User-agent: *
Allow: /
# Cloudflare's email-obfuscation landing page. Scrape Shield rewrites mailto:
# links to this URL; with the feature off it 404s, and Google has been holding
# one since a 2026-08-07 crawl. Scoped to the one endpoint on purpose: blocking
# all of /cdn-cgi/ would also hide email-decode.min.js, which Googlebot needs to
# render the page if Scrape Shield is ever switched back on.
Disallow: /cdn-cgi/l/email-protection
Sitemap: https://novamsl.com/sitemap.xml

# Text and data mining rights are RESERVED. See /.well-known/tdmrep.json for the
# machine-readable declaration and /terms for the policy.
#
# The distinction below is the whole point: AI providers run separate crawlers
# for search and for training. Crawlers that cite and link back are welcome.
# Crawlers that absorb the content into a model are not.

# ---- Allowed: these cite and link back ----

User-agent: OAI-SearchBot
Allow: /

User-agent: ChatGPT-User
Allow: /

User-agent: Claude-SearchBot
Allow: /

User-agent: Claude-User
Allow: /

User-agent: PerplexityBot
Allow: /

User-agent: Perplexity-User
Allow: /

# ---- Disallowed: model training ----

User-agent: GPTBot
Disallow: /

User-agent: ClaudeBot
Disallow: /

# Google-Extended has no search/training split: it governs Gemini grounding AND
# training together. Disallowing it reserves training at the cost of Gemini
# citation. Allow it if that reach matters more than the reservation.
User-agent: Google-Extended
Disallow: /

# ---- Disallowed: bulk dataset crawlers, no citation and no referral ----

User-agent: CCBot
Disallow: /

User-agent: Amazonbot
Disallow: /