Free llms.txt Generator

What is llms.txt?

It works the same way robots.txt does, but for a different audience. robots.txt talks to search-engine crawlers. llms.txt talks to large language models — ChatGPT, Claude, Perplexity, Gemini, and the crawlers behind Google AI Overviews and Bing Copilot.

When a person asks ChatGPT a question and ChatGPT needs to consult your site to answer, it can read a short llms.txt instead of crawling your whole sitemap. That gives you control over what gets quoted and how you get described.

The file format was proposed in September 2024 by Jeremy Howard at Answer.AI. There is no W3C-style standard yet, but adoption has grown quickly: Anthropic, Vercel, Supabase, and most modern docs sites publish one. Reading is opt-in for each AI vendor, and the major crawlers (OpenAI, Anthropic, Perplexity) have all been observed fetching /llms.txt in the wild.

How to create an llms.txt file

Think of it as a one-page summary of your website, written so AI chatbots can read it in a second. Three parts. Only the first two are required, and most files look about the same.

llms.txt examples

Two complete files you can copy and adapt for your own site. Both are about 15 lines. Most sites do not need more than this.

# Acme CRM

> Acme CRM helps small sales teams track leads, log calls, and
> close more deals without the overhead of enterprise tools.

## Pages

- [Home](https://acme.com): Product overview and signup
- [Pricing](https://acme.com/pricing): Plans from $29 to $99/mo
- [How it works](https://acme.com/how-it-works): The 3-step sales loop
- [Integrations](https://acme.com/integrations): Gmail, Outlook, Slack, Zapier
- [Help docs](https://acme.com/docs): Setup guides and API reference

## Optional

- [Blog](https://acme.com/blog): Sales operations articles
- [Press](https://acme.com/press): Media coverage

# OpenLog

> A lightweight structured logger for Node.js with zero dependencies,
> built for serverless functions and edge runtimes.

## Pages

- [Home](https://openlog.dev): Overview and install
- [Quickstart](https://openlog.dev/docs/quickstart): 5-minute setup

## Docs

- [API reference](https://openlog.dev/docs/api): All exported functions
- [Transports](https://openlog.dev/docs/transports): Built-in destinations
- [Recipes](https://openlog.dev/docs/recipes): Common patterns
- [Migrating from Winston](https://openlog.dev/docs/migrating-from-winston)

## Optional

- [Changelog](https://openlog.dev/changelog)
- [Contributing](https://openlog.dev/contributing)

For real-world examples, look at the live files at anthropic.com/llms.txt, vercel.com/llms.txt, and supabase.com/llms.txt.

How to add llms.txt to your site

The goal is the same on every platform: serve your file at https://yoursite.com/llms.txt with content type text/plain. The path matters — AI crawlers only look at the apex.

const BODY = `# Your Site
> ...
`;

export function GET() {
  return new Response(BODY, {
    headers: { 'content-type': 'text/plain; charset=utf-8' },
  });
}
export const dynamic = 'force-static';

About the llms.txt standard

The llms.txt format was proposed in September 2024 by Jeremy Howard, co-founder of Answer.AI. The canonical spec lives at llmstxt.org. There is no formal W3C standardization yet, but the format is stable: a single H1, an optional blockquote summary, then H2 sections of Markdown link lists, with a final ## Optional for low-priority pages.

The longer companion file llms-full.txt includes the full contents of each linked page concatenated together. It is only worth shipping for docs-heavy sites where you want models to ingest the whole corpus in a single fetch.

Validating your file: there is no official validator yet. The practical check is to fetch your URL with curl, confirm it returns HTTP 200 and content-type: text/plain, and that the body parses as Markdown with at least one H1 and one link list. The generator above produces a valid file by construction.

Frequently asked questions

What is llms.txt and why do I need one?

It's a short text file that tells AI chatbots what your website is. When ChatGPT or Claude visits your site to help someone, they read this file first. It says what your site is, which pages matter most, and how to describe you. There's no official rulebook yet, but the big AI companies (OpenAI, Anthropic, Perplexity) have all been spotted reading these files. Making one takes about a minute.

Where do I put the file once I download it?

Right at the front of your website, at https://yoursite.com/llms.txt — the same way robots.txt sits. Most CMS platforms let you upload it through a "static files" or "public folder" setting. If you run your site in multiple languages, you can put one in each language folder.

Will AI chatbots actually read this?

Yes, and more than you'd guess. Yesterday alone, ChatGPT visited our tool and blog pages 28 times to read them while answering people. We can see this in our server logs. Whether your specific file changes one specific answer is hard to prove case by case, but the file is free to make and the big AI tools are reading them.

How is llms.txt different from robots.txt?

robots.txt tells crawlers what they can and can't access on your site, like a bouncer. llms.txt is more like a welcome note for AI chatbots: here's what we are and what you should read. The two don't conflict. You usually want both.

Do I need to update it as my site grows?

Update it when you launch a big new page, change what your business is about, or every few months as a habit. AI crawlers cache files for a while, so changing it every day is wasted effort. A short list of the right pages always beats a giant list of every page on your site.