# Lava Pizza YYC Robots.txt
# Last Updated: 2026-08-17

# Search engine crawlers
# Note: llms.txt is intentionally left crawlable here (not Disallowed) so these
# bots can fetch it and see the X-Robots-Tag: noindex header set in .htaccess.
# Disallowing it instead would block the crawl entirely, meaning the noindex
# signal would never be seen and the URL could still surface in search results
# without a snippet if linked to from elsewhere.
User-agent: Googlebot
User-agent: Bingbot
User-agent: Slurp
User-agent: DuckDuckBot
User-agent: YandexBot
Allow: /
Disallow: /admin/
Disallow: /config/
Disallow: /search/
Disallow: /api/
Disallow: /static/
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html
Disallow: /logs/

# Main directives for all other bots
User-agent: *
Allow: /
Disallow: /admin/
Disallow: /config/
Disallow: /search/
Disallow: /api/
Disallow: /static/
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html
Disallow: /logs/

# AI Bot specific directives — allow full access including llms.txt
User-agent: GPTBot
User-agent: ChatGPT-User
User-agent: CCBot
User-agent: anthropic-ai
User-agent: Google-Extended
User-agent: FacebookBot
User-agent: Claude-Web
User-agent: cohere-ai
User-agent: PerplexityBot
User-agent: YouBot
Allow: /
Allow: /llms.txt
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html
Disallow: /logs/

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

# Sitemap location
Sitemap: https://lavapizzayyc.com/sitemap.xml

# Crawl delay for heavy bots
User-agent: Baiduspider
Crawl-delay: 10