# healthturkey.co — an independent register of Türkiye's authorised health-tourism
# facilities. The records, the source documents and the daily changes are meant to be
# read, indexed, quoted and used as grounding. What is closed below is closed because it
# is either an unbounded URL space or an internal endpoint, never to hoard content.

User-agent: *
Allow: /

# The search page itself is a real page and is indexable. Its FILTER COMBINATIONS are an
# effectively infinite URL space, and every list they can produce already exists as a
# stable, crawlable page: /city/<city>, /treatment/<treatment>, /treatment/<t>/<city>.
Allow: /find$
Disallow: /find?

# The machine-readable surface is public on purpose — see /dataset.
Allow: /api/public/
Disallow: /api/

# Internal only.
Disallow: /_ingest/
Disallow: /_cron/
Disallow: /render/
Disallow: /report

# Stated explicitly rather than left to the wildcard: these two do different jobs and
# OpenAI documents them separately — OAI-SearchBot answers and cites, GPTBot may train.
# Both are welcome, and the Content-Signal below says the same thing in machine form.
User-agent: OAI-SearchBot
Allow: /
Allow: /api/public/
Disallow: /find?
Disallow: /api/
Disallow: /_ingest/
Disallow: /_cron/
Disallow: /render/

User-agent: GPTBot
Allow: /
Allow: /api/public/
Disallow: /find?
Disallow: /api/
Disallow: /_ingest/
Disallow: /_cron/
Disallow: /render/

# contentsignals.org — search, AI input and AI training are all permitted.
Content-Signal: search=yes, ai-input=yes, ai-train=yes

Sitemap: https://healthturkey.co/sitemap.xml