User-agent: *
Crawl-delay: 4

# Directories
Disallow: /app/
Disallow: /bin/
Disallow: /dev/
Disallow: /lib/
Disallow: /phpserver/
Disallow: /pkginfo/
Disallow: /report/
Disallow: /setup/
Disallow: /update/
Disallow: /var/
Disallow: /vendor/

# Paths (clean URLs)
Disallow: /index.php/
Disallow: /catalog/product_compare/
Disallow: /catalog/category/view/
Disallow: /catalog/product/view/
Disallow: /catalogsearch/
Disallow: /checkout/
Disallow: /control/
Disallow: /contacts/
Disallow: /customer/
Disallow: /customize/
Disallow: /newsletter/
Disallow: /review/
Disallow: /sendfriend/
Disallow: /wishlist/
Disallow: /amcookie/cookie/cookies/
Disallow: /blog/search/

# Files
Disallow: /composer.json
Disallow: /composer.lock
Disallow: /CONTRIBUTING.md
Disallow: /CONTRIBUTOR_LICENSE_AGREEMENT.html
Disallow: /COPYING.txt
Disallow: /Gruntfile.js
Disallow: /LICENSE.txt
Disallow: /LICENSE_AFL.txt
Disallow: /nginx.conf.sample
Disallow: /package.json
Disallow: /php.ini.sample
Disallow: /RELEASE_NOTES.txt

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

# Do not index pages that are sorted or filtered.
Disallow: /*?*product_list_mode=
Disallow: /*?*product_list_order=
Disallow: /*?*product_list_limit=
Disallow: /*?*product_list_dir=

# Do not index layered navigation pages
Disallow: /*?*merken=
Disallow: /*?*haartype=
Disallow: /*?*ingedienten=
Disallow: /*?*query=

# Do not index session ID
Disallow: /*?SID=
# Disallow: /*? disabled ref T45489 ivm sessie van channable
Disallow: /*.php$

# CVS, SVN directory and dump files
Disallow: /*.CVS
Disallow: /*.Zip$
Disallow: /*.Svn$
Disallow: /*.Idea$
Disallow: /*.Sql$
Disallow: /*.Tgz$

### User agent specific ###

User-Agent: bingbot
Crawl-Delay: 20

User-Agent: msnbot
Crawl-Delay: 20

User-Agent: PetalBot
Crawl-Delay: 20

User-agent: Seekport Crawler
Disallow: /

User-Agent: Jyxobot
Disallow: /

User-Agent: MJ12bot
Disallow: /

User-Agent: ShopWiki
Disallow: /

User-Agent: Fasterfox
Disallow: /

User-Agent: Baiduspider
Disallow: /

User-Agent: CazoodleBot
Disallow: /

# OpenAI trainingsbots (géén toegang)
User-agent: GPTBot
Disallow: /

User-agent: OAI-ImageBot
Disallow: /

User-agent: OAI-ResearchBot
Disallow: /

# AI-training en extra scraping blokkeren
User-agent: Google-Extended
Disallow: /

User-agent: GoogleOther
Disallow: /

User-agent: GoogleOther-Image
Disallow: /

User-agent: PerplexityBot
Disallow: /

User-agent: PlexBot
Disallow: /

User-agent: PerplexityAI
Disallow: /

# ===== Anthropic / Claude blokkeren =====
User-agent: anthropic-ai
Disallow: /

User-agent: ClaudeBot
Disallow: /

User-agent: claude-web
Disallow: /

# ===== Overige AI-trainings- en scrapingbots blokkeren =====
User-agent: meta-externalagent
Disallow: /

User-agent: cohere-ai
Disallow: /

User-agent: CCBot
Disallow: /

User-agent: AI2Bot
Disallow: /

User-agent: omgili
Disallow: /

User-agent: omgilibot
Disallow: /

User-agent: Bytespider
Disallow: /


Sitemap: https://www.haarpro.nl/media/sitemap.xml