Free tool · GEO

Get your site ready for ChatGPT and friends

Create an llms.txt that explains your business to large language models, check which AI crawlers your robots.txt allows and generate a setup that keeps you visible in AI assistants without handing over your content for training.

Follows the format proposed at llmstxt.org.
The name is required: it's the file's title (H1).
Turns paths like /services into full links.Enter the full URL, including https://
Write a short summary: it's the most important part.
Facts you want AI to repeat correctly: location, hours, prices, what sets you apart.

Sections with links

llms.txt

—
    
      
      
    
      
    Upload it to the root of your domain so it lives at yoursite.com/llms.txt. It's plain-text Markdown, no HTML tags.
    
    
    
  
    

  
  

  
  

    

    

    
  
    
    
    
      
    01How it works
    Three steps and done
    
      
    
    
    
    1. 01
      Create your llms.txt
      Enter your name, a summary, key facts and your most important pages with a short description. Download the file and upload it to the root of your domain.
    2. 02
      Check your robots.txt
      Paste your robots.txt and see, bot by bot, whether Googlebot, GPTBot, ClaudeBot, PerplexityBot and the rest can read your site, and which rule decides it.
    3. 03
      Apply the recommended policy
      Choose whether to appear in AI answers without allowing training, fine-tune each crawler and copy the generated robots.txt.
    
  
    

    

    
  
    
    
    
      
    02GEO: generative engine optimization
    Not every AI bot does the same thing
    
      
    
    
    
    AI crawlers fall into three families. Training crawlers (GPTBot, ClaudeBot, CCBot, meta-externalagent, Bytespider) download content to train future models; blocking them doesn't cost you visibility today. AI search crawlers (OAI-SearchBot, Claude-SearchBot, PerplexityBot) index your site so it can be shown and cited when someone asks ChatGPT, Claude or Perplexity; blocking them does take you out of those answers. And user-triggered agents (ChatGPT-User, Claude-User, Perplexity-User) visit a specific page because someone asked the assistant to.
    There are also two control tokens: Google-Extended and Applebot-Extended. They aren't separate crawlers; Google and Apple crawl with their regular bots and use these names only to know whether they may use your content to train Gemini or Apple's models. Blocking Google-Extended does not affect your Google rankings or your presence in AI Overviews, which depend on Googlebot.
    A robots.txt file is read in groups: each bot looks for the group with its exact name and, if there isn't one, uses the User-agent: * group. Within a group, the Allow or Disallow rule with the longest matching path wins, and Allow wins ties. A common mistake is creating a group for an AI bot and forgetting to repeat your private paths: that bot no longer reads the * rules. The checker shows you exactly which rule applies in each case.
    llms.txt is an open proposal (llmstxt.org) for a Markdown summary at the root of your site with the essentials of your business and links to your key pages. It isn't yet a standard every assistant reads, but it takes five minutes, helps agents that browse your site and forces you to organize the information you want AI to repeat: who you are, what you offer, where and at what price.
    • llms.txt builder
      Summary, key facts, sections with links and an Optional section, with a live preview and file download.
    • Spec-accurate checker
      Reads User-agent groups and * and $ wildcards, and resolves by the longest rule, just like Google and RFC 9309.
    • 16 search and AI crawlers
      Googlebot, Bingbot, GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, Claude-SearchBot, PerplexityBot, Google-Extended and more.
    • Test any path
      Besides the root, check a specific path like /blog/ or /pricing to see which bots can read it.
    • Recommended robots.txt
      Three ready-made policies plus per-bot tweaks, with your private paths repeated in every group so nothing slips through.
    • Everything in your browser
      We don't upload your robots.txt or your content to any server.
    
  
    

    

    
  
    
    
    
      
    03More free tools
    Keep measuring
    
      
    
    
    
    SEO + GEO AuditFree SEO and AI search auditWebsite Cost CalculatorPrice and timeline in secondsROI and ROAS CalculatorAre your ads making money?UTM BuilderTrackable links + QR codeGoogle SERP PreviewTitle and description in pixelsLocal Business Schema GeneratorLocalBusiness and FAQ JSON-LDEisenhower MatrixEisenhower MatrixPassword generatorStrong, random and never leaves your deviceLink shortenerShort links with QR codes and click statsQR code generatorWi-Fi, WhatsApp, vCard and more, with your logoDigital transformation assessmentYour digital maturity in 5 minutesAutomation savings calculatorHours and money you get backWhatsApp link generatorLink, QR code and website buttonEmail signature generatorHTML signature for Gmail and OutlookMexico IVA calculator (SAT)Add, remove and withhold Mexican VATSleep calculatorWhen to go to bed or wake upIntermittent fasting calculatorEating window and fasting stages
    
  
    

    

    
  
    04FAQ
    What people ask us
    
  
    Should I block GPTBot and other AI bots?
    It depends on your goal. If you want ChatGPT, Claude or Perplexity to recommend you, don't block their search bots (OAI-SearchBot, Claude-SearchBot, PerplexityBot). You can block the training bots (GPTBot, ClaudeBot, CCBot) without losing visibility in today's answers.
    Does blocking Google-Extended remove me from Google or AI Overviews?
    No. Google-Extended only controls whether your content is used to train and improve Gemini. Google Search and AI Overviews depend on Googlebot, which follows its own rules.
    What is llms.txt and who reads it?
    It's a Markdown file at the root of your site that summarizes your business and links to your key pages so language models can understand it quickly. It's read mainly by agents and tools that browse websites; the big search engines don't use it as an official signal yet, but it's cheap to implement.
    Do bots respect robots.txt?
    The crawlers run by major companies (Google, Microsoft, OpenAI, Anthropic, Apple, Perplexity) say they do. Agents acting on a user's request may be treated differently, so check each company's documentation. To block bots that don't comply, you need rules on your server or firewall.
    How often should I review my robots.txt?
    Every time you redesign your site, switch platforms or a relevant new bot appears. A robots.txt left over from a site under construction with Disallow: / is one of the most common reasons a site disappears from Google.
    

    

    
  

    

    

Message us on WhatsApp