# CrowdHum robots.txt
# Public marketing pages are crawlable. Application, presenter, voter, and
# auth-gated surfaces are noindex and excluded here to keep the crawl focused.

User-agent: *
Allow: /

# Application + real-time surfaces (also carry a noindex meta tag)
Disallow: /create
Disallow: /create2.html
Disallow: /dashboard
Disallow: /ops
Disallow: /p/
Disallow: /pm/
Disallow: /v/
Disallow: /r/
Disallow: /qa/
Disallow: /magic
Disallow: /mcp-authorize
Disallow: /poll-recap
Disallow: /connect

# POST-only API endpoints. Googlebot only issues GET, so crawling these
# returns 405/400 — a non-404 4xx, which Search Console reports as
# "Blocked due to other 4xx issue". Nothing here is ever indexable.
Disallow: /api/

# NOTE: /s/ (session share links) is deliberately NOT disallowed here.
# Those pages carry a noindex meta tag instead, because LinkedIn's crawler
# respects robots.txt and blocking it would break share-link unfurls —
# which is the entire purpose of that page.

Sitemap: https://crowdhum.com/sitemap.xml