# Friendz Dhaba — Indian & Punjabi Dhaba Restaurant
# 10A Bram Court, Brampton, Ontario, L6W 3R6
# https://www.friendzdhaba.ca/

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

# Rendering assets must stay crawlable — the header and footer are
# fetched at runtime by components.js, so blocking them would hide
# the site navigation from search engines.
Allow: /assets/
Allow: /components/
Allow: /*.css$
Allow: /*.js$

# No query strings are used by this static site; block crawl traps.
Disallow: /*?*author=*
Disallow: /*?*tag=*
Disallow: /*?*month=*
Disallow: /*?*view=*
Disallow: /*?*format=*
Disallow: /*?*replytocom=*

# Housekeeping files that should never appear in search results.
Disallow: /.DS_Store
Disallow: /.vscode/
Disallow: /.github/
Disallow: /update-menu.js

# llms.txt is published for AI agents to read, but it should not be
# indexed as a search result. Keep it crawlable, keep it out of the index.
Noindex: /llms.txt

# ---------- AI / LLM crawlers ----------
# Full access to the content surface, minus the legal boilerplate.
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: Claude-Web
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
Allow: /
Disallow: /privacy-policy.html
Disallow: /terms-of-use.html
Noindex: /llms.txt

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

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

User-agent: YandexBot
Crawl-delay: 10

User-agent: SemrushBot
Crawl-delay: 10

User-agent: AhrefsBot
Crawl-delay: 10

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