# Median - Know what's actually making you money
#
# A named user-agent group REPLACES the * group rather than inheriting it, so
# every group below repeats the Disallow lines. Without them, naming a crawler
# to welcome it also hands it /private/, which is where the hosted investor memo
# lives. Add the four Disallows to any group you add.

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

User-agent: *
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

# ---------------------------------------------------------------------------
# Search crawlers
# ---------------------------------------------------------------------------

User-agent: Googlebot
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

User-agent: Bingbot
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

# ---------------------------------------------------------------------------
# Answer engines and AI crawlers.
#
# Two distinct jobs, and the difference decides whether we get cited:
#   - training/indexing crawlers (GPTBot, ClaudeBot, Google-Extended, CCBot)
#   - live retrieval agents that fetch a page to answer a question right now
#     (OAI-SearchBot, ChatGPT-User, Perplexity-User, Claude-User)
# Blocking the second kind removes us from citations even when the first kind
# has already read the site. Both are allowed here on purpose.
# ---------------------------------------------------------------------------

# OpenAI: GPTBot trains, OAI-SearchBot powers ChatGPT Search citations,
# ChatGPT-User fetches on a user's behalf. OAI-SearchBot was missing.
User-agent: GPTBot
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

User-agent: OAI-SearchBot
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

User-agent: ChatGPT-User
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

# Anthropic
User-agent: ClaudeBot
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

User-agent: Claude-User
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

User-agent: Claude-Web
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

# Perplexity: PerplexityBot indexes, Perplexity-User fetches to answer.
User-agent: PerplexityBot
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

User-agent: Perplexity-User
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

# Google AI surfaces (AI Overviews, AI Mode, Gemini grounding)
User-agent: Google-Extended
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

# Apple Intelligence and Siri
User-agent: Applebot
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

User-agent: Applebot-Extended
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

# Meta AI
User-agent: meta-externalagent
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

# Microsoft Copilot
User-agent: MicrosoftPreview
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

# Common Crawl, which several models train from
User-agent: CCBot
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

# Mistral, Cohere, You.com, Amazon
User-agent: MistralAI-User
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

User-agent: cohere-ai
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

User-agent: YouBot
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

User-agent: Amazonbot
Allow: /
Disallow: /admin/
Disallow: /api/
Disallow: /_next/
Disallow: /private/

# Plain-language summary of what Median does, for agents that look for it:
# https://medianfi.com/llms.txt  and  https://medianfi.com/llms-full.txt