# llm.txt for Any Studios
# Location: https://www.anystudios.com/llm.txt
# Purpose: Declare how AI crawlers and LLM providers may use this site

site: https://www.anystudios.com
sitemap: https://www.anystudios.com/sitemap.xml
owner: Any Studios, a Maison Lotan company
contact: hello@anystudios.com
effective-date: 2025-09-30
policy-version: 1.0

############################
# 1) Allowed usage
############################
# You may:
# - crawl public pages for search indexing and snippet generation
# - display short excerpts when a user asks about a specific page
# - follow standard robots.txt and rate limits

allow:
  - /$
  - /work/
  - /services/
  - /about/
  - /journal/
  - /privacy/
  - /terms/

############################
# 2) Prohibited usage
############################
# You may not:
# - use our content or media for training or fine-tuning models
# - build embeddings or long-term caches beyond the cache rules below
# - reproduce full pages or case studies
# - use logos, imagery, or brand assets for model output or generation
# - create derivative datasets or sell our data

disallow:
  - /admin/
  - /api/
  - /cms/
  - /cart/
  - /checkout/
  - /dashboard/
  - /login/
  - /account/
  - /*?*    # do not crawl querystring variants

media-disallow:
  - all-images
  - all-videos
  - all-logos
  - all-typefaces

training: disallow
derivative-datasets: disallow

############################
# 3) Cache and retention
############################
# Keep only ephemeral caches needed to serve a live user request
cache: transient
max-cache-ttl: 24h
embedding-retention: 0

############################
# 4) Attribution
############################
# When summarizing or quoting our pages:
# - cite the page title and canonical URL
# - limit excerpts to 160 words
attribution: required
max-excerpt: 160-words
canonical-required: true

############################
# 5) Rate limits
############################
# Default: 1 request per second, burst up to 5 for 10 seconds
rate-limit: 1 rps
burst: 5
burst-window: 10s

############################
# 6) Recognized AI user-agents
############################
# These agents must comply with robots.txt and llm.txt together
user-agent-allow:
  - GPTBot
  - ChatGPT-User
  - CCBot
  - ClaudeBot
  - Google-Extended
  - Google-LLM
  - PerplexityBot
  - anthropic-ai
  - Apple-AI
  - cohere-ai
  - FacebookExplore
  - Bytespider
  - Amazonbot
  - DuckAssistBot

# Explicitly denied agents
user-agent-disallow:
  - any-bot-violating-robots
  - unidentified-ai-crawler

############################
# 7) Enforcement
############################
# Noncompliance revokes any permission above
noncompliance: permissions-revoked
legal-reserved: true
blocked-use: model-training, embeddings, dataset-resale, brand-simulation

############################
# 8) Appeals and permissions
############################
# For licenses or research exceptions, email with:
# - purpose, duration, model name, storage plan, removal plan
appeals: hello@anystudios.com
licensing: hello@anystudios.com