# robots.txt for bloome.im — served by the bloome-seo worker.
#
# Policy: the bloome-seo surface is open to every crawler, including AI answer
# engines (GPTBot, ClaudeBot, Google-Extended, PerplexityBot, …). The SEO/GEO
# pages, sitemap, robots.txt, llms.txt, llms-full.txt, and SEO assets are
# crawlable by default.
#
# The app / account / public-dynamic surfaces stay sealed here. They are owned
# by the Reson app, not the SEO app, and either return the SPA shell, require
# auth, or are dynamic invite/share pages that should not become SERP entries.
# Reson should also return X-Robots-Tag: noindex, nofollow for public dynamic
# pages such as /join/:code and /s/:code; robots.txt keeps crawlers off them
# after any urgent GSC removal / deindexing cleanup is complete.

User-agent: *
Allow: /

# --- SEO/GEO surfaces owned by bloome-seo: explicitly crawlable ---
Allow: /features/
Allow: /guides/
Allow: /alternatives/
Allow: /showcase/
Allow: /zh-CN/
Allow: /ja/
Allow: /es/
Allow: /pt-BR/
Allow: /seo-assets/
Allow: /sitemap.xml
Allow: /robots.txt
Allow: /llms.txt
Allow: /llms-full.txt

# --- Public entry pages: intentionally crawlable (NOT disallowed) ---
#   /login, /register — public sign-in / sign-up entry pages.
#   /explore           — public agent gallery (individual /agents/:id stay sealed).

# --- App / account surface (SPA routes + API): sealed ---
Disallow: /api/
Disallow: /admin
Disallow: /app
Disallow: /forgot-password
Disallow: /reset-password
Disallow: /verify-email-notice
Disallow: /activate-invite
Disallow: /welcome
Disallow: /onboarding/
Disallow: /inbox
Disallow: /credits/
Disallow: /guest-chat/
Disallow: /image-standalone/
Disallow: /widget-standalone/
Disallow: /embed/

# --- Public dynamic pages owned by Reson: sealed after GSC cleanup ---
Disallow: /agent/
Disallow: /agents/
Disallow: /join/
Disallow: /s/
Disallow: /arena
Disallow: /chat
Disallow: /contacts
Disallow: /resources
Disallow: /me
Disallow: /create
Disallow: /new

Sitemap: https://bloome.im/sitemap.xml