# Upcode Studios - Robots.txt
# https://upcode.com

# ============================================
# SEARCH ENGINE CRAWLERS
# ============================================

User-agent: Googlebot
Allow: /

User-agent: Googlebot-Image
Allow: /

User-agent: Bingbot
Allow: /

User-agent: Slurp
Allow: /

User-agent: DuckDuckBot
Allow: /

User-agent: Yandex
Allow: /

User-agent: Baiduspider
Allow: /

# ============================================
# AI & LLM CRAWLERS (Explicitly Allowed)
# ============================================

# OpenAI's GPTBot - Powers ChatGPT web browsing
User-agent: GPTBot
Allow: /

# OpenAI's ChatGPT-User - ChatGPT browsing mode
User-agent: ChatGPT-User
Allow: /

# Google's AI crawler
User-agent: Google-Extended
Allow: /

# Anthropic's Claude web crawler
User-agent: ClaudeBot
Allow: /
User-agent: Claude-Web
Allow: /

# Perplexity AI crawler
User-agent: PerplexityBot
Allow: /

# Cohere AI crawler
User-agent: cohere-ai
Allow: /

# Meta AI crawler
User-agent: Meta-ExternalAgent
Allow: /
User-agent: FacebookBot
Allow: /

# Amazon/Alexa crawler
User-agent: Amazonbot
Allow: /

# Apple's crawler (Siri, Spotlight)
User-agent: Applebot
Allow: /
User-agent: Applebot-Extended
Allow: /

# Microsoft/Bing AI crawler
User-agent: bingbot
Allow: /

# Common Crawl (used by many AI training)
User-agent: CCBot
Allow: /

# ============================================
# SOCIAL MEDIA CRAWLERS
# ============================================

User-agent: Twitterbot
Allow: /

User-agent: facebookexternalhit
Allow: /

User-agent: LinkedInBot
Allow: /

User-agent: Slackbot
Allow: /

User-agent: Discordbot
Allow: /

User-agent: WhatsApp
Allow: /

User-agent: TelegramBot
Allow: /

# ============================================
# GENERAL RULES
# ============================================

User-agent: *
Allow: /

# Block non-essential paths (internal/test pages)
Disallow: /api/
Allow: /dashboard
Disallow: /home-old
Disallow: /home2
Disallow: /home3
Disallow: /home4
Disallow: /home5
Disallow: /home9
Disallow: /platformcards

# ============================================
# SITEMAPS & AI-READABLE CONTENT
# ============================================

# Primary sitemap
Sitemap: https://upcode.com/sitemap.xml

# LLM-friendly content files
# llms.txt - Concise summary for AI inference
# llms-full.txt - Comprehensive documentation for AI

# ============================================
# HOST & CRAWL SETTINGS
# ============================================

Host: https://upcode.com

# Respectful crawl delay
Crawl-delay: 1