llms.txt and structured data for AI search: the technical basics

llms.txt and structured data

What is llms.txt?

llms.txt is a new, proposed standard that lets website owners give AI search engines targeted information about their site. Much like robots.txt addresses classic search engine crawlers, llms.txt is aimed at large language models and their retrieval systems.

The file sits in the root directory of a website. It contains structured information on which content is most relevant, what expertise the provider has and how the content is best used.

How to create an llms.txt file, step by step

Step 1: Set up the file structure

Create a text file named llms.txt in the root directory of your website. The file uses a simple, readable format with Markdown-like syntax. Start with a title and a short description of your organization.

Step 2: Define your core pages

List the most important pages of your website and briefly describe what information each one contains. Focus on pages with strong subject-matter relevance, because AI systems prioritize content that shows clear expertise.

Step 3: State your expertise and context

Name the areas in which your company has particular expertise. This context helps AI systems classify your content as an authoritative source on specific topics.

Structured data for AI search

Alongside llms.txt, structured data (schema markup) is a decisive factor for AI visibility. AI search engines use Schema.org markup to understand the content of a page semantically.

These schema types matter most for AI systems:

  • Article / BlogPosting: helps AI systems recognize the author, publication date and topic of an article
  • FAQPage: question-and-answer pairs are used especially often by AI systems as a source for answers
  • Organization / Person: strengthens E-E-A-T signals and helps attribute content to a trustworthy source
  • HowTo: AI systems like to take over step-by-step instructions as a structured answer
  • DefinedTerm: definitions and glossary entries are ideal sources for AI-generated answers

Quick check: is your website AI-ready?

Use this checklist to see whether your website meets the technical basics for AI visibility:

  1. Is there an llms.txt file in the root directory?
  2. Do all important pages use Schema.org JSON-LD markup?
  3. Is your sitemap up to date and complete?
  4. Does your robots.txt accidentally block AI crawlers?
  5. Are author profiles marked up with Person schema?
  6. Do your FAQ sections have FAQPage schema?

If you answer “no” to more than two points, there is clear room for improvement. Our schema markup guide walks you through the implementation.

Technical AI optimization is an ongoing process, not a one-off project. The sooner you put the basics in place, the better positioned you are for the AI search of the future.

Need help with technical AI optimization? Our technical SEO team analyzes your website and implements everything that is needed. As an SEO & GEO agency, we make sure your website is set up well for Google and for AI search engines.

Also see which content formats AI systems process best and how to optimize multimodal content for AI search.

Baris Gündogdu

Baris Gündogdu

Founder & CEO of SEOX. More than 25 years of experience in SEO, GEO and digital marketing.

Baris Gündogdu, SEO & GEO expert →

Frequently asked questions

What is llms.txt, and how is it different from robots.txt?

robots.txt controls which pages search engine crawlers may visit. llms.txt is an additional file that gives AI systems context about your website, such as your areas of expertise and which pages matter most. The two files work together. In an SEO relaunch, robots.txt and llms.txt both belong on the checklist before go-live.

Does structured data help AI visibility?

Yes. AI search engines use Schema.org markup to understand the content of a page semantically. The most effective schema types for AI visibility are Article (with author and date), FAQPage (direct answers to common questions) and Organization/Person (E-E-A-T signals). Implement at least these three.

Do I have to maintain llms.txt manually?

Yes, at the moment llms.txt is a manually maintained file. Update it when the structure or focus of your website changes. Some CMS plugins already generate the file automatically from your content, though.

Ready for
technically optimized AI visibility?

We check your website's technical AI readiness and implement all the optimizations it needs.