# llms.txt for mrpatch.us
# Policy: Allow AI search previews/summaries. Disallow model training & bulk dataset use.

Owner: MrPatch USA LLC
Contact: info@mrpatch.us
Sitemap: https://www.mrpatch.us/sitemap.xml
Last-Updated: 2025-09-25

# Global rules (interpreted by cooperating LLM crawlers)
User-Agent: *
Crawl-Delay: 5
Allow: /
Disallow: /cart/
Disallow: /checkout/
Disallow: /account/
Disallow: /admin/
Disallow: /ajax/
Disallow: /api/private/
Purpose-Search: allow          # permit retrieval for answer snippets / previews
Purpose-Summarization: allow   # allow short summaries/QA about public pages
Purpose-Training: disallow     # do NOT use content to train/fine-tune models
Purpose-Cache: allow
Purpose-Commercial-Reuse: disallow
Attribution: required          # must cite "MrPatch" with a link to source URL

# Explicit opt-out for known training/bulk crawlers
User-Agent: GPTBot
Disallow: /

User-Agent: ChatGPT-User
Allow: /                        # browsing/answers allowed

User-Agent: Google-Extended
Disallow: /                     # opt-out of Google AI model training

User-Agent: ClaudeBot
Disallow: /

User-Agent: CCBot
Disallow: /                     # Common Crawl (used for AI corpora)

User-Agent: PerplexityBot
Disallow: /                     # block bulk crawling/training

User-Agent: Applebot-Extended
Disallow: /                     # Apple model training opt-out

User-Agent: Amazonbot
Disallow: /                     # Amazon AI data collection

User-Agent: Bytespider
Disallow: /                     # ByteDance data collection

User-Agent: Meta-ExternalAgent
Disallow: /                     # Meta AI data collection

# Optional: media tightening (images/PDFs)
# Many providers also respect "noimageai" via X-Robots-Tag headers (see below).