# Traditional search engines
User-agent: *
Allow: /
Disallow: /profile
Disallow: /wallet
Disallow: /history
Disallow: /auth/
Disallow: /dev/
Disallow: /banned
Sitemap: https://kaban.gg/sitemap.xml

# Yandex — same policy + UTM param folding (prevents duplicate-URL indexing).
# Critical for the RU/CIS market: without Clean-param, Yandex treats
# kaban.gg/?utm_source=telegram, ?utm_source=vk, etc. as distinct pages and
# splits ranking weight across the copies instead of consolidating on one URL.
User-agent: Yandex
Allow: /
Disallow: /profile
Disallow: /wallet
Disallow: /history
Disallow: /auth/
Disallow: /dev/
Disallow: /banned
Clean-param: utm_source&utm_medium&utm_campaign&utm_content&utm_term&gclid&yclid /
Sitemap: https://kaban.gg/sitemap.xml

# AI crawlers — Tier 1 (AI search/answer engines) + Tier 2 (AI features) + training.
# Explicitly allowed for AI-search visibility; private pages still blocked.
User-agent: GPTBot
User-agent: OAI-SearchBot
User-agent: ChatGPT-User
User-agent: ClaudeBot
User-agent: PerplexityBot
User-agent: Google-Extended
User-agent: GoogleOther
User-agent: Applebot-Extended
User-agent: Amazonbot
User-agent: FacebookBot
User-agent: CCBot
User-agent: anthropic-ai
User-agent: cohere-ai
Allow: /
Disallow: /profile
Disallow: /wallet
Disallow: /history
Disallow: /auth/
Disallow: /dev/
Disallow: /banned
Sitemap: https://kaban.gg/sitemap.xml

# Aggressive scraper — blocked (high crawl load, no AI-search value)
User-agent: Bytespider
Disallow: /

# NOTE(seo): /crate-pvp/create, /crate-pvp/:uid (battle rooms) and
# /giveaway/create are intentionally NOT Disallowed here. They're either
# auth-gated (create pages) or ephemeral (battle rooms) and get
# <meta name="robots" content="noindex, nofollow"> at runtime instead (see
# app/router/index.ts meta.noindex + src/main.ts). A Disallow would block
# Googlebot from ever crawling the page and seeing that noindex tag, which
# risks Google indexing the bare URL with no snippet if it's ever linked
# externally — noindex-while-crawlable is the safer signal for pages that
# are reachable but not meant to rank.