# Honey Health LLMs.txt
# Goal: Maximize AI visibility & SEO, allow training for select reputable providers,
# and protect internal operational areas.

############################
# GLOBAL DEFAULTS
############################

# By default, allow AI systems to index our content for AI search / answers.
Content-Indexing: allow

# By default, do NOT allow model training, unless explicitly allowed below.
Model-Training: disallow

# Apply basic path protection for internal/operational areas.
# (Update these paths to match your actual internal URLs, if different.)
User-Agent: *
Disallow: /internal/
Disallow: /admin/
Disallow: /dashboard/
Disallow: /portal/
Disallow: /provider-portal/

############################
# ALLOW TRAINING FOR SELECT REPUTABLE PROVIDERS
############################

# OpenAI (ChatGPT, etc.)
User-Agent: OpenAI
Content-Indexing: allow
Model-Training: allow

# Google (Gemini, AI search)
User-Agent: Google-LLM
Content-Indexing: allow
Model-Training: allow

# Anthropic (Claude)
User-Agent: Anthropic-AI
Content-Indexing: allow
Model-Training: allow

# Microsoft / Bing AI
User-Agent: BingAI
Content-Indexing: allow
Model-Training: allow

# Apple Intelligence
User-Agent: Applebot-Extended
Content-Indexing: allow
Model-Training: allow

# Amazon (Bedrock / Titan)
User-Agent: Amazon-AI
Content-Indexing: allow
Model-Training: allow

# Meta AI
User-Agent: MetaAI
Content-Indexing: allow
Model-Training: allow

# Perplexity (you explicitly want visibility here)
User-Agent: PerplexityBot
Content-Indexing: allow
Model-Training: allow

############################
# FALLBACK FOR ALL OTHER AI CRAWLERS
############################

# For any other AI crawler not listed above:
# - They can index (good for visibility and SEO)
# - They cannot train on our content.
User-Agent: *
Content-Indexing: allow
Model-Training: disallow