# =============================================================================
# robots.txt — gregkilmer.com
# =============================================================================
# Strict anti-scraping policy.
#
#   - Google/Bing/etc. (legitimate search) are ALLOWED for SEO.
#   - /api/ and /contact are disallowed for everyone (the form + endpoints
#     should not be indexed or crawled).
#   - Known AI training bots are blocked by User-Agent below. Note that
#     robots.txt is voluntary; determined scrapers will ignore it. This is a
#     first layer (paired with noindex on /contact and email obfuscation).
# =============================================================================

# --- Default policy: allow crawling of public pages, block the API + contact
User-agent: *
Allow: /
Disallow: /api/
Disallow: /contact

# --- Block known AI training / scraping bots explicitly ---------------------
# OpenAI
User-agent: GPTBot
Disallow: /

User-agent: ChatGPT-User
Disallow: /

User-agent: OAI-SearchBot
Disallow: /

# Anthropic
User-agent: ClaudeBot
Disallow: /

User-agent: Claude-Web
Disallow: /

User-agent: anthropic-ai
Disallow: /

# Google's AI/Vertex crawlers (separate from regular Googlebot, which is allowed)
User-agent: Google-Extended
Disallow: /

# Common Crawl (feeds many AI training datasets)
User-agent: CCBot
Disallow: /

# ByteDance / TikTok
User-agent: Bytespider
Disallow: /

# Meta
User-agent: meta-externalagent
Disallow: /

User-agent: Meta-ExternalAgent
Disallow: /

User-agent: FacebookBot
Disallow: /

# Amazon / Alexa
User-agent: Amazonbot
Disallow: /

# Apple's AI/Applebot-Extended (regular Applebot remains allowed for Siri/Suggestions)
User-agent: Applebot-Extended
Disallow: /

# Perplexity
User-agent: PerplexityBot
Disallow: /

# Cohere
User-agent: cohere-ai
Disallow: /

# Diffbot
User-agent: Diffbot
Disallow: /

# Imagesift / OC-72 / other image scrapers
User-agent: ImagesiftBot
Disallow: /

User-agent: omgili
Disallow: /

User-agent: Omgilibot
Disallow: /

# Sitemap (uncomment once a sitemap.xml is generated)
# Sitemap: https://gregkilmer.com/sitemap.xml
