# ================================
# AI / LLM Crawlers
# ================================

User-agent: GPTBot
Allow: /

User-agent: OAI-SearchBot
Allow: /

User-agent: Google-Extended
Allow: /

User-agent: ClaudeBot
Allow: /

User-agent: Claude-SearchBot
Allow: /

User-agent: CCBot
Allow: /


# ================================
# Default crawlers (Shopify base)
# ================================

User-agent: *
Disallow: /a/downloads/-/*
Disallow: /admin
Disallow: /cart
Disallow: /carts
Disallow: /checkout
Disallow: /checkouts/
Disallow: /orders
Disallow: /account

Disallow: /53202780351/checkouts
Disallow: /53202780351/orders

# Faceted / filtered URLs
Disallow: /collections/*sort_by*
Disallow: /*/collections/*sort_by*
Disallow: /collections/*+*
Disallow: /collections/*%2B*
Disallow: /collections/*%2b*
Disallow: */collections/*filter*&*filter*

Disallow: /blogs/*+*
Disallow: /blogs/*%2B*
Disallow: /blogs/*%2b*

# Tracking & preview URLs
Disallow: /*?*oseid=*
Disallow: /*preview_theme_id*
Disallow: /*preview_script_id*

# Internal / legal
Disallow: /policies/
Disallow: /*/policies/

# Locale / session params
Disallow: /*/*?*ls=*&ls=*
Disallow: /*/*?*ls%3D*%3Fls%3D*
Disallow: /*/*?*ls%3d*%3fls%3d*

# Internal search
Disallow: /search

# Shopify internals
Disallow: /sf_private_access_tokens
Disallow: /apple-app-site-association
Disallow: /.well-known/shopify/monorail
Disallow: /cdn/wpm/*.js

# Product recommendations
Disallow: /recommendations/products
Disallow: /*/recommendations/products

# Shopify remote product URLs
Disallow: /products/*-[a-f0-9]{8}-remote
Disallow: /*/products/*-[a-f0-9]{8}-remote
Disallow: /collections/*/products/*-[a-f0-9]{8}-remote
Disallow: /*/collections/*/products/*-[a-f0-9]{8}-remote


# ================================
# Google Ads bot
# ================================

User-agent: adsbot-google
Disallow: /checkouts/
Disallow: /checkout
Disallow: /carts
Disallow: /cart
Disallow: /orders
Disallow: /53202780351/checkouts
Disallow: /53202780351/orders
Disallow: /*?*oseid=*
Disallow: /*preview_theme_id*
Disallow: /*preview_script_id*
Disallow: /cdn/wpm/*.js
Disallow: /sf_private_access_tokens


# ================================
# Ahrefs (allowed, rate limited)
# ================================

User-agent: AhrefsBot
Crawl-delay: 10
Allow: /

User-agent: AhrefsSiteAudit
Crawl-delay: 10
Allow: /


# ================================
# Other crawlers
# ================================

User-agent: Nutch
Disallow: /

User-agent: MJ12bot
Crawl-delay: 10
Disallow: /

User-agent: Pinterest
Crawl-delay: 1
Allow: /


# ================================
# Sitemap
# ================================

Sitemap: https://thirtytwo.com/sitemap.xml