Knowledge base

What is llms.txt? your site, introduced to AI.

Robots.txt told crawlers where they may go; llms.txt tells language models what your site actually is. A small file with a growing role — this guide explains it in plain language.

Updated 1 August 2026

The short answer

llms.txt is a plain-text file in the root of your website (yourdomain.com/llms.txt) written for large language models. It introduces your site in machine-friendly Markdown: who you are, what you offer, and a curated list of your most important pages with one-line descriptions. The goal is guidance: when AI systems read your domain, the file points them to the pages that carry your knowledge — raising the odds that they represent you correctly and cite the right URLs. Creating one takes about an hour; the signal stands from that moment on.

Curious where your own website stands? The first findings appear on your screen right away.

Free website scan · twenty checkpoints · full report in your inbox

What goes into the file

The convention is simple Markdown: a first-level heading with your name, a short factual description, and sections of links — each link followed by a one-line summary of what that page answers. Strong files curate rather than dump: your services, your key knowledge articles, your contact route. Think of it as the front desk of your site, speaking machine.

Why it matters in 2026

Language models meet thousands of pages per domain and limited time to read them. A curated introduction shapes what they take away — the same way a good sitemap shaped crawlers. Adoption is young and growing across publishers, documentation sites and agencies; the cost is an hour, the downside is absent, and the upside compounds as more systems read it. Early signals reward early movers.

How to create yours

Write it by hand: name, one honest paragraph, then sections — Services, Knowledge, Contact — with your best URLs and one line each. Save as llms.txt in the web root, keep it current when pages change, and reference your true strengths first. Ovolum ships an llms.txt with every build and maintains it as part of AI visibility programmes; this site’s own file is live at ovolum.nl/llms.txt as a working example.

Frequently asked questions

Is llms.txt an official standard?

It is an open convention, proposed in 2024 and adopted organically since — comparable to how robots.txt and sitemaps spread. Systems that read it gain a cleaner picture of your site; systems that skip it simply fall back to crawling.

Does llms.txt replace robots.txt or my sitemap?

They serve different jobs: robots.txt sets access, the sitemap lists URLs, llms.txt explains meaning. A well-run site carries all three.

What belongs in a good llms.txt?

Curation: your identity in one paragraph, then your strongest pages with one-line descriptions each. Quality over quantity — the file is an introduction, and introductions are short.

Further reading

Free advice

Shall Ovolum take a look at your website?

Ask your question — you will receive a personal, written answer within one working day. Free of charge, with full attention.

Answer within one working day

Your site deserves a proper introduction.