# Robots.txt - COPILOT GESTORIA
# Actualizado 2026-07-23 - Bots IA de búsqueda/citación PERMITIDOS (GEO)

# ============================================
# BOTS IA PERMITIDOS (visibilidad en ChatGPT,
# Claude, Perplexity y Gemini — GEO)
# Sin grupo propio, siguen las reglas de
# "User-agent: *" de abajo (zonas privadas
# bloqueadas, resto permitido):
#   GPTBot, OAI-SearchBot, ChatGPT-User,
#   ClaudeBot, Claude-SearchBot, Claude-User,
#   PerplexityBot, Perplexity-User,
#   Google-Extended
# ============================================

# ============================================
# BLOQUEO BOTS IA SIN VALOR DE CITACIÓN
# (solo entrenamiento, no citan fuentes)
# ============================================

# Common Crawl (alimenta muchos modelos IA)
User-agent: CCBot
Disallow: /

# ByteDance / TikTok AI
User-agent: Bytespider
Disallow: /

# Meta AI training
User-agent: FacebookBot
Disallow: /

# Apple AI training (diferente de Applebot search)
User-agent: Applebot-Extended
Disallow: /

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

# ============================================
# BLOQUEO CRAWLERS AGRESIVOS
# (descargan PDFs sin aportar valor)
# ============================================

User-agent: Barkrowler
Disallow: /

User-agent: MJ12bot
Disallow: /

# ============================================
# BUSCADORES LEGÍTIMOS (permitidos)
# Google, Bing, Apple search, Yandex, etc.
# ============================================

User-agent: *
Disallow: /admin/
Disallow: /user/
Disallow: /install/
Disallow: /api/internal/
Disallow: /webhooks/
# Datos personales de clientes servidos vía symlink: no indexar (defensa SEO;
# el control de acceso real es el 403 de Apache en public/.htaccess)
Disallow: /storage/payrolls/
Disallow: /storage/payroll/
Disallow: /storage/rgpd-contratos/
Disallow: /storage/temp/
Disallow: /storage/tpv-tmp/
# Endpoint de descarga de recursos: no rastrear (las guías requieren login)
Disallow: /recursos/*/download
Allow: /

# Sitemap URL
Sitemap: https://copilotgestoria.com/sitemap.xml