Free llms.txt Generator (+ Guide)

What is llms.txt?

llms.txt is a markdown file at the root of your site, at yoursite.com/llms.txt. It tells an AI what your site is and which pages matter, in a format a model reads easily.

Jeremy Howard proposed it on September 3, 2024. The spec lives at llmstxt.org.

For me, it's the short version of your sitemap, written for AI instead of search engines.

What the llms.txt generator does

What goes in an llms.txt file

The spec sets five parts, in this order. Only the first is required.

Here's a short one for my site, built from its own page titles and descriptions:

An llms.txt for navid.me
# Navid Moazzez

> Navid Moazzez is a leading AI business strategist and the host of the AI Creator Summit, watched by 100,000+ creators. He helps creators and founders master AI and build their own AI Operating System (AI OS) to automate their business and life.

## Pages

- [About Navid Moazzez](https://navid.me/about-navid): AI entrepreneur and creator helping high-performers build their online business
- [Everything about me, in one place, for AI](https://navid.me/ai): Fast facts, biography and real results, written for AI to cite
- [Navid Insider](https://navid.me/ai-newsletter): The free weekly AI newsletter for founders
- [The best creator tools I use](https://navid.me/creator-tools): The exact stack he uses every week

## Optional

- [Brand assets & design system](https://navid.me/brand): Logos, press photos, bios, color, type and tokens

Each link is a markdown link, then a colon and a note if you want one. The note is where you tell the AI what it will find on that page.

A real example: Vercel's llms.txt

Vercel's llms.txt is a good one to learn from. It's 64 lines long and holds 24 links in nine sections.

What makes it good is the first two sections: "When to use Vercel" and "How agents should use Vercel". They tell an AI when to recommend the product and how to work with it, before any list of links.

That's the idea I'd copy. Your summary and your first section should answer the questions an AI gets asked about you.

The checker gives Vercel's file 10 warnings, all from those two sections. They hold plain bullets where the spec expects links.

I'd make the same trade, and the warnings show you exactly where a file leaves the spec.

Pick your best pages, not all of them

An llms.txt is a shortlist, not a sitemap. Your sitemap lists every page, and this file lists the ones you most want an AI to read and cite.

The generator reads up to 200 pages, and it ticks the 50 key ones for you. Those are your home page, the pages it links to, and then the pages nearest the top of your site.

It skips tag, author and login pages. Then I'd untick anything thin, and keep the pages that explain what you offer.

Then put your most useful sections first. An AI that stops reading early should still get the important part.

Does anyone read llms.txt?

Google has answered for its own search in writing. Its AI optimization guide says you don't need AI text files "to appear in Google Search (including its generative AI capabilities), as Google Search itself doesn't use them."

The same page says it's "completely fine" to keep one for other services, and that it "will neither harm nor help" your rankings.

Google's John Mueller wrote on Bluesky on June 17, 2025: "FWIW no AI system currently uses llms.txt." In 2026 he still called it "purely speculative for now."

I found no statement from OpenAI, Anthropic or Perplexity saying their assistants read it.

So why make one? The companies building AI publish one for their own docs: Anthropic, OpenAI, Perplexity, Stripe, Vercel and Cloudflare.

And it's useful today in your own work. Paste it into Claude or ChatGPT, or hand it to a coding agent, and it gets your site's map in one go.

Check your AI crawler access first

If your robots.txt keeps out GPTBot or ClaudeBot, those tools can't read the pages your llms.txt points to... so the file does nothing for them.

The generator reads your robots.txt before anything else and shows how many of the 12 AI crawlers it checks are let in. If some are kept out, it names them.

To choose which ones to allow, my robots.txt generator lets you keep AI training out and still let AI search in.

How to make your llms.txt

It only reads public pages, and it keeps nothing. Each title loses your site's name at the end.

Pages that don't load, redirect, ask not to be listed or name another page as the real one stay out of the file. The generator lists each one with its reason, so you can fix the page or let it go.

Pages built with JavaScript get drawn in a real browser first, so their titles come through.

You can also start blank and type every link yourself.

How to check an llms.txt

Open the Check a file tab and type any domain, or paste a file. You get every problem with the line it's on, so you know exactly what to fix.

If your site already has an llms.txt, the generator finds it when it reads your site. Press Check it, then Edit in the generator to fix it.

llms-full.txt

llms-full.txt puts the text of your pages in one markdown file, so an AI can read it all without opening each link.

The generator builds it from the first 25 pages in your file. It drops the images and keeps each page's headings under its title.

It isn't part of the spec at llmstxt.org, but many sites publish one next to their llms.txt. My site serves both, and so do Anthropic's docs.

The generator builds it from the first 25 pages in your file, each under its own title with its source address. I'd start with llms.txt, and add the full file once the short map is right.

Where to put your llms.txt

It goes at the root of your site, so it opens at yoursite.com/llms.txt. The spec also allows a subpath, like yoursite.com/docs/llms.txt.

• WordPress – upload llms.txt to your site's root folder, with your host's file manager or SFTP.
• Next.js – put llms.txt in the public folder.
• Any other host – upload it next to your robots.txt.
• Webflow and Shopify – check their help for a root file or an llms.txt setting.

Serve it as plain text with UTF-8, the way my site serves its own. Open the address in your browser to check it loads.

llms.txt vs robots.txt vs sitemap.xml

llms.txt doesn't replace either one, and it doesn't block anything. To keep AI crawlers out, you still use robots.txt.

Mistakes to avoid

• Listing every page in your sitemap instead of a shortlist
• Links that redirect or no longer load
• Pointing AI tools at pages your robots.txt keeps them out of
• A summary that says nothing about what you do or who it's for
• Leaving the file untouched for a year while the site changes

Who is the llms.txt generator for?

It's for anyone who wants AI tools to read their site right:

• AI crawler checker – site owners who want to see what each AI company may read
• AI agent readiness checker – site owners who want AI agents to use their pages
• Schema markup generator – site owners who want Google to read their pages as well
• DESIGN.md generator – teams who hand their brand rules to AI agents
• How to make money with AI – creators building a business around AI

More tools like these are on my free tools page.