# ==========================================================
# Watta Sushi — Winnipeg, MB
# https://www.wattasushi.ca/
# 837 Sherbrook St, Winnipeg, MB, Canada | +1 (204) 774-7777
# ==========================================================

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

# Keep infrastructure and query-string duplicates out of the index
Disallow: /components/
Disallow: /.github/
Disallow: /.vscode/
Disallow: /*?*utm_*
Disallow: /*?*fbclid=*
Disallow: /*?*gclid=*
Disallow: /*?*replytocom=*
Disallow: /*?*view=*
Disallow: /*?*format=*
Disallow: /*?*sort=*
Disallow: /*?*search=*

# llms.txt is written for AI assistants, not for search results.
# Crawling is allowed so agents can read it; indexing is not.
Noindex: /llms.txt

# ---------- AI / LLM crawlers ----------
# Menu, dish, neighbourhood and contact content may be used to answer
# questions about Watta Sushi. Legal pages are off limits.
User-agent: GPTBot
User-agent: OAI-SearchBot
User-agent: ChatGPT-User
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: FacebookBot
User-agent: Bytespider
User-agent: Amazonbot
User-agent: DuckAssistBot
Allow: /
Allow: /llms.txt
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html
Noindex: /llms.txt

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

# ---------- Heavy / low-value crawlers ----------
User-agent: Baiduspider
Crawl-delay: 10

User-agent: YandexBot
Crawl-delay: 10

User-agent: SemrushBot
Crawl-delay: 10

User-agent: AhrefsBot
Crawl-delay: 10

User-agent: MJ12bot
Crawl-delay: 10

# ---------- Sitemap ----------
Sitemap: https://www.wattasushi.ca/sitemap.xml