Skip to content
ansezz.

▸ Free tool

llms.txt, Written and Checked.

Build an llms.txt file from a short form, or paste one to check it against the llmstxt.org format. Everything runs in your browser.

▸ Checked October 10, 2026

Follows v2 of the llmstxt.org proposal, updated August 10, 2026. This site's own /llms.txt passes the validator.

1. Generate a file.

Becomes the H1. This is the only required part.

Written as a blockquote under the title. Agents read it first.

Markdown paragraphs and lists. No headings here; the format reserves them for the title and the link sections.

Link sections

Start from a sitemap or a URL list

Read in your browser, nothing is fetched. URLs are grouped by their first folder. Up to 200 links, so trim the list to the pages that matter.

Your llms.txt

 
    
 
     
      
     
     
    
The link stores the form in the URL. Do not put anything private in
          it.

      
     
     
    2. Validate a file.
     
      
       
    
Checked live in your browser. Never saved in the URL.

     
     
       
     
    
Paste a file to check it.

      
        
         
        
       
       
      What goes in an llms.txt file
       
      
An llms.txt file is a markdown guide to your site for AI agents. A
        coding agent fetches it to find the right API page, and a chat assistant
        reads it to answer questions about your product. Web pages carry
        navigation, scripts and layout that cost tokens and hide the content, so
        the file points agents straight at the pages worth reading.

       
      The format is plain markdown in a fixed order:
       
         
      1. 
An H1 with the site or project name. This is the only required
          part.

        2.  
      2. 
A blockquote with a short summary, the key facts an agent
          needs to read the rest.

        3.  
      3. 
Any number of paragraphs or lists with more detail. No
          headings in this part.

        4.  
      4. 
Any number of H2 sections, each a list of links
          written as - [name](url): notes. The
          notes are optional.

        5.  
       
      
By convention an H2 named Optional holds secondary
        links an agent can skip. The generator writes exactly this structure and escapes
        brackets in titles and spaces or parentheses in URLs, so the links parse.
        The validator reads a pasted file the same way a simple parser would and reports
        each problem with its line number: content before the title, headings in the
        wrong place, list items without a link, broken or relative URLs, duplicates
        and empty sections.

       
      What changed in v2
       
      
The proposal was updated on August 10, 2026. A file can now sit under
        any path, like /docs/llms.txt, and covers
        the pages under it; the most specific file wins. Markdown copies of
        pages can use either
page.html.md or page.md. Pages can point to their markdown copy with rel="alternate" type="text/markdown" and to the llms.txt that covers them with rel="describedby", as a link tag or a Link: header. The Optional
        section is still allowed but no longer has a special processing meaning.

       
      Writing a good one
       
      
Keep it short. The file should fit in an agent's context with room to
        spare, and the detail lives behind the links. Write notes that say what
        a page answers, not what it is called. Link to markdown copies when you
        have them. Then test it: give an agent only your llms.txt and ask it
        three questions a customer would ask. If it picks the wrong page, fix
        the notes.

       
      
Sources, checked October 10, 2026:  The /llms.txt file, v2 (llmstxt.org),  llmstxt.org changes for v2.

       
       
        
       
      Questions, answered
       
       
        What is an llms.txt file?   
      A markdown file at /llms.txt that gives AI agents a short guide to your site: what it is, and which pages to read for which questions. It was proposed by Jeremy Howard in September 2024 and updated to v2 in August 2026. Agents read it, then follow the links that match the task, so the file itself stays small.
       
        What does the format require?   
      Only an H1 with the name of the site or project. After that, in this order: an optional blockquote summary, optional paragraphs or lists with more detail (no headings), then optional H2 sections that each hold a list of links written as - [name](url), with an optional colon and a note. The validator checks that order and every link line.
       
        Is llms.txt the same as robots.txt or sitemap.xml?   
      No. robots.txt tells crawlers what they may fetch, and sitemap.xml lists every page for search engines. llms.txt is a short, hand-picked reading list for agents that are helping someone right now. You can start from your sitemap here with the import box, then cut it down to the pages that answer real questions.
       
        Should the links point to markdown files?   
      If you have them, yes. The proposal suggests serving a clean markdown copy of a page at the same URL with .md added, like page.html.md or page.md, because markdown costs far fewer tokens than HTML. Plain HTML links still work. The validator only counts how many of your links are markdown, it does not fail the file for it.
       
        What is the Optional section for?   
      By convention, an H2 called Optional holds secondary links an agent can skip when it is short on context. In v1 of the proposal some tools dropped that section automatically. Since v2 it is only a convention with no special processing, so you can use it or leave it out.
       
        Where do I put the file, and can I have more than one?   
      Put it at the root, https://yoursite.com/llms.txt. v2 also allows a file under a path, like /docs/llms.txt, which covers the pages under that path. When more than one file applies, agents should use the most specific one. Anything you type here stays in your browser; the share link holds your form in the URL so you can send it to a teammate.
       
       
       
       
       
      
▸ Last verified:  
         
      
Need this in production?

       
      Shipping AI features? I build RAG pipelines, MCP servers and agents that hold up in production.
       
       AI & MCP Integration  
               
      
Also related:   Shopify Plus Apps  
.

       
       
      Keep reading