VibeSecurity

Free tool, runs in your browser

Generate robots.txt and llms.txt, with AI crawler controls

Choose which paths search engines should skip, decide bot by bot whether AI companies may crawl you for training, search or on-demand fetches, and download a file that is ready to deploy. There is an llms.txt builder too.

Your site
Keep out of search results (all crawlers)

Paths that only logged-in users can reach are already hidden from crawlers. Disallowing them here just saves crawl effort.

AI crawlers
  • GPTBotOpenAI
    Training

    Collects content that may be used to train OpenAI's models.

  • OAI-SearchBotOpenAI
    Search

    Finds and indexes pages so they can appear as results in ChatGPT search.

  • ChatGPT-UserOpenAI
    User-triggered

    Fetches a page when a ChatGPT user or custom GPT asks it to.

    OpenAI says that because these requests are started by a user, robots.txt rules may not apply.

  • ClaudeBotAnthropic
    Training

    Collects public web content that may be used to train Anthropic's models.

  • Claude-SearchBotAnthropic
    Search

    Crawls pages to improve the quality of Claude's search results.

  • Claude-UserAnthropic
    User-triggered

    Fetches a page when a Claude user asks a question that needs it.

  • PerplexityBotPerplexity
    Search

    Indexes pages so they can be shown and linked in Perplexity answers. Perplexity says it is not used to train foundation models.

  • Perplexity-UserPerplexity
    User-triggered

    Visits a page when a Perplexity user asks a question that needs it.

    Perplexity says this fetcher generally ignores robots.txt, because a user requested the page.

  • Google-ExtendedGoogle
    Training

    Not a separate crawler. A control token that decides whether content Google crawls may be used for Gemini model training and grounding. Blocking it does not affect Google Search.

  • Applebot-ExtendedApple
    Training

    Not a separate crawler. Controls whether pages Applebot crawls may be used to train Apple's foundation models. Blocking it does not remove you from Apple search features.

  • CCBotCommon Crawl
    Training

    Builds the Common Crawl open web archive, which many AI training datasets are built from.

  • BytespiderByteDance
    Training

    ByteDance's crawler. ByteDance publishes little documentation about how the data is used.

  • meta-externalagentMeta
    Training + other

    Meta says it is used for use cases such as training AI models or indexing content directly.

  • AmazonbotAmazon
    Training + other

    Amazon says it is used to improve its products and services, which may include training Amazon AI models.

robots.txt

User-agent: *
Disallow: /api/

robots.txt is not access control

Anyone can read this file at /robots.txt, and crawlers follow it voluntarily. Never list secret or admin paths here to hide them: that advertises them. Anything private needs a login and a server-side check.

Place the file at the root of your domain, for example https://your-app.example/robots.txt. Allowed AI crawlers fall under the * group, so they follow the same path rules as everyone else. Everything runs in your browser. Nothing you type is sent anywhere, and Download saves the file directly from this page.

Training, search and user-triggered bots are different

AI companies now run several crawlers each, and blocking one does not block the others. Deciding what to allow is easier once you know which job each one does.

TypeWhat it doesWhat blocking it means
Training (GPTBot, ClaudeBot, CCBot, Google-Extended, Applebot-Extended)Collects pages that may be used to train AI models. Google-Extended and Applebot-Extended are opt-out tokens read by the normal crawler, not separate crawlersYour content is less likely to end up in future model training. It does not remove you from normal search results
Search (OAI-SearchBot, Claude-SearchBot, PerplexityBot)Indexes pages so an AI assistant can find and cite themYou are less likely to be linked as a source in that assistant's answers
User-triggered (ChatGPT-User, Claude-User, Perplexity-User)Fetches one page because a person asked the assistant to read itThe assistant cannot open your page when a user asks. Some vendors say these fetches may not follow robots.txt

Rules that cause trouble

  • Blocking /_next/, /static/ or other folders of JavaScript and CSS. Google renders pages like a browser and needs those files. Without them your pages can look empty or broken to Google.
  • Using Disallow to remove a page from Google. A disallowed page can still be indexed if other sites link to it. Use a noindex meta tag or header instead, and leave the page crawlable so Google can see it.
  • Relative sitemap addresses. Sitemap lines must be full URLs, such as https://your-app.example/sitemap.xml.
  • Leaving a staging rule of Disallow: / on the live site. It asks every crawler to skip the whole site.

What llms.txt is, and what it is not

llms.txt is a proposed convention from llmstxt.org: a Markdown file at /llms.txt that gives language models a short summary of your site and links to the pages that matter most. The format is simple. An H1 with the name, a blockquote summary, optional paragraphs, then H2 sections containing lists of links. A section called Optional marks links that can be skipped when a shorter context is needed.

It is not a standard, and it does not control crawling. Adoption by AI tools is limited and not guaranteed, so treat it as a cheap extra rather than a ranking lever. Google says Google Search does not use llms.txt, including for its AI features, and that the file neither helps nor hurts rankings. It never replaces robots.txt or a sitemap.

llms.txt
# Acme Invoices

> Invoicing for freelancers, with a REST API and Stripe payouts.

## Docs

- [Quickstart](https://acme.example/docs/quickstart.md): Create an account and send a first invoice

## Optional

- [Changelog](https://acme.example/changelog)

Check what is live

After deploying, fetch both files from your own domain and read them back.

Terminal
curl -s https://your-app.example/robots.txt
curl -s https://your-app.example/llms.txt

Frequently asked questions

Is anything I type sent to a server?

No. Both generators run entirely in your browser. Copy uses your clipboard and Download creates the file on your device. Nothing is sent, stored or logged.

Will blocking GPTBot or ClaudeBot remove my site from ChatGPT or Claude?

Not by itself. Those are the training crawlers. OpenAI and Anthropic run separate search and user-triggered agents with their own names, which you control separately. Blocking only the training bots is what the "Allow search, block AI training" preset does.

Does blocking Google-Extended hurt my Google rankings?

Google says Google-Extended does not affect inclusion or ranking in Google Search. It only controls whether content Google crawls may be used for Gemini model training and grounding.

Can robots.txt protect my admin pages or API?

No. It is a public file that polite crawlers choose to follow. Protect admin pages and APIs with authentication and server-side authorization. Listing a sensitive path in robots.txt makes it easier to find.

Do I need an llms.txt file?

No. It is optional and adoption is limited, and Google says Google Search ignores it. If you have documentation or product pages that are useful to AI assistants, it costs little to add. Only link to pages that are already public.

Sources

  1. 1.RFC 9309: Robots Exclusion Protocol
  2. 2.Google Search Central: Introduction to robots.txt
  3. 3.Google Search Central: Google's common crawlers (Google-Extended)
  4. 4.Google Search Central: Optimizing for generative AI features (llms.txt)
  5. 5.OpenAI: Overview of OpenAI crawlers
  6. 6.Anthropic: Does Anthropic crawl data from the web?
  7. 7.Perplexity: Perplexity crawlers
  8. 8.llmstxt.org: The /llms.txt file