# ====================================================================
# robots.txt — Čištění a Nátěry Střech Praha (cistenistrechpraha.online)
# Generated from sitemap architecture
# ====================================================================

# Standard Search Engine Crawlers — Full Access
User-agent: *
Allow: /
Disallow: /api/
Disallow: /_astro/*.map
Disallow: /admin/
Disallow: /private/
Disallow: /*.json$
Allow: /manifest.json

# ====================================================================
# AI / LLM Crawlers — Allowed for Search & Citation
# ====================================================================

# OpenAI (ChatGPT Search, Browse)
User-agent: GPTBot
Allow: /
Disallow: /api/

User-agent: ChatGPT-User
Allow: /

# Google AI (Gemini, AI Overviews)
User-agent: Google-Extended
Allow: /

User-agent: Googlebot
Allow: /

# Anthropic (Claude)
User-agent: Anthropic-ai
Allow: /

User-agent: ClaudeBot
Allow: /

User-agent: Claude-Web
Allow: /

# Apple (Apple Intelligence, Siri)
User-agent: Applebot-Extended
Allow: /

# Perplexity AI
User-agent: PerplexityBot
Allow: /

# Cohere
User-agent: Cohere-ai
Allow: /

# You.com
User-agent: YouBot
Allow: /

# ====================================================================
# Aggressive Scrapers & Non-Search Bots — Blocked
# ====================================================================

User-agent: CCBot
Disallow: /

User-agent: Bytespider
Disallow: /

User-agent: PetalBot
Disallow: /

User-agent: AhrefsBot
Disallow: /

User-agent: SemrushBot
Disallow: /

User-agent: MJ12bot
Disallow: /

User-agent: DotBot
Disallow: /

# ====================================================================
# Sitemaps & LLM Discovery Files
# ====================================================================

Sitemap: https://cistenistrechpraha.online/sitemap-index.xml

# LLMs-Full: https://cistenistrechpraha.online/llms-full.txt
# LLMs: https://cistenistrechpraha.online/llms.txt