What exactly is llms.txt?
llms.txt is a plain-text file in Markdown format that sits at the root of your website, at /llms.txt. It gives AI crawlers and agents (from ChatGPT, Claude, Perplexity and other large language models) a structured overview of your most important pages at a glance: who you are, what you do, and where your core pages live.
The spec comes from Answer.AI (Jeremy Howard) and was first proposed in 2024 at llmstxt.org. By August 2026, version 2 of the spec is widely adopted: alongside the compact /llms.txt, there's now also a fuller variant, /llms-full.txt, containing the full content in plain text for models that need more context before citing.
Why is it needed?
AI models crawl differently than Google. An LLM agent that wants to quickly gather context about your business doesn't have time to search your entire site, render every page and parse the navigation. llms.txt answers that directly: a short, readable summary without HTML noise, pop-ups or cookie banners.
For your business that means faster and cheaper inclusion in AI answers, with less risk that a model cites an outdated or incomplete page. It's not a Google ranking factor, but it is a practical tool for GEO (Generative Engine Optimization).
How do you implement an llms.txt?
Structure the file like a short table of contents: an H1 with your brand name, an H2 with your positioning in one sentence, and below that sections with markdown links to your most important pages — services, core pages, blog, contact. Add your NAP details (name, address, phone) so AI models recognise you as a concrete, local entity.
- Open a new text file and write an H1 with your business name.
- Add an H2 with your positioning in one sentence max.
- Put your NAP details (name, address, phone, registration number) in a short block.
- Group links by category: Services, Core pages, Blog, Contact.
- Publish the file at /llms.txt with content-type text/plain.
- Optionally publish a full content dump at /llms-full.txt.
Difference from sitemap.xml
sitemap.xml is for search engine crawlers like Googlebot: a machine-readable list of every URL with metadata (lastmod, priority) so Google knows what to crawl and when. llms.txt is for AI models: not a full list, but a manually curated, prioritised selection of the content most valuable to read — written to be understood by a language model rather than a crawler that walks every URL.
The two files complement each other. sitemap.xml gets you indexed in Google; llms.txt lets an AI model quickly and correctly cite who you are.
Example: our own llms.txt
At organiqgrow.nl/llms.txt you'll find our own file as a concrete example: company details, an overview of all our services with indicative pricing, every city page, industry page and the full blog list by category. Feel free to use it as a reference for your own implementation.
Want to know if your site is ready for AI search engines?
Request a GEO scan →