Insights · KI-Suche

Controlling AI crawlers: robots.txt for GPTBot and ClaudeBot

Jul 7, 2026 · 5 min · Daniel Giebelhaus

AI crawlers such as GPTBot, ClaudeBot or PerplexityBot visit your website so AI systems know and can cite your content. Your robots.txt controls who gets access. For most businesses the rule is: allow them, because whoever blocks disappears from AI answers.

What AI crawlers do

Classic search engine crawlers like Googlebot fill the search index. AI crawlers have two additional jobs: they collect content for training language models, and they fetch live information when a user asks a question in ChatGPT or Perplexity. Only sites reachable for these bots can appear as sources in AI answers.

The most important bots at a glance

Block or allow?

This is a business decision, not a reflex. Publishers with exclusive paid content have good reasons to block. For service providers, retailers and SMEs the maths is usually reversed: you want your content to be found, named and recommended. More and more buying decisions start with a question to an AI, and whoever does not appear there does not exist for those customers.

How to check and steer access

The robots.txt sits in the root of your website (yourdomain.ch/robots.txt). A line like "User-agent: GPTBot" followed by "Disallow: /" locks that bot out completely, "Allow: /" permits it. Check what is in there first: some SEO plugins and hosting providers block AI crawlers without asking. As a complement, an llms.txt explains your offer to AI systems in compact form.

Whether your website is optimally set up for AI search engines is exactly what we check in our GEO optimisation, including a free audit.

Daniel Giebelhaus

Daniel Giebelhaus

Founder, Mindful Consulting Giebelhaus, ZurichLinkedIn ↗

Want to put this to work for your business? We review your website for free in classic and AI search and show the biggest levers.

More articles

KI-Suche5 min

AI visibility for dental practices: what you can check

Five test questions for ChatGPT and co. and five things a dental practice can improve itself: practice data, Google profile, treatment pages, reviews, crawlers.

KI-Suche5 min

AI visibility for law firms: a self-test

Five test questions for ChatGPT and co. and five things a law firm can improve itself: firm data, Google profile, practice-area pages, people, crawlers.

KI-Suche5 min

Perplexity for companies: how to get found in Perplexity

Perplexity delivers answers with clearly visible sources. How the AI search engine works and how you get named as a source.

KI-Suche6 min

How does my business get found in ChatGPT?

Crawler access, citable content, brand authority: the three levers that get your business into AI answers.

KI-Suche5 min

Google AI Overviews: what they are and how to get cited

The AI summary above Google results is changing search behaviour. How AI Overviews work and how you appear as a cited source.

Visibility is no accident.

Let us find out together where your biggest levers are.

Request a free audit