Paste a domain or a list of URLs. Get a spec-compliant llms.txt you can copy or download and drop into your site root. Free, no signup, runs in your browser.
https:// needed. A sitemap is optional but improves grouping.Upload the file so it answers at yourdomain.com/llms.txt.
Update it when you ship a page worth citing. A stale index is worse than none.
Open the URL in a browser. If you see Markdown, the file is live.
An llms.txt is a Markdown file at the root of your site that tells language models what your site is, what it contains, and which pages matter most. The format was proposed at llmstxt.org and follows the same idea as robots.txt — a small, predictable file at a well-known path.
robots.txt tells crawlers what not to fetch. llms.txt tells models what is here, in a shape they can read directly, instead of making them infer your structure from raw HTML.
# Example Corp > A one-sentence description of what this site is. Context for the model: the site covers X, is authoritative on Y, and these pages are the best starting points. ## Docs - [Getting started](https://example.com/start): Install and run your first build - [API reference](https://example.com/api): Complete endpoint documentation ## Optional - [Full page index](https://example.com/full-index.md): Every URL on the site
Three structural rules from the spec: the H1 is your site name, a blockquote line carries the summary, and H2 sections group links in the form - [Title](URL): note.
Wherever you host, the requirement is the same: the file must be reachable at the root path.
| Platform | Where it goes |
|---|---|
| Static site / Vercel / Netlify | Put llms.txt in the folder that is published as the site root — usually public/ or static/. |
| WordPress | Upload to the domain root via SFTP or your host's file manager, alongside wp-config.php. Some SEO plugins also expose a root-file editor. |
| Shopify | Not writable through the theme editor. Use an app or a redirect/proxy that serves the file at /llms.txt. |
| Webflow | Add it as a static file in the site's root directory, or serve it from a redirect rule. |
| Next.js (App Router) | Put llms.txt in app/llms.txt/route.ts and return the text, or place it in public/. |
| Docs generators (Docusaurus, Mintlify, MkDocs) | Drop it in the static assets directory; most ship it automatically once present. |
Then confirm it: open yourdomain.com/llms.txt in a browser. If you see Markdown, it is live.
| File | Written for | Answers |
|---|---|---|
| robots.txt | Crawlers | Which paths may be fetched |
| sitemap.xml | Search engines | Which URLs exist and when they changed |
| llms.txt | Language models | What this site is and which pages are worth reading |
They are complementary, not alternatives. Keep all three.
A minimal docs site:
# Acme Docs > API documentation for the Acme platform. ## Docs - [Authentication](https://acme.dev/docs/auth): API keys and OAuth - [Rate limits](https://acme.dev/docs/limits): Quotas and headers ## Optional - [OpenAPI schema](https://acme.dev/openapi.json): Machine-readable API spec
An e-commerce store with an optional index:
# Northwind Supply > Industrial fasteners and tooling, shipped from Rotterdam. Context for the model: this is a product catalogue. Prices are in EUR and exclude VAT. Stock is shown per variant. ## Catalogue - [Bearings](https://northwind.example/bearings): All bearing types and sizes - [Fasteners](https://northwind.example/fasteners): Bolts, nuts, washers - [Tooling](https://northwind.example/tooling): Hand and power tools ## Support - [Shipping & returns](https://northwind.example/shipping): Delivery times and return policy ## Optional - [Full product index](https://northwind.example/index.md): Every SKU with prices
The honest answer is that it is unproven. Some crawler logs show model agents do fetch the file; other analyses argue the effect on citations is small or unmeasured. Treat it as a cheap, low-risk addition to your site — not a guaranteed traffic source.
No. It is a community proposal hosted at llmstxt.org, adopted by a number of documentation platforms. There is no governing body, and support varies by crawler.
Yes. They serve different readers: sitemaps list URLs for search engines, robots.txt controls fetching, and llms.txt summarises the site for models. Keep all three.
Keep the main file short — a summary plus your most important links. If you have hundreds of pages, link out to a longer index file from an ## Optional section instead of inlining everything.
The generator works from any public URL list. Use it to draft a file for a client site, or to inspect how a competitor's site would be summarised.
No. This page runs entirely in your browser. Nothing you paste is uploaded, stored, or logged.