# --------------------------------------------
# Default rules
# --------------------------------------------
User-agent: *

Sitemap: https://lvivholod.com/wp-sitemap.xml

# Keep admin private but let front-end AJAX work
Disallow: /wp-admin/
Allow: /wp-admin/admin-ajax.php

# Optional: keep REST endpoints out of the index (still accessible)
Disallow: /wp-json/
Disallow: /?rest_route=

# Avoid thin/duplicate pages in the index
Disallow: /trackback/
Disallow: */trackback/
Disallow: */comment-page-*
Disallow: /*?*replytocom=
Disallow: /search/
Disallow: /*?s=

# Blocking pagination and filters (causing "Scanned - not currently indexed")
Disallow: */page/*?
Disallow: *?query_type_*
Disallow: *?filter_*
Disallow: *?orderby=*

# Blocking "Add to cart" parameter (calls up "Alternative page...")
Disallow: /?add-to-cart=*
Disallow: *?add-to-cart=*

# Blocking junk links
Disallow: /feed/
Disallow: /trackback/
Disallow: /comments/
Disallow: /wp-content/plugins/
Disallow: /wp-content/cache/
Disallow: /wp-json/

# IMPORTANT: do NOT block /wp-includes/, /wp-content/plugins/, /wp-content/themes/
# Assets must be crawlable for proper rendering (Core Web Vitals/SEO).
Allow: /wp-content/uploads/

# Optional (USE META NOINDEX INSTEAD if you can):
# Disallow: /tag/
# Disallow: /author/

# Common technical files
Disallow: /readme.html
Disallow: /license.txt

# --------------------------------------------
# AI / data-mining crawlers (blocked)
# --------------------------------------------
User-agent: GPTBot
User-agent: ChatGPT-User
User-agent: OAI-SearchBot
User-agent: CCBot
User-agent: ClaudeBot
User-agent: Claude-Web
User-agent: Claude-User
User-agent: anthropic-ai
User-agent: PerplexityBot
User-agent: Perplexity-User
User-agent: Applebot-Extended
User-agent: Google-Extended
User-agent: Meta-ExternalAgent
User-agent: Meta-ExternalFetcher
User-agent: Bytespider
User-agent: Amazonbot
User-agent: cohere-ai
User-agent: cohere-training-data-crawler
User-agent: ImagesiftBot
User-agent: img2dataset
User-agent: omgilibot
User-agent: PetalBot
User-agent: TikTokSpider
User-agent: Timpibot
User-agent: YouBot
Disallow: /