# Viva Restaurant — robots.txt
# Website: https://www.vivarestaurant.ca

# ─── General Search Crawlers ────────────────────────────────────────────────
User-agent: *
Allow: /

# Exclude non-public utility paths
Disallow: /config/
Disallow: /search/
Disallow: /account/
Disallow: /api/
Disallow: /static/
Disallow: /components/

# Exclude legal / policy pages (low SEO value)
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html

# Noindex llms.txt for traditional search engines
# (kept allowed for AI/LLM crawlers below)
Disallow: /llms.txt

# Exclude parameterised URLs
Disallow: /*?*author=*
Disallow: /*?*tag=*
Disallow: /*?*month=*
Disallow: /*?*view=*
Disallow: /*?*format=*

# ─── AI & LLM Bots ──────────────────────────────────────────────────────────
User-agent: GPTBot
User-agent: ChatGPT-User
User-agent: CCBot
User-agent: anthropic-ai
User-agent: Claude-Web
User-agent: ClaudeBot
User-agent: Google-Extended
User-agent: FacebookBot
User-agent: cohere-ai
User-agent: PerplexityBot
User-agent: YouBot
User-agent: Applebot-Extended
User-agent: Bytespider
User-agent: Amazonbot
User-agent: Meta-ExternalAgent
Allow: /
Allow: /llms.txt
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html
Disallow: /components/

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

# ─── Slow / Aggressive Crawlers ─────────────────────────────────────────────
User-agent: Baiduspider
Crawl-delay: 10

User-agent: SemrushBot
Crawl-delay: 5

User-agent: AhrefsBot
Crawl-delay: 5

# ─── Sitemap ────────────────────────────────────────────────────────────────
Sitemap: https://www.vivarestaurant.ca/sitemap.xml