# OpenCap robots.txt
# Public: the marketing site, share pages, and docs are crawlable.
# Private: signed-in app surfaces and CLI auth ceremony are not.

User-agent: *
Allow: /
Allow: /docs
Allow: /docs/

Disallow: /app
Disallow: /settings
Disallow: /settings/
Disallow: /edit/
Disallow: /cli-auth
Disallow: /login

# Hosted share recordings are intentionally public. They're the share
# surface. Crawling them gets the page metadata but the video itself is
# behind a short-TTL signed URL, so re-indexing is fine.
Allow: /r/

# Explicit allow for AI / LLM crawlers. We expose llms.txt for richer
# generative-engine context. Each platform has its own bot — listing them
# explicitly makes citation policy auditable from one place.
User-agent: GPTBot
Allow: /

# ChatGPT-User is OpenAI's live-browsing fetcher (vs GPTBot which is the
# training crawler). Required separately for ChatGPT to cite us in answers.
User-agent: ChatGPT-User
Allow: /

User-agent: ClaudeBot
Allow: /

User-agent: anthropic-ai
Allow: /

User-agent: PerplexityBot
Allow: /

# Google-Extended controls Gemini / AI Overviews visibility independently
# from the regular Googlebot crawl.
User-agent: Google-Extended
Allow: /

# Bingbot also powers Copilot's source set.
User-agent: bingbot
Allow: /

User-agent: CCBot
Allow: /

Sitemap: https://opencap.dev/sitemap.xml