# Karachi Bites Robots.txt
# Halal Pakistani & Indian Restaurant — NE Calgary
# https://www.karachibitescalgary.ca

# ==============================
# Default rules for all crawlers
# ==============================
User-agent: *
Allow: /
Disallow: /config/
Disallow: /search/
Disallow: /account/
Disallow: /api/
Allow: /api/ui-extensions/
Disallow: /static/
Disallow: /*?*author=*
Disallow: /*?*tag=*
Disallow: /*?*month=*
Disallow: /*?*view=*
Disallow: /*?*format=*

# Keep the AI/LLM guide file out of search indexes.
# Noindex (unofficial, ignored by most engines) is kept as a hint,
# and Disallow blocks crawling of the file entirely.
Noindex: /llms.txt
Disallow: /llms.txt

# ==============================
# AI / LLM crawlers
# ==============================
User-agent: GPTBot
User-agent: ChatGPT-User
User-agent: OAI-SearchBot
User-agent: CCBot
User-agent: anthropic-ai
User-agent: Claude-Web
User-agent: ClaudeBot
User-agent: Google-Extended
User-agent: FacebookBot
User-agent: Meta-ExternalAgent
User-agent: PerplexityBot
User-agent: cohere-ai
User-agent: Applebot-Extended
Allow: /
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html

# ==============================
# Google Ads
# ==============================
User-agent: AdsBot-Google
User-agent: AdsBot-Google-Mobile
User-agent: AdsBot-Google-Mobile-Apps
Allow: /

# ==============================
# Heavy / aggressive bots
# ==============================
User-agent: Baiduspider
Crawl-delay: 10

# ==============================
# Sitemap
# ==============================
Sitemap: https://www.karachibitescalgary.ca/sitemap.xml