llms.txt
llms.txt is a proposed text file at the root of your site, like yoursite.com/llms.txt, that lists your most important pages in plain Markdown for AI tools. A SaaS company might list its product overview, pricing, docs and key guides, as a quick extra after crawler access and good pages.
What is llms.txt?
llms.txt is a plain text file in Markdown that sits at the root of a website, like yoursite.com/llms.txt, and lists the pages an AI tool should read to understand the site. Jeremy Howard of Answer.AI proposed the format in September 2024, and the spec lives at llmstxt.org.
The format is simple. The only required part is an H1 with your site or product name. After that come a short summary in a blockquote, a few lines of context and H2 sections with lists of links, each with a short note. A section called "Optional" holds links an AI tool can skip when it's short on space. The spec also suggests offering a clean Markdown copy of each important page.
Let's say you sell an API for sending invoices. Your llms.txt might open with "# InvoiceAPI", then one line on what the product does, then sections for "Docs" (quickstart, authentication, webhooks), "Product" (pricing, security) and "Optional" (changelog, blog). A coding agent that's asked to build an integration can read that one file and go straight to the right pages.
Why llms.txt matters for B2B SaaS
The honest answer is that it matters less than the hype suggests, and more for some companies than others.
Adoption is still small. The HTTP Archive's 2025 Web Almanac found llms.txt on 2.1% of websites. And in Ahrefs' study of 137,000 sites, 97% of llms.txt files got no requests at all in May 2026.
Google has been clear about its side. Its guidance on AI features says you don't need llms.txt or other special files to appear in AI Overviews or AI Mode. In 2026, Chrome's Lighthouse tool added an llms.txt check, but under a separate category for browser agents. It isn't part of the SEO score.
Here's where I think it still earns its place:
- Developer docs: Coding agents like Claude Code and Cursor read docs directly while they write code. A clean llms.txt with Markdown pages helps them find the right page and quote it correctly.
- API products: If developers build on your product, an agent that reads your auth and webhook docs well writes better integration code, and that makes your product easier to adopt.
- A cheap extra: Writing one takes an hour or two. As long as your crawler access and key pages are already in good shape, there's little downside.
How to create an llms.txt file
Treat llms.txt as a short, curated reading list. Here's my process:
- Fix crawler access first: Check that your robots.txt and CDN let in the AI search bots, like OAI-SearchBot and Claude-SearchBot. A file pointing to pages the bots can't fetch does nothing.
- Pick 15 to 40 pages: Choose the pages that explain what you do: product overview, pricing, security, integrations, the docs quickstart and your main guides. Leave out blog archives, tag pages and anything behind a login.
- Write the summary in plain words: The blockquote should say in one or two sentences what the product does and who it's for, the same way you'd say it to a buyer.
- Add a note to each link: After each link, write a short description, like "Pricing: plans, seat limits and what's in each tier". It helps an AI tool pick the right page without opening all of them.
- Offer Markdown copies of key pages: If your docs platform can serve a .md version of each page, link to those. They're easier for AI tools to read than pages full of menus and scripts.
- Put it on a schedule: Review the file whenever you launch a product or change pricing, so it never points to retired pages.
llms.txt examples
Some of the clearest examples come from developer-focused companies:
- Stripe: The llms.txt at docs.stripe.com opens with instructions for AI coding tools, like checking the package registry for the latest SDK version, before it lists the docs.
- Cloudflare: Its developer docs have a top-level llms.txt that links to a separate llms.txt for each product, from Workers to R2, so an agent loads only the product it needs.
- Vercel: The file at vercel.com/llms.txt opens with a one-line summary in a blockquote, then a short "When to use Vercel" section before the links.
Notice that all three are written for agents doing a task, mostly with code. That's where llms.txt does its best work today. Our post on B2B tech SEO covers how docs, Markdown copies and llms.txt fit into SEO for technical buyers.
How to measure llms.txt
Most of the value of llms.txt is hard to see in analytics, so measure what you can directly. Here's what I'd check:
- Requests to the file: In your server logs or CDN analytics, count requests to /llms.txt and note the user agents. If nothing requests it after a few months, keep it but don't spend more time on it.
- Requests to your Markdown pages: Track visits to the .md copies of your docs by user agent. Those show which pages agents actually read.
- Answers from coding agents: Ask Claude Code, Cursor or ChatGPT to build a small integration with your API. Check whether they use current endpoints and the right SDK version.
- AI answers about your product: Keep running your buyer prompts through prompt tracking. If llms.txt helps at all, you'd see it in more accurate answers about your product.
Common llms.txt mistakes
- Expecting a ranking boost: Google doesn't use llms.txt for rankings or for AI Overviews. Judge it by how well agents read your docs.
- Doing it before the basics: A blocked crawler, a JavaScript-only pricing page or a thin product page all matter far more. Fix those first, and our GEO checklist lists them in order.
- Dumping your sitemap into it: A file with 3,000 links isn't a reading list. Keep it short enough that a person could read it in two minutes.
- Using it to block AI tools: llms.txt has no allow or block rules. If you want to keep a crawler out, that belongs in robots.txt.
- Letting it go stale: A file that still links to last year's pricing page sends agents to the wrong facts. Update it with every launch.
So, does your product have public docs that developers build with? If it does, an llms.txt file is a good use of one afternoon this month.
Read more
FAQ
Does llms.txt help SEO?
No. Google says you don't need it to appear in AI Overviews or AI Mode, and it has no effect on regular rankings. Its main use today is helping AI coding tools and agents read developer docs.
Where do you put an llms.txt file?
At the root of your domain, so it loads at yoursite.com/llms.txt, the same place robots.txt lives. Large docs sites sometimes add one per product section too, like Cloudflare does.
What's the difference between llms.txt and robots.txt?
robots.txt tells crawlers which pages they may or may not visit. llms.txt doesn't block or allow anything. It's a reading list that points AI tools to the pages you think matter most.
Should every B2B SaaS company add llms.txt?
It's worth an hour if you have public developer docs or an API, since coding agents read those directly. For a marketing site without docs, fix crawler access and your key pages first.


