# robots.txt – SEOX / Publisher - 2026-01

# ------------------------------------------------------------------
# BLOQUEIO GERAL & ROTAS ADMINISTRATIVAS
# ------------------------------------------------------------------
User-agent: *
Disallow: /wp-admin/
Allow: /wp-admin/admin-ajax.php
Disallow: /wp-login.php
Disallow: /xmlrpc.php
Disallow: /?s=
Disallow: /search
Disallow: /trackback/
Disallow: /readme.html
Disallow: /license.txt
Disallow: /cdn-cgi/
Disallow: /private/

# ------------------------------------------------------------------
# AI CRAWLERS & LLM SCRAPERS
# ------------------------------------------------------------------
User-agent: CCBot
User-agent: PerplexityBot
User-agent: anthropic-ai
User-agent: ClaudeBot
User-agent: Omgilibot
User-agent: Diffbot
User-agent: Applebot-Extended
Disallow: /

# ------------------------------------------------------------------
# SEO / SCRAPING BOTS AGRESSIVOS
# ------------------------------------------------------------------
User-agent: Amazonbot
User-agent: PetalBot
User-agent: SemrushBot
User-agent: SemrushBot-SA
User-agent: Bytespider
User-agent: SeobilityBot
User-agent: AhrefsBot
User-agent: BLEXBot
User-agent: DotBot
User-agent: linkdexbot
User-agent: Scrapy
User-agent: YandexBot
User-agent: Baiduspider
User-agent: Sogou
User-agent: MJ12bot
Disallow: /

# ------------------------------------------------------------------
# SITEMAPS
# ------------------------------------------------------------------
Sitemap: https://network.grupodicas.com/news-sitemap.xml
Sitemap: https://network.grupodicas.com/sitemap_index.xml