# Cafe Lake Louise - robots.txt
# https://cafelakelouise.com/

# ==========================================
# All crawlers
# ==========================================
User-agent: *
Allow: /
Allow: /tags/
Allow: /assets/

# Source files and fragments that are not pages
Disallow: /Store%20Information.txt
Disallow: /update-menu.js
Disallow: /.github/
Disallow: /.vscode/

# Parameterised and duplicate URLs
Disallow: /*?*author=*
Disallow: /*?*tag=*
Disallow: /*?*month=*
Disallow: /*?*view=*
Disallow: /*?*format=*
Disallow: /*?*replytocom=*

# llms.txt is published for AI agents to read, not for search results
Allow: /llms.txt
Noindex: /llms.txt

# ==========================================
# AI / LLM crawlers and assistants
# ==========================================
User-agent: GPTBot
User-agent: ChatGPT-User
User-agent: OAI-SearchBot
User-agent: ClaudeBot
User-agent: Claude-Web
User-agent: anthropic-ai
User-agent: PerplexityBot
User-agent: Perplexity-User
User-agent: Google-Extended
User-agent: Applebot-Extended
User-agent: Meta-ExternalAgent
User-agent: FacebookBot
User-agent: Amazonbot
User-agent: DuckAssistBot
User-agent: MistralAI-User
User-agent: cohere-ai
User-agent: CCBot
Allow: /
Allow: /llms.txt
Disallow: /Store%20Information.txt
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
User-agent: Bytespider
Crawl-delay: 10

# ==========================================
# Sitemap
# ==========================================
Sitemap: https://cafelakelouise.com/sitemap.xml