# Merkato Restaurant — robots.txt
# https://merkatorestaurantwpg.ca

# ── All crawlers ──────────────────────────────────────────
User-agent: *
Allow: /

# Block legal pages from indexing
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html

# Block component partials (not standalone pages)
Disallow: /components/

# ── AI / LLM crawlers ─────────────────────────────────────
# llms.txt is intentionally public for AI agents
User-agent: GPTBot
Allow: /llms.txt
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html
Disallow: /components/

User-agent: ChatGPT-User
Allow: /llms.txt
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html
Disallow: /components/

User-agent: Claude-Web
Allow: /llms.txt
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html
Disallow: /components/

User-agent: anthropic-ai
Allow: /llms.txt
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html
Disallow: /components/

User-agent: CCBot
Allow: /llms.txt
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html
Disallow: /components/

User-agent: Google-Extended
Allow: /llms.txt
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html
Disallow: /components/

User-agent: cohere-ai
Allow: /llms.txt
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html
Disallow: /components/

User-agent: FacebookBot
Allow: /llms.txt
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html
Disallow: /components/

User-agent: PerplexityBot
Allow: /llms.txt
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html
Disallow: /components/

# ── Google Ads bots ────────────────────────────────────────
User-agent: AdsBot-Google
Allow: /

User-agent: AdsBot-Google-Mobile
Allow: /

# ── Courteous crawl delay for heavy scrapers ───────────────
User-agent: Baiduspider
Crawl-delay: 10

User-agent: SemrushBot
Crawl-delay: 5

User-agent: AhrefsBot
Crawl-delay: 5

# ── Sitemap ────────────────────────────────────────────────
Sitemap: https://merkatorestaurantwpg.ca/sitemap.xml