# Robots.txt for halostatue.ca

# Standard search engines - welcome
User-agent: Googlebot
Allow: /

User-agent: Bingbot
Allow: /

User-agent: DuckDuckBot
Allow: /

User-agent: Applebot
Allow: /

# AI Training Crawlers - explicitly blocked
# OpenAI
User-agent: GPTBot
Disallow: /

User-agent: ChatGPT-User
Disallow: /

User-agent: OAI-SearchBot
Disallow: /

# Anthropic
User-agent: anthropic-ai
Disallow: /

User-agent: Claude-Web
Disallow: /

User-agent: ClaudeBot
Disallow: /

# Google AI (Gemini training, not search)
User-agent: Google-Extended
Disallow: /

# Meta
User-agent: FacebookBot
Disallow: /

User-agent: Meta-ExternalAgent
Disallow: /

User-agent: meta-externalagent
Disallow: /

# Apple Intelligence
User-agent: Applebot-Extended
Disallow: /

# Common Crawl (used for AI training datasets)
User-agent: CCBot
Disallow: /

# Perplexity
User-agent: PerplexityBot
Disallow: /

# Cohere
User-agent: cohere-ai
Disallow: /

# Amazon/AWS
User-agent: Amazonbot
Disallow: /

# ByteDance
User-agent: Bytespider
Disallow: /

# Musk-owned garbage - explicit blocks
# X/Twitter
User-agent: Twitterbot
Disallow: /

# xAI/Grok
User-agent: Grok
Disallow: /

User-agent: xAI
Disallow: /

User-agent: xai-ai
Disallow: /

# Generic AI scrapers
User-agent: AI2Bot
Disallow: /

User-agent: Ai2Bot-Dolma
Disallow: /

User-agent: Diffbot
Disallow: /

User-agent: ImagesiftBot
Disallow: /

User-agent: Omgilibot
Disallow: /

User-agent: Omgili
Disallow: /

User-agent: YouBot
Disallow: /

User-agent: PetalBot
Disallow: /

User-agent: Scrapy
Disallow: /

User-agent: img2dataset
Disallow: /

User-agent: Webzio-Extended
Disallow: /

User-agent: sentibot
Disallow: /

User-agent: iaskspider
Disallow: /

# Default - allow legitimate crawlers
User-agent: *
Allow: /

# Sitemap
Sitemap: https://halostatue.ca/sitemap.xml