants, llms.txt is a small piece of a larger job. Our AI search optimization team works on the parts that move visibility most: access, content structure and third-party presence.