Crawlsonar

llms.txt Generator

Build a draft llms.txt for your site in seconds — from your sitemap.xml and homepage metadata, grouped into sensible sections and ready to curate.

What llms.txt is

llms.txt is a simple Markdown file at the root of your site that offers AI assistants a curated map of your most important pages — the way a table of contents guides a reader. Where a sitemap dumps every URL, llms.txt is hand-picked: an H1 with your site name, a one-line summary, then a few sections of links each with a short description. When an LLM tool fetches it, it spends its limited context budget on the pages you chose rather than crawling blindly. This generator drafts one from your existing site so you only have to curate, not write from scratch.

How it works

  1. Reads your homepage <title>, og:site_name and meta description for the H1 and summary line.
  2. Fetches sitemap.xml (sitemap indexes supported); if none exists, it falls back to your homepage navigation links.
  3. Groups URLs into sections — Documentation, API, Products, Pricing, Blog, Company, Legal — and caps the list so the file stays curated, not dumped.

Honest note: Google Search ignores llms.txt — it will not improve your Google rankings. Publish it for broader LLM tools, AI assistants and documentation consumers.

Editing tips before you publish

  • Add a one-line description after a colon for each important link: - [Pricing](https://…): plans, limits and billing FAQ.
  • Delete pages that only make sense to humans mid-journey (cart, login, search).
  • Keep it under ~50 links — an assistant’s context budget is the scarce resource.
  • Re-generate (or at least re-check) after site restructures so links don’t rot.

FAQ

Where do I put the file?

At the root of your host — https://example.com/llms.txt — served as text/plain. On Next.js put it in public/; on WordPress upload it to the web root or use your SEO plugin if it supports llms.txt.

Why is the output capped at ~40 links?

llms.txt is a curation format. Hundreds of links defeat its purpose — tools that want everything can read your sitemap; llms.txt is for what matters most.

My site has no sitemap — will this still work?

Yes, the generator falls back to your homepage navigation links. But you should fix the missing sitemap too — it hurts your ordinary SEO far more than any llms.txt question.

Related tools

  • llms.txt Checker — validate the file you just published (structure, content type, link health).

Last reviewed: 2026-07-09. Output follows the llmstxt.org proposal current at that date.

Frequently asked questions

How is the draft generated?

From your sitemap and homepage metadata — it lists your key pages with short descriptions in the llms.txt Markdown format.

Do I just publish the output as-is?

Review and curate it first, then serve it at /llms.txt with a text/plain or text/markdown content type.