# Signed-in and transactional surfaces only. Everything else is open to all crawlers.
#
# Content-Signal (contentsignals.org) states what the content may be used for once fetched, which
# Allow and Disallow cannot say: `search` is an index that links back, `ai-input` is an assistant
# reading a page to answer a question. Both are how a script page reaches an agent. `ai-train` is
# left unstated, which grants and restricts nothing.

User-agent: *
Content-Signal: search=yes, ai-input=yes
Disallow: /admin
Disallow: /api/
Disallow: /api-keys
Disallow: /browser/
Disallow: /dev/
Disallow: /devices
Disallow: /email-change/
Disallow: /invitations/
Disallow: /oauth/
Disallow: /projects
Disallow: /runs
Disallow: /setup
Disallow: /signin
Disallow: /subscribe/
Disallow: /connectors

Sitemap: https://reduck.ai/sitemap.xml