# Café la Reine Robots.txt
# French Bakery & Café - Edmonton, AB
# Last Updated: July 18, 2026

# ------------------------------------------
# All bots (search engines + others)
# ------------------------------------------
User-agent: *
Allow: /

# Internal / non-content paths
Disallow: /config/
Disallow: /search/
Disallow: /account/
Disallow: /api/
Allow: /api/ui-extensions/
Disallow: /static/
Disallow: /components/

# Query-parameter URLs (avoid duplicate content)
Disallow: /*?*author=*
Disallow: /*?*tag=*
Disallow: /*?*month=*
Disallow: /*?*view=*
Disallow: /*?*format=*

# Forms and legal pages
Disallow: /reservation.html
Disallow: /enquiry.html
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html

# NOTE: /llms.txt is intentionally left crawlable so search engines can
# read its X-Robots-Tag: noindex header (served via _headers) and drop it
# from their index. AI crawlers still fetch the full file normally.

# ------------------------------------------
# AI / LLM crawlers
# ------------------------------------------
User-agent: GPTBot
User-agent: ChatGPT-User
User-agent: OAI-SearchBot
User-agent: CCBot
User-agent: anthropic-ai
User-agent: ClaudeBot
User-agent: Claude-Web
User-agent: Google-Extended
User-agent: FacebookBot
User-agent: cohere-ai
User-agent: Omgilibot
User-agent: PerplexityBot
Allow: /
Allow: /llms.txt
Disallow: /reservation.html
Disallow: /enquiry.html
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: /

# ------------------------------------------
# Bing
# ------------------------------------------
User-agent: Bingbot
Allow: /
Crawl-delay: 1

# ------------------------------------------
# Heavy / lower-priority bots
# ------------------------------------------
User-agent: Baiduspider
Crawl-delay: 10

User-agent: Yandex
Crawl-delay: 2

# ------------------------------------------
# Sitemap
# ------------------------------------------
Sitemap: https://cafelareine.online/sitemap.xml