# Pressefeuer.at - Robots.txt
# Optimiert für Google News und Suchmaschinen

# Allow all search engines to crawl public content.
# Content Signals: Suche und KI-Antworten ja, Training auf unseren Inhalten nein.
User-agent: *
Content-Signal: search=yes, ai-input=yes, ai-train=no
Allow: /
Allow: /news/
Allow: /newsroom/
Allow: /features
Allow: /pricing
Allow: /api
Allow: /compare
Allow: /compare/
Allow: /docs
Allow: /docs/
Allow: /glossar
Allow: /about
Allow: /contact
Allow: /tools
Allow: /llms.txt
Allow: /llms-full.txt
Allow: /agents.txt
Allow: /openapi.json
Allow: /auth.md
Allow: /pricing.md
Allow: /index.md
Allow: /.well-known/
Allow: /impressum
Allow: /datenschutz
Allow: /agb
# Block private/admin areas
Disallow: /dashboard/
Disallow: /api/
Disallow: /login
Disallow: /login/verify
Disallow: /settings
Disallow: /submit
Disallow: /approve/

# Allow tRPC API for prerendering
Allow: /api/trpc/

# --- KI-Crawler: gestufte Freigabe --------------------------------------
# Antwortmaschinen (Suche/Assistenz) sind willkommen; die Archive bleiben zu,
# weil jeder Cache-Miss dort einen vollen Re-Render plus Datenbank-Reads
# auslöst. Reine Trainings-Crawler ohne Gegenwert bleiben komplett draußen.
# Am CDN gilt dieselbe Aufteilung — robots.txt beschreibt sie nur.

User-agent: GPTBot
User-agent: OAI-SearchBot
User-agent: ChatGPT-User
User-agent: ClaudeBot
User-agent: Claude-User
User-agent: Claude-SearchBot
User-agent: PerplexityBot
User-agent: Google-Extended
User-agent: Applebot-Extended
Allow: /
Allow: /llms.txt
Allow: /llms-full.txt
Allow: /agents.txt
Allow: /openapi.json
Allow: /auth.md
Allow: /pricing.md
Allow: /index.md
Allow: /.well-known/
Disallow: /news/
Disallow: /newsroom/
Disallow: /schlagwort/
Disallow: /themen
Disallow: /dashboard/
Disallow: /api/
Disallow: /login
Disallow: /settings
Disallow: /submit
Disallow: /approve/

# Reine Trainings-Crawler: kein Zugriff.
User-agent: CCBot
User-agent: Bytespider
User-agent: meta-externalagent
User-agent: Amazonbot
Disallow: /

# Google News specific
User-agent: Googlebot-News
Allow: /news/
Allow: /newsroom/
Disallow: /dashboard/
Disallow: /api/

# Machine-readable entry points for agents:
#   /llms.txt      - short site summary and canonical page list
#   /llms-full.txt - extended LLM context
#   /agents.txt    - product guardrails for agents
#   /openapi.json  - OpenAPI 3.1 spec of the external HTTP surface
# Note: AI user agents are additionally filtered at the CDN edge; this file
# describes crawl policy, not the current edge rules.

# Sitemaps
Sitemap: https://pressefeuer.at/sitemap.xml
Sitemap: https://pressefeuer.at/sitemap-news.xml

# NLWeb Schema Feeds — strukturierte Datenfeeds
Schemamap: https://pressefeuer.at/schemamap.xml