# Dubreads — https://www.dubreads.com
#
# AI / generative engines (ChatGPT/GPTBot, Claude/ClaudeBot, Perplexity,
# Google-Extended for AI Overviews, Applebot-Extended, …) are WELCOME to crawl
# and cite content pages. We keep a single "*" group as the source of truth so
# every crawler — search and generative — gets the same content access and the
# same sensible exclusions below. A curated map for LLMs lives at /llms.txt.

User-agent: *
Allow: /
Disallow: /api/
Disallow: /uzivatel/
Disallow: /wishlist
Disallow: /oblibene
Disallow: /suggest
Disallow: /suggest-authors
# Query-string URLs are all non-canonical: faceted filters (?mood[]=, ?category[]=,
# ?language=, ?provider=, ?tag=), search (?search=), pagination (?page=), the
# expensive AI search (?prompt=) and tracking params. NONE of them may be crawled —
# the combinatorial facet space is a crawler trap that exhausts the origin (worker
# pool → 504). All indexable content lives on clean paths and is listed in the
# sitemaps below. Landing pages (/, /ai-search, /list, /mood/…, /zanr/…) have no
# query string, so they stay crawlable.
Disallow: /*?

# Google's ads.txt/app-ads.txt validator fetches only /ads.txt. It reads only
# the group that names it, so grant it explicit access (it does not inherit "*").
User-agent: Google-adstxt
Allow: /app-ads.txt

# Curated site overview for large language models (GEO).
# (Not a robots directive; listed here for discoverability.)
# LLMs: https://www.dubreads.com/llms.txt

Sitemap: https://www.dubreads.com/en/sitemap.xml
Sitemap: https://www.dubreads.com/cs/sitemap.xml