# robots.txt — Papa Nash's | Indian & Middle Eastern Restaurant, Airdrie, Alberta
# https://papanashs.com/

# ----------------------------------------------------------
# All search engine crawlers
# ----------------------------------------------------------
User-agent: *
Allow: /
Allow: /tags/
Allow: /assets/
Allow: /llms.txt

# Keep legal boilerplate out of the index
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html

# Internal / non-content paths
Disallow: /config/
Disallow: /search/
Disallow: /account/
Disallow: /api/
Disallow: /.github/

# Parameterised duplicates
Disallow: /*?*author=*
Disallow: /*?*tag=*
Disallow: /*?*month=*
Disallow: /*?*view=*
Disallow: /*?*format=*
Disallow: /*?*utm_*

# llms.txt is for AI agents to read, not for search engines to index
Noindex: /llms.txt

# ----------------------------------------------------------
# AI / LLM crawlers — welcome to read the site and llms.txt
# ----------------------------------------------------------
User-agent: GPTBot
User-agent: ChatGPT-User
User-agent: OAI-SearchBot
User-agent: ClaudeBot
User-agent: Claude-User
User-agent: Claude-SearchBot
User-agent: anthropic-ai
User-agent: PerplexityBot
User-agent: Perplexity-User
User-agent: Google-Extended
User-agent: Applebot-Extended
User-agent: CCBot
User-agent: cohere-ai
User-agent: meta-externalagent
User-agent: Amazonbot
User-agent: Bytespider
Allow: /
Allow: /llms.txt
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html
Noindex: /llms.txt

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

# ----------------------------------------------------------
# Image crawlers
# ----------------------------------------------------------
User-agent: Googlebot-Image
Allow: /assets/

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

User-agent: YandexBot
Crawl-delay: 10

# ----------------------------------------------------------
# Sitemap
# ----------------------------------------------------------
Sitemap: https://papanashs.com/sitemap.xml