# llms.txt for vr-avatar.net
# Purpose: Declare how this site’s content may be used by LLMs and AI crawlers.
# Notes:
# - This is a voluntary, emerging convention. Please also respect robots.txt and
#   known AI crawler user-agents’ directives (e.g., GPTBot, CCBot, anthropic-ai,
#   PerplexityBot, Google-Extended). When in doubt, obtain written permission.

version: 1.0
owner: vr-avatar.net
contact: webmaster@vr-avatar.net  # TODO: update to your preferred contact
last-updated: 2025-09-26
scope:
  - https://vr-avatar.net
  - https://www.vr-avatar.net

# Site content characteristics
# - This site hosts original avatar images and associated metadata/text.
# - Images and creative assets are high-value and copyrighted.
# - Be conservative: indexing/summarization of public text is OK; model training on
#   text or images is NOT allowed without explicit written permission.

usage-policy:
  training: disallow            # Disallow training on any text or images
  dataset-creation: disallow    # Disallow inclusion in datasets or embeddings
  fine-tuning: require-permission
  image-training: disallow      # No use of images for training or vision datasets
  summarization: allow          # Allow non-persistent summarization of public pages
  indexing: allow               # Allow search indexing of public pages
  research-noncommercial: require-permission
  commercial-use: disallow      # No commercial use without written permission
  caching: limited              # May cache for up to 7 days for compliance ops only
  attribution: required         # Link back to the page URL when quoting/snippetting

allowed-snippets:
  # Short quotes are allowed to aid search/discovery/UI previews.
  text-max-characters: 200
  images: no-full-resolution    # Thumbnails only; do not redistribute originals

rate-limits:
  polite-min-delay-ms: 1000     # At most ~1 request/sec
  concurrency: 1

compliance:
  robots: follow                # Must follow /robots.txt directives
  contact-for-exceptions: required
  audit-log: recommended        # Keep a record of access honoring these terms

notes:
  - To request exceptions (e.g., research, fine-tuning), email the contact above.
  - If you operate an AI crawler, identify via User-Agent and honor both robots.txt and this file.
  - This declaration does not grant any license to redistribute images or data.