llms.txt: The Complete Guide (2026)

What llms.txt is, how to create one, how it differs from robots.txt and sitemap.xml, whether AI engines actually use it, and where it fits in a GEO strategy. An honest, practical guide for site owners.

By Andrej Ruckij · · 5 min read

llms.txt: The Complete Guide (2026)

By Andrej Ruckij · June 16, 2026

TL;DR: llms.txt is a proposed standard — a curated markdown file at your site root that gives AI systems a high-signal map of your most important content. It’s quick to create and harmless to add, but it is a comprehension aid, not access control or a ranking lever. As of 2026, no major AI engine has confirmed it consumes the file, and Google says it doesn’t use it. Add it as a cheap supplement; don’t build strategy around it.

What you’ll learn

  • What llms.txt is and the problem it tries to solve
  • How to create one (with a template) and what to put in it
  • How it differs from robots.txt (access) and sitemap.xml (discovery)
  • Whether AI engines actually use it in 2026 — the honest answer
  • Where it fits (and doesn’t) in an SEO/GEO strategy

What llms.txt is

llms.txt is a plain markdown file you place at https://yoursite.com/llms.txt. It contains a curated map of your site — a title, a one-line summary, and grouped links to your most important pages, each with a short description. Proposed by Jeremy Howard (Answer.AI) in September 2024, the idea is to give large language models a concise, high-signal overview instead of forcing them to infer your site’s structure from raw HTML. For the one-line version, see glossary/llms-txt.

The key word is curated. Unlike a sitemap, it doesn’t list every URL — it lists the handful that matter, with context. That curation is the entire value proposition.

How to create one

The format is deliberately minimal: an H1 (your brand), a blockquote summary, optional context prose, and H2 sections of links. It takes about 20 minutes. The full step-by-step and a copy-paste template are in how-to-create-llms-txt. An optional companion file, llms-full.txt, holds the full expanded text of your key pages for reference-heavy sites.

The discipline that separates a useful file from a useless one is curation and description quality — see llms-txt-best-practices. The most common mistake is dumping your entire URL list in; that produces a worse-formatted sitemap, not an llms.txt.

How it differs from robots.txt and sitemap.xml

These three files get conflated constantly. They solve different problems:

  • robots.txt controls access — what crawlers may fetch. It’s advisory (compliant bots honor it; scrapers don’t). Full comparison: llms-txt-vs-robots-txt.
  • sitemap.xml drives discovery — an exhaustive list of every URL, so search engines find all your pages. Full comparison: llms-txt-vs-sitemap.
  • llms.txt aids comprehension — a curated map of your best pages for AI to understand quickly.

In a line: robots.txt is a permission sign, sitemap.xml is a complete index, and llms.txt is a guided tour. They stack; none replaces another.

Does it actually work?

This is the question that matters, and the honest 2026 answer is: unproven on the consumption side. Publishing llms.txt is widespread, but the AI engines that matter haven’t confirmed they read it — and Google has stated plainly that it doesn’t use it (John Mueller likened it to the old keywords meta tag). The full evidence review is in does-llms-txt-work.

That doesn’t make it worthless — it makes it a cheap bet on an unadopted standard. The cost is 20 minutes and the downside is zero, so adding it is reasonable. Expecting it to move your AI visibility is not.

Where it fits in SEO and GEO

For traditional SEO: no effect. Google ranks your actual pages; llms.txt isn’t a ranking signal.

For GEO (getting cited in AI answers): no proven effect either. What actually drives AI citations is earned media and brand mentions, topical authority, clean crawlable content, and not blocking the retrieval bots that cite you. llms.txt isn’t on that evidence-backed list. The full breakdown is in llms-txt-for-seo-geo, and the broader citation playbook is seo/ai-visibility.

So the correct priority order: do the real AI-visibility work first, add llms.txt last as a low-cost supplement.

Common questions

  • Q: Is llms.txt worth it? A: Yes, as a cheap, harmless addition — not as a visibility lever. See does-llms-txt-work.
  • Q: Does llms.txt help me rank on Google? A: No. Google doesn’t use it. See llms-txt-for-seo-geo.
  • Q: Is llms.txt the same as robots.txt? A: No — robots.txt is access control, llms.txt is comprehension. See llms-txt-vs-robots-txt.
  • Q: Should I just convert my sitemap to llms.txt? A: No — curation is the point; an exhaustive list defeats it. See llms-txt-vs-sitemap.

Key takeaways

  • llms.txt is a curated markdown map of your best pages, for AI comprehension — at /llms.txt.
  • It’s distinct from robots.txt (access) and sitemap.xml (discovery); all three are advisory except where a firewall enforces.
  • No major AI engine has confirmed it uses llms.txt in 2026; Google says it doesn’t.
  • It has no effect on Google SEO and no proven effect on AI citations.
  • Add it because it’s cheap and harmless — after the real AI-visibility work, never instead of it.

Sources