User-agent: *
Crawl-delay: 10

Allow: /

Disallow: /wp-admin/
Allow: /wp-admin/admin-ajax.php

Disallow: /xmlrpc.php
Disallow: /readme.html
Disallow: /license.txt

Disallow: /?s=
Disallow: /search/

Disallow: /*?replytocom=
Disallow: /*?preview=
Disallow: /*?elementor-preview=

Disallow: /author/

Disallow: /wp-content/uploads/wp-import-export-lite/

Sitemap: https://stmargaretstl.org/sitemap.xml

Disallow: /events/*
Disallow: news/events/*

# Facebook excess crawl
User-agent: meta-externalagent
User-agent: meta-webindexer
User-agent: facebookexternalhit
Crawl-delay: 5

# Apple excess crawl
User-agent: Applebot
User-agent: iTMS
Crawl-delay: 5

# Block problem bots
User-agent: Baiduspider
User-agent: 360Spider
User-agent: Yisouspider
User-agent: PetalBot
User-agent: Amazonbot
Disallow: /

User-agent: Bytespider
Disallow: /

User-agent: Sogou web spider
Disallow: /

User-agent: Sogou inst spider
Disallow: /

# Block Apple GenAI
User-agent: Applebot-Extended
Disallow: /

# Block OpenAI
User-agent: GPTBot
Disallow: /
User-agent: ChatGPT-User
Disallow: /

# Block Google Bard AI
User-agent: Google-Extended
Disallow: /

# Block Common Crawl AI scraper
User-agent: CCBot
Disallow: /

# Block Perplexity AI
User-agent: PerplexityBot
Disallow: /

# Block other misc AI scrapers
User-agent: anthropic-ai
Disallow: /

User-agent: Omgilibot
Disallow: /

User-agent: Omgili
Disallow: /

User-agent: Diffbot
Disallow: /

User-agent: ImagesiftBot 
Disallow: /

User-agent: cohere-ai
Disallow: /

User-agent: claudebot
Disallow: /