Your whole site, in one file an AI can read.
Paste your URL. We read your real pages and assemble a valid llms.txt — then you edit every line of it. Nothing is written by a model, so no page in your file can be one that does not exist.
# Your site
Save it as llms.txt at the root of your site, so it answers at https://yourdomain.com/llms.txt. It must be served as plain text.
- No blockquote summary. One sentence saying what the site is, as `> …`, is the part most likely to be quoted back.
- No links. A file with no pages in it tells an engine very little.
- No engine has publicly committed to reading llms.txt, and adoption is early. It costs nothing to add and may help; treat anyone promising otherwise with suspicion.
A sitemap lists everything. This one leaves things out.
That is the whole idea. A crawler reading your site wades through navigation, cookie banners, footers and three hundred blog tags before it reaches anything that says what you do. llms.txt is one short Markdown file, written by you, that says it directly — and links only to the pages actually worth reading.
What a crawler may not fetch. A set of prohibitions.
Every URL you have, so nothing is missed. Exhaustive by design.
The few pages that matter, described in your words. Curated by design.
Three different jobs. You want all three — none replaces another.
# Acme
> Acme turns messy spreadsheets into dashboards your finance team will actually open.
Built for finance teams at companies of 50–500 people. Connects to Xero, QuickBooks
and NetSuite, and ships 40 pre-built reports so nobody starts from a blank canvas.
## Product
- [Features](https://acme.com/features): what the product does, module by module
- [Integrations](https://acme.com/integrations): the accounting systems we connect to
- [Pricing](https://acme.com/pricing): three plans, published prices
## Documentation
- [Getting started](https://acme.com/docs/start): first dashboard in ten minutes
- [API reference](https://acme.com/docs/api): REST endpoints and auth
## Optional
- [Terms](https://acme.com/terms)
- [Privacy](https://acme.com/privacy)Nobody can prove it does. Add it anyway.
You will find pages telling you llms.txt improves your AI visibility. None of them can show you the evidence, because there isn't any yet — and we would rather you heard that from the people giving you the tool than find out later.
What is true
The convention exists, it is documented at llmstxt.org, and a growing number of sites publish one. Anthropic, Perplexity and several developer-tool companies have files live today.
What is not established
No major AI provider has publicly committed to reading it, and there is no published study showing it changes how a brand is cited. Anyone quoting you a percentage is guessing.
Why to add it regardless
It takes ten minutes, costs nothing, and cannot hurt you. If the convention is adopted you are already there; if it is not, you have lost an afternoon and gained a clear summary of your own site.
This is also what our page audit says about it. The llms.txt check is worth 3 of 100 and its fix text reads: “cheap to add, don't expect much from it alone.” We are not going to sell it harder on a page than we score it in the product.
Root of the domain, served as plain text.
Static sites & Vercel / Netlify
Drop llms.txt into your public/ (or static/) folder and deploy. It is served at the root automatically.
Next.js
public/llms.txt works. For a generated one, a route handler at app/llms.txt/route.ts returning text/plain also works — that is how this site does it.
WordPress
Upload it to the web root over FTP or your host's file manager. Do not create it as a Page — that returns HTML, not plain text.
Anything else
Wherever robots.txt lives, llms.txt goes beside it. If you can reach /robots.txt you can reach /llms.txt the same way.
Then check it. Open yourdomain.com/llms.txt in a browser — you should see plain text, not a styled page and not a 404. Our free page checker tests for it as one of its 31 checks.
Questions about llms.txt
What is llms.txt?+
A single Markdown file at the root of your site — yoursite.com/llms.txt — that describes what the site is and lists its most useful pages. The idea, proposed at llmstxt.org, is that an AI system reading one short curated file understands you better than one crawling a hundred pages of navigation and cookie banners.
Does llms.txt actually do anything yet?+
Honestly: nobody can prove it does. No major AI provider has publicly committed to reading llms.txt, and there is no published evidence of it changing how a brand is cited. It costs nothing to add and it cannot hurt you, which is why it is worth doing — but treat anyone promising it will improve your AI visibility with suspicion. That is a claim none of us can currently support.
How is it different from robots.txt or a sitemap?+
robots.txt says what a crawler may not fetch. A sitemap lists every URL so a crawler misses nothing. llms.txt does the opposite of a sitemap: it is a short, curated, human-written summary of what matters, with the noise left out. Three different jobs — you want all three, and none replaces another.
Is this generator writing the file with AI?+
No. It reads your site's title, description and real internal links and assembles the file from those. It cannot invent a page, because there is no step in which anything writes one. The summary line is yours to rewrite — that is the part a machine cannot do well.
Where exactly do I put the file?+
At the root of your domain, served as plain text, so a request to https://yourdomain.com/llms.txt returns it. On most static hosts that means dropping it in your public folder. On WordPress it is usually easiest to add via a plugin or by uploading to the web root — not as a page, which would return HTML.
How long should it be?+
Short. It is a map, not a copy of the site. If an engine has to read everything you have anyway, the file has gained you nothing. A few hundred words and your genuinely useful pages is the right shape; over about 20,000 characters you have written a second website.
What about llms-full.txt?+
A related convention where the file contains your full documentation as Markdown rather than links to it. It suits documentation sites specifically. This generator builds the standard llms.txt — the link-based map — which is what most sites want.
Will you keep my file updated?+
Not from this free tool — build it, copy it, and you are done. Inside the product, llms.txt presence is one of the 31 checks the site audit runs each week, so you find out if it disappears or goes stale.
llms.txt is worth 3 points out of 100.
That is genuinely how much our audit gives it. The other 97 are the checks that decide whether an AI can read, understand and quote your page at all — run them free on any URL.