# ----------------------------------------------------
# robots.txt for Come2Indonesia
# SEO + Security + AI Alignment + Cloudflare + Wordfence optimized
# ----------------------------------------------------

# ----------------------------------------------------
# GLOBAL RULES
# ----------------------------------------------------
User-agent: *
Disallow: /wp-login.php
Disallow: /xmlrpc.php
Disallow: /cart/
Disallow: /checkout/
Disallow: /my-account/

# WordPress admin protected but allow Ajax
Disallow: /wp-admin/
Allow: /wp-admin/admin-ajax.php

# Block Wordfence internal endpoints
Disallow: /?wfls-*
Disallow: /*?wc-ajax=*

# ----------------------------------------------------
# BLOCK BAD / AGGRESSIVE / LOW-QUALITY CRAWLERS
# ----------------------------------------------------
User-agent: AhrefsBot
Disallow: /
User-agent: AhrefsSiteAudit
Disallow: /
User-agent: SemrushBot
Disallow: /
User-agent: MJ12bot
Disallow: /
User-agent: DotBot
Disallow: /
User-agent: PetalBot
Disallow: /
User-agent: Sogou
Disallow: /
User-agent: Yandex
Disallow: /
User-agent: YandexBot
Disallow: /

# Unwanted AI scrapers (massive data extraction)
User-agent: omgili
Disallow: /
User-agent: Omgilibot
Disallow: /
User-agent: Bytespider
Disallow: /
User-agent: cohere-training-data-crawler
Disallow: /

# ----------------------------------------------------
# ALLOW GOOD AI INDEXERS (for visibility in LLMs)
# ----------------------------------------------------
User-agent: GPTBot
Allow: /

User-agent: ChatGPT-User
Allow: /

User-agent: OAI-SearchBot
Allow: /

User-agent: ClaudeBot
Allow: /

User-agent: Claude-Web
Allow: /

User-agent: Claude-SearchBot
Allow: /

User-agent: PerplexityBot
Allow: /

User-agent: Perplexity-User
Allow: /

User-agent: Google-Extended
Allow: /

User-agent: Gemini-Deep-Research
Allow: /

User-agent: Meta-ExternalFetcher
Allow: /

User-agent: meta-externalagent
Allow: /

User-agent: Applebot-Extended
Allow: /

User-agent: Amazonbot
Allow: /

User-agent: YouBot
Allow: /

User-agent: MistralAI-User
Allow: /

User-agent: cohere-ai
Allow: /

User-agent: Webzio
Allow: /

User-agent: Webzio-Extended
Allow: /

# Generic fallback for future LLMs
User-agent: anthropic-ai
Allow: /

# ----------------------------------------------------
# SITEMAPS
# ----------------------------------------------------
Sitemap: https://come2indonesia.com/sitemap_index.xml
Sitemap: https://come2indonesia.com/es/sitemap_index.xml

# ----------------------------------------------------
# LLM RULES (separate file — optional)
# ----------------------------------------------------
# Large AI training bots should read:
# https://come2indonesia.com/llms.txt
# AI policy
# See https://come2indonesia.com/llms-policy.txt