# Crawlers Setup
User-agent: *
Crawl-delay: 20

Sitemap: https://www.unibrew-nederland.nl/sitemap.xml

# Directories
Disallow: /404/
Disallow: /app/
Disallow: /cgi-bin/
Disallow: /downloader/
Disallow: /errors/
Disallow: /includes/
#Disallow: /js/
Disallow: /lib/
Disallow: /magento/
Disallow: /pkginfo/
Disallow: /report/
Disallow: /scripts/
Disallow: /shell/
#Disallow: /skin/
Disallow: /stats/
Disallow: /var/
Disallow: /catalogsearch/result/

# Media 
Disallow: /media/captcha/
#Disallow: /media/css/
#Disallow: /media/css_secure/
Disallow: /media/customer/
Disallow: /media/dhl/
Disallow: /media/downloadable/
Disallow: /media/import/
#Disallow: /media/js/
Disallow: /media/pdf/
Disallow: /media/sales/
Disallow: /media/tmp/
Disallow: /media/wysiwyg/
Disallow: /media/xmlconnect/

# Paths (clean URLs)
Disallow: /*?p=
Disallow: /index.php/
Disallow: /catalog/product_compare/
Disallow: /catalog/category/view/
Disallow: /catalog/product/view/
Disallow: /catalog/product/gallery/
Disallow: /catalogsearch/
Disallow: /checkout/
Disallow: /control/
Disallow: /contacts/
Disallow: /customer/
Disallow: /customize/
Disallow: /newsletter/
Disallow: /poll/
Disallow: /review/
Disallow: /sendfriend/
Disallow: /tag/
Disallow: /wishlist/
Disallow: /inloggen

# Files
Disallow: /cron.php
Disallow: /cron.sh
Disallow: /error_log
Disallow: /install.php
Disallow: /LICENSE.html
Disallow: /LICENSE.txt
Disallow: /LICENSE_AFL.txt
Disallow: /STATUS.txt
Disallow: /get.php

# Paths (no clean URLs)
#Disallow: /*.js$
#Disallow: /*.css$
Disallow: /*.php$
Disallow: /*?p=*&
Disallow: /*?SID= 
Disallow: /*?$
Disallow: /rss*
Disallow: /*PHPSESSID

# Bot-trap 
User-Agent: *
Disallow: /bot-trap

# Webmail
Disallow: /roundcube
Disallow: /webmail
Disallow: /squirrelmail

# MageHost: prevent Google from going wild on filter/sorting params
Disallow: *?*%20*
Disallow: *?*%25*
Disallow: *?*%2C*
Disallow: *?*,*
Disallow: *?*%5B*
Disallow: *?limit=*
Disallow: *&limit=*
Disallow: *?order=*
Disallow: *&order=*
Disallow: *?dir=*
Disallow: *&dir=*
Disallow: *?mode=*
Disallow: *&mode=*
Disallow: *?___store=*
Disallow: *&___store=*
Disallow: *?___from_store=*
Disallow: *&___from_store=*
Disallow: *?SID=*
Disallow: *&SID=*
Disallow: *?___SID=*
Disallow: *&___SID=*
Disallow: */sort-by/
Disallow: */sort-direction/
Disallow: */limit/
Disallow: */dir/
Disallow: */order/
Disallow: */catalogsearch/
Disallow: */search/
Disallow: */customer/
Disallow: */checkout/
Disallow: */errors/
Disallow: */onestepcheckout/
Disallow: */maten/
Disallow: */maat/
Disallow: */price/
Disallow: *?p=

# Specifically block Googlebot from indexing filter pages
User-agent: Googlebot
Disallow: */catalogsearch/
Disallow: */search/
Disallow: */customer/
Disallow: */checkout/

# User-Agents
User-Agent: Baiduspider
Disallow: /
User-Agent: CazoodleBot
Disallow: /
User-Agent: Fasterfox
Disallow: /
User-Agent: Jyxobot
Disallow: /
User-Agent: MJ12bot
Disallow: /
User-Agent: ShopWiki
Disallow: /
User-Agent: AhrefsBot
Disallow: /
User-Agent: voltron
Disallow: /
User-agent: Googlebot
Disallow:
User-agent: Googlebot-image
Disallow:

# Allow AI search and agent use
User-agent: OAI-SearchBot
User-agent: ChatGPT-User
User-agent: PerplexityBot
User-agent: AndiBot
User-agent: PhindBot
User-agent: YouBot
Allow: /

# Disallow AI training data collection
User-agent: GPTBot
User-agent: CCBot
User-agent: Google-Extended
Disallow: /