Make Your Website AI-Friendly with Structured Data and llms.txt

What you’ll get

You will learn how to add structured data (like Schema.org JSON-LD) and create an llms.txt file for your website. This makes your content easier for AI search engines like ChatGPT, Gemini, and Perplexity to understand and use, improving how they present your information.

Tools you need

• ChatGPT (freemium): A powerful AI chatbot for generating text and code, useful here for creating structured data examples.
• Gemini (freemium): Google’s AI chatbot, similar to ChatGPT, for generating text and code.
• Perplexity (freemium): An AI-powered search engine that can help you understand how AI models process information.
• Google’s Rich Results Test (free): A tool from Google to check if your structured data is correctly formatted and can be understood by search engines.

Steps

1. Understand AI Search Engine Needs: AI search engines like ChatGPT and Gemini often fetch content on demand and have limited space (context window) to read it. Unlike traditional search engines, they don’t build a long-term index. This means your content needs to be clear and easy to find quickly. If the AI can’t understand your page fast, it might skip it entirely.

2. Add Schema.org Structured Data: This is code that helps AI understand what your content is about. You can add it to the <head> section of your website’s HTML.

   • For articles: Use Article schema. Here’s an example you can adapt:

     <script type="application/ld+json">
     {
       "@context": "https://schema.org",
       "@type": "Article",
       "headline": "How AI Search Engines Actually Retrieve Content",
       "author": {
         "@type": "Person",
         "name": "Your Name",
         "url": "https://yourdomain.com/about"
       },
       "datePublished": "2026-07-01",
       "dateModified": "2026-07-20",
       "publisher": {
         "@type": "Organization",
         "name": "Your Company",
         "logo": "https://yourdomain.com/logo.png"
       },
       "mainEntityOfPage": "https://yourdomain.com/blog/ai-search-retrieval"
     }
     </script>

   • For question-and-answer pages: Use FAQPage schema. This is very effective for AI answer engines. Here’s an example:

     <script type="application/ld+json">
     {
       "@context": "https://schema.org",
       "@type": "FAQPage",
       "mainEntity": [{
         "@type": "Question",
         "name": "Does Perplexity cite websites in its answers?",
         "acceptedAnswer": {
           "@type": "Answer",
           "text": "Yes, Perplexity typically cites the sources it retrieves from directly in its response."
         }
       }]
     }
     </script>

   • Other useful schemas: Consider HowTo, Product, Organization, and BreadcrumbList where they fit your content.

3. Validate Your Structured Data: Use Google’s Rich Results Test to make sure your schema code is correct. Incorrect code can be worse than no code.

   • Go to Google’s Rich Results Test.
   • Paste your website URL or paste your HTML code into the tool.
   • Click “Test URL” or “Test Code”.
   • You should see a message indicating your structured data is valid, or it will highlight any errors to fix.

4. Create an llms.txt File: This file acts as a guide for AI models, similar to how robots.txt guides web crawlers. It should be placed at the root of your website (e.g., https://yourdomain.com/llms.txt).

   • Create a plain text file named llms.txt.

   • Add a brief description of your company or website.

   • Include links to important sections of your site, like documentation or key pages, using Markdown format.

   • Here’s a basic example:

     # Your Company Name
     One-sentence description of what you do, written the way you'd pitch it in a boardroom.
     
     ## Docs
     - [Getting Started](https://yourdomain.com/docs/getting-started): Setup and installation
     - [API Reference](https://yourdomain.com/docs/api): Full endpoint documentation

   • You can use AI tools like ChatGPT or Gemini to help you write the description and structure the links for your llms.txt file. For example, you could ask:

     Write a one-sentence description for a website that offers [describe your website's purpose].
     Then, list the main sections of my website as Markdown links, like:
     - [Section Name](URL)
     - [Another Section](URL)
     Format this as the content for an llms.txt file.

5. Test with AI Tools: After implementing these changes, you can use tools like Perplexity, ChatGPT, or Gemini to see how they interpret your site. Ask them questions about your content and check if they can find and use the information effectively.

Original source

This guide is based on insights from a blog post by synfinity-dynamics-pvt-ltd on DEV Community, which explains technical steps for making websites more readable by AI search engines.

Notes & variations

• Common mistake: Using malformed or incorrect structured data. Always validate with Google’s Rich Results Test before assuming it’s helping.
• Tip for better results: Keep your llms.txt file updated. If you add new important sections to your website, add them to this file so AI models can discover them.
• Free tier alternative: While the official Schema.org types are standard, you can use ChatGPT or Gemini to help generate the JSON-LD code. Just be sure to validate it afterwards.