# robots.txt - Connections (connections.icu)
# GENERATED by scripts/icons/generate-plane-robots.mjs - DO NOT EDIT BY HAND.
#   Edit infra/shared/seo/ai-crawler-roster.ts (the crawler DECLARATION every tier below
#   derives from) or this script's PLANES table, then run `bun run robots:sync`.
#
# Crawl policy is split into three intents. Rationale: docs/how-to/SEO_AND_OPEN_GRAPH.md §6.
#   1. AI ANSWER / SEARCH crawlers - they cite us and drive traffic, so allow. This is the
#      tier that decides whether Connections shows up when someone asks ChatGPT, Claude,
#      Gemini, Meta AI, Copilot, Perplexity, Alexa, or Siri a question.
#   2. AI TRAINING crawlers - allowed on this plane's own surface (being learned by models is
#      how Connections shows up in AI answers), but kept off the MEMBER-GENERATED public
#      paths listed below - the only thing a training block here actually protects is a
#      member's own public content, not our marketing pages (Jacob, 2026-09-13).
#   3. Everyone else - allow the public site, keep app/admin/oauth paths out.
#
# Every token below is DECLARED in infra/shared/seo/ai-crawler-roster.ts with its operator,
# what it does, and why we chose this disposition. A crawler that exists but is not declared
# there is a gating finding (ai-crawler-roster-declaration) - because an undeclared crawler
# is not blocked or allowed on purpose, it just falls through to tier 3.

# ── AI answer / search crawlers (cite + link back) ──
User-agent: OAI-SearchBot
Allow: /

User-agent: ChatGPT-User
Allow: /

User-agent: Claude-SearchBot
Allow: /

User-agent: Claude-User
Allow: /

User-agent: Google-Extended
Allow: /

User-agent: Google-CloudVertexBot
Allow: /

User-agent: meta-webindexer
Allow: /

User-agent: meta-externalfetcher
Allow: /

User-agent: PerplexityBot
Allow: /

User-agent: Perplexity-User
Allow: /

User-agent: Applebot
Allow: /

User-agent: Bingbot
Allow: /

User-agent: DuckAssistBot
Allow: /

User-agent: MistralAI-User
Allow: /

User-agent: YouBot
Allow: /

User-agent: Amazonbot
Allow: /

# ── AI training crawlers - allowed here, kept off member-generated content. Every operator
#    below also ships an answer-side token in the tier above, so allowing the rest of the
#    site costs us nothing in answer-engine pickup while it still lets these models learn
#    Connections well enough to recommend it. ──
User-agent: GPTBot
Allow: /
Disallow: /mc/
Disallow: /book/
Disallow: /f/
Disallow: /com/
Disallow: /e/

User-agent: ClaudeBot
Allow: /
Disallow: /mc/
Disallow: /book/
Disallow: /f/
Disallow: /com/
Disallow: /e/

User-agent: Claude-Web
Allow: /
Disallow: /mc/
Disallow: /book/
Disallow: /f/
Disallow: /com/
Disallow: /e/

User-agent: anthropic-ai
Allow: /
Disallow: /mc/
Disallow: /book/
Disallow: /f/
Disallow: /com/
Disallow: /e/

User-agent: meta-externalagent
Allow: /
Disallow: /mc/
Disallow: /book/
Disallow: /f/
Disallow: /com/
Disallow: /e/

User-agent: FacebookBot
Allow: /
Disallow: /mc/
Disallow: /book/
Disallow: /f/
Disallow: /com/
Disallow: /e/

User-agent: Applebot-Extended
Allow: /
Disallow: /mc/
Disallow: /book/
Disallow: /f/
Disallow: /com/
Disallow: /e/

User-agent: CCBot
Allow: /
Disallow: /mc/
Disallow: /book/
Disallow: /f/
Disallow: /com/
Disallow: /e/

User-agent: Bytespider
Allow: /
Disallow: /mc/
Disallow: /book/
Disallow: /f/
Disallow: /com/
Disallow: /e/

User-agent: Diffbot
Allow: /
Disallow: /mc/
Disallow: /book/
Disallow: /f/
Disallow: /com/
Disallow: /e/

User-agent: cohere-ai
Allow: /
Disallow: /mc/
Disallow: /book/
Disallow: /f/
Disallow: /com/
Disallow: /e/

User-agent: omgili
Allow: /
Disallow: /mc/
Disallow: /book/
Disallow: /f/
Disallow: /com/
Disallow: /e/

User-agent: omgilibot
Allow: /
Disallow: /mc/
Disallow: /book/
Disallow: /f/
Disallow: /com/
Disallow: /e/

User-agent: ImagesiftBot
Allow: /
Disallow: /mc/
Disallow: /book/
Disallow: /f/
Disallow: /com/
Disallow: /e/

User-agent: PanguBot
Allow: /
Disallow: /mc/
Disallow: /book/
Disallow: /f/
Disallow: /com/
Disallow: /e/

User-agent: Timpibot
Allow: /
Disallow: /mc/
Disallow: /book/
Disallow: /f/
Disallow: /com/
Disallow: /e/

# ── Everyone else ──
User-agent: *
Disallow: /admin
Disallow: /dev
Disallow: /oauth

Sitemap: https://connections.icu/sitemap.xml