# New Delhi Indian Restaurant - Robots.txt
# https://newdelhiindian.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 AI-guidance file out of search indexes
Noindex: /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: PerplexityBot
User-agent: Applebot-Extended
User-agent: FacebookBot
User-agent: cohere-ai
Allow: /
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html

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

# ------------------------------------------------------------
# Crawl delay for heavy bots
# ------------------------------------------------------------
User-agent: Baiduspider
Crawl-delay: 10

# ------------------------------------------------------------
# Sitemap
# ------------------------------------------------------------
Sitemap: https://newdelhiindian.ca/sitemap.xml