# Thruact — https://thruact.com
#
# Everything public here is meant to be read: the capability catalogue, the
# measured reliability board, the vendor directory. Nothing below /v1/ is worth
# crawling — it is an API that answers JSON to authenticated callers — and the
# dashboard routes behind sign-in have nothing to show a crawler.

User-agent: *
Allow: /
Disallow: /v1/
Disallow: /mcp
Disallow: /oauth/
Disallow: /.well-known/
Disallow: /overview
Disallow: /runs
Disallow: /billing
Disallow: /secrets
Disallow: /settings
Disallow: /policy
Disallow: /team
Disallow: /signin
Disallow: /device
Disallow: /shared

# The answer-engine crawlers, named explicitly.
#
# Not because the wildcard above misses them — it does not — but because a
# blanket rule is easy to tighten later by accident, and the golden board is the
# one page here worth citing: measured success rates and latencies for real
# providers, which is a question these engines are asked and currently answer
# from marketing copy.
User-agent: GPTBot
Allow: /
User-agent: OAI-SearchBot
Allow: /
User-agent: ChatGPT-User
Allow: /
User-agent: ClaudeBot
Allow: /
User-agent: Claude-Web
Allow: /
User-agent: PerplexityBot
Allow: /
User-agent: Google-Extended
Allow: /
User-agent: Applebot-Extended
Allow: /
User-agent: CCBot
Allow: /

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