# Content signals — declarative reservation of rights, replacing the equivalent
# line Cloudflare's managed robots.txt used to inject (that managed config also
# sent Disallow: / to ClaudeBot and GPTBot, which are the same user-agents Claude
# and ChatGPT use to RETRIEVE and cite pages — so it made us uncitable. It is now
# set to "Disable robots.txt configuration" and this file is the only source.)
#
#   search=yes     index us and return us in results
#   ai-input=yes   fetch us to ground/answer a live user question, with citation
#   ai-train=no    do not use this content to train or fine-tune models
#   use=reference  quote and link, don't reproduce wholesale
#
# ANY RESTRICTIONS EXPRESSED VIA CONTENT SIGNALS ARE EXPRESS RESERVATIONS OF
# RIGHTS UNDER ARTICLE 4 OF EU DIRECTIVE 2019/790 ON COPYRIGHT IN THE DSM.

User-agent: *
Content-Signal: search=yes,ai-input=yes,ai-train=no,use=reference
Allow: /

# Bulk training scrapers — no attribution, no referral traffic, no citation
# upside. Blocked at the Cloudflare WAF too; stated here for well-behaved bots.
User-agent: CCBot
Disallow: /

User-agent: Bytespider
Disallow: /

User-agent: Amazonbot
Disallow: /

User-agent: meta-externalagent
Disallow: /

User-agent: Applebot-Extended
Disallow: /

# App / auth / utility surfaces — not for the index
Disallow: /auth
Disallow: /checkout
Disallow: /onboarding
Disallow: /settings
Disallow: /home
Disallow: /meet
Disallow: /create
Disallow: /store
Disallow: /sessions
Disallow: /meetings
Disallow: /scenarios
Disallow: /personas
Disallow: /billing
Disallow: /embed/
Disallow: /shares/
Disallow: /vc/
Disallow: /end-call
Disallow: /incomplete-signup
Disallow: /consent-cancel
Disallow: /successful-login

# Internal / test pages
Disallow: /pipecat-meeting-test
Disallow: /notion-editor-demo
Disallow: /widget
Disallow: /categories
Disallow: /use-cases-kanban

# Sitemap + feeds (feeds are polled more often than the sitemap)
Sitemap: https://waterr.ai/sitemap.xml