llms.txt Generator & Validator

1. Enter your website. The generator reads your homepage, its links and your sitemap.xml, and groups your key pages into sections such as Docs, Product, Blog and Company.
2. Edit the draft. Rewrite the summary in your own words, add a short note after each link (": what this page covers"), and delete pages an AI doesn't need. The checker re-grades the file as you type.
3. Download and upload it. Save it as llms.txt and put it in the root of your site, so it loads at yoursite.com/llms.txt.
4. Validate the live file. Switch to Validate and enter your URL. We fetch /llms.txt the way an AI agent would and check it's served correctly.

llms.txt is a plain-text Markdown file at the root of a website that tells AI assistants and agents what the site is and which pages matter most. It was proposed in September 2024 at llmstxt.org as a lightweight, human-readable index made for language models, which struggle with navigation menus, scripts and ads on regular web pages.

Think of it as a curated table of contents. Where a sitemap lists every URL, llms.txt picks the handful of pages that explain your product (docs, pricing, guides) and says what each one covers, so an AI can find the right page quickly and describe you accurately.

1. An H1 title with your site or product name. The only required part.
2. A blockquote summary (a line starting with >): one or two sentences on what it is and who it's for.
3. Optional detail in normal paragraphs or lists, without headings.
4. H2 sections containing lists of links in the form - [Title](https://url): notes.
5. An Optional section for links an AI can skip when it's short on space.

# Acme

> Acme is invoicing software for freelancers. Send invoices, take card
> payments and track who owes you, from one dashboard.

Acme works in the browser and on iOS. Prices are in USD.

## Docs

- [Quick start](https://acme.com/docs/start): create an account and send your first invoice
- [Payments](https://acme.com/docs/payments): connect Stripe and accept cards
- [API reference](https://acme.com/docs/api): REST endpoints for invoices and clients

## Product

- [Pricing](https://acme.com/pricing): Free, Pro ($12/mo) and Team plans
- [Integrations](https://acme.com/integrations): Stripe, QuickBooks, Zapier

## Optional

- [Changelog](https://acme.com/changelog): every release since 2023
- [Privacy policy](https://acme.com/privacy)

Whatever you use, the file must load at exactly /llms.txt with status 200 and a text content type. A common mistake is a site that returns its normal web page (or a styled 404) for every unknown path: the validator flags that as "Served as an HTML page".

They work together. llms.txt doesn't grant or block access: if your robots.txt blocks AI crawlers like GPTBot or ClaudeBot, they won't read your pages however good your llms.txt is. Check both with our free SEO Chrome extension.

llms-full.txt is a companion file with the full text of your important pages in one Markdown document, so an AI can read everything without visiting each link. llms.txt stays a short index; llms-full.txt holds the content. If you have one, link it from the Optional section of llms.txt. The generator adds that link automatically when it finds your llms-full.txt.

Honestly, it's early. llms.txt is a proposal, not an adopted standard, and no major AI search engine has confirmed it uses the file for ranking or citations. Google has said it doesn't use llms.txt for Search. Where it clearly helps today is with AI agents and coding assistants that fetch a site's docs, and many developer documentation sites now publish one for that reason.

It takes ten minutes, can't hurt, and forces you to write a clear summary of your product, which is what AI assistants need to recommend you. To see whether ChatGPT and Perplexity actually mention your product, run the free AI visibility check.

• The file loads at /llms.txt with status 200, served as text rather than an HTML page.
• It starts with exactly one H1 title.
• It has a blockquote summary that actually says something.
• Pages are grouped under H2 sections as Markdown links, with full https:// URLs.
• Most links have a short description after the colon.
• No duplicate links, and a sensible size: a short index, not your whole site.

Each failed check costs points from 100, and every result comes with the fix. DodoLaunch's own llms.txt is a working example.