SchemaValid
ENDE

llms.txt Generator

Enter your website URL and get a ready-to-use llms.txt file that follows the official llmstxt.org format. We crawl your homepage and its internal pages, then build the file for you — free, no account needed.

How the generator works

Everything runs on our server: we fetch your homepage, find the internal links on it, then read the title and description of each linked page to build the file. Nothing is stored and no account is needed.

  1. Enter your URL. Use your homepage — that's where the links to everything else live.
  2. We crawl up to ten pages. Your homepage plus the first ten internal pages we find, read in parallel.
  3. You get a ready file. Grouped into sections by URL path, ready to copy or download.

What is llms.txt and why would you want one?

An llms.txt is a plain-text file at the root of your domain that gives AI assistants a curated map of your site: what it is, and which pages matter. The idea comes from llmstxt.org and mirrors what robots.txt does for search crawlers — except robots.txt says what not to read, while llms.txt says what's worth reading. The honest caveat: it's a proposal, not a ratified standard, and no major provider has publicly committed to reading it. But the file takes minutes to publish and costs nothing, and if assistants do start reading it, you'd rather have curated your own summary than let a model infer one from your navigation menu.

The format

The file is Markdown with a fixed shape: a title line, an optional blockquote summarising the site, then sections of links with short descriptions. Our generator produces exactly this, grouping pages by their first URL path segment — everything under /blog/ lands in a Blog section, everything under /tools/ in a Tools section:

# Your Site Name

> A short description of what your site does.

## Tools

- [IBAN Validator](https://example.com/tools/iban-validator/): Check any IBAN for free.
- [FAQ Schema Generator](https://example.com/tools/faq-schema-generator/): Generate FAQPage JSON-LD.

## Blog

- [How IBANs are structured](https://example.com/blog/iban-structure/): The anatomy of an IBAN, explained.

Why only ten pages?

Two reasons, one practical and one about quality. Practically, each page is a separate network request and the whole job has to finish inside a serverless time limit — crawling a hundred pages would time out and hammer your server. But the better reason is that llms.txt was never meant to be a sitemap. A sitemap lists everything; an llms.txt curates. Ten pages gives you the structure and a working draft, and the editing you do afterwards is what makes the file actually useful.

An llms.txt is one signal among many that decide whether AI assistants can use your site. Run the free AI Search Readiness Check

Frequently asked questions

Do AI assistants actually read llms.txt?
Not reliably, and anyone telling you otherwise is overselling it. It's a proposal from llmstxt.org that has gained traction among developer-facing sites, but no major provider has publicly committed to reading it. The case for publishing one is that it costs you fifteen minutes and nothing else — a cheap bet on a standard that may or may not land.
Where exactly does the file go?
The root of your domain, so it loads at yourdomain.com/llms.txt — the same place as robots.txt. Not in a subfolder, not on a subdomain. If your site is on Next.js, Astro or a similar framework, that means your public/ or static/ directory.
Should I use the generated file as-is?
No. Treat it as a first draft. We read whatever titles and meta descriptions your pages already have, which means the file inherits any vagueness in them. Cut pages that don't earn their place, rewrite descriptions so they say what the page actually does, and reorder sections so the important ones come first. The curation is the value.
Why do some of my pages not appear?
Three common reasons: we only follow links from your homepage, so anything reachable solely from a submenu or another deep page won't be found; we cap the crawl at ten pages; and we skip pages that fail to load or have no title tag, because an untitled entry would be useless in the file. Add missing pages by hand.
How are the sections decided?
By the first segment of each URL path. Everything under /blog/ becomes a Blog section, everything under /tools/ a Tools section, and pages sitting at the root go under Main. It's a mechanical rule that gets you a sensible starting structure — rename the sections to whatever suits your site.
Is llms.txt the same as robots.txt?
No, and they solve opposite problems. robots.txt is a permission file: it tells crawlers what they may not fetch, and it's an established standard every crawler respects. llms.txt is a content file: it tells assistants what's worth reading and summarises it for them. You want both — they don't overlap.
Does having an llms.txt help my Google rankings?
No. Google has never said it reads the file, and there's no evidence it affects classic rankings. Don't publish one expecting a ranking bump — publish it because it might help AI assistants describe your site accurately. Anyone selling llms.txt as an SEO ranking factor is making that up.
How often should I update it?
Whenever your site structure changes meaningfully — a new section, a product line, a batch of articles worth surfacing. There's no freshness signal to chase here, so there's no benefit in touching it for its own sake. Re-running this generator after a redesign is a quick way to catch what's new.
Do you store my URL or the generated file?
No. The crawl runs live and the result exists only in your browser until you copy or download it. Nothing is logged, no database, no account. Refresh the page and it's gone.
Can I generate this for a site I don't own?
Technically yes — the tool fetches public pages, and it can't tell who owns what. But the file only helps if you can publish it to that domain's root, so generating one for someone else's site is a curiosity rather than a use case. We only fetch publicly reachable pages and nothing behind a login.
Why is my file mostly empty?
Usually because your homepage has few internal links, or your pages render their content client-side so the raw HTML we fetch has no titles or descriptions in it. That second case is worth taking seriously: if we can't see your content in the HTML, neither can most AI crawlers.
What's the difference between llms.txt and llms-full.txt?
llms.txt is the curated index — links and one-line descriptions. llms-full.txt is a proposed companion that inlines the full text of your documentation into one file, so an assistant can read everything without following links. This tool generates the index version. The full variant only really makes sense for documentation sites, and it gets large fast.

Related tools

AI Search Readiness Checker

Score how ready your site is for ChatGPT, Perplexity and AI Overviews.

FAQ Schema Generator

Generate FAQPage JSON-LD for AI search and structured Q&A data.

Video Schema Generator

Generate VideoObject JSON-LD for video rich results and AI search.

Schema Validator

Validate a live URL or JSON-LD code: SEO score, errors, warnings and rich-results readiness.

IBAN Validator

Check whether an IBAN is valid, instantly and for free.

Article & Blog Posting Schema Generator

Generate Article, Blog Posting or NewsArticle JSON-LD for your posts.