# llms.txt for madeby.studio
# Last updated: 2025-08-20
# Contact: hello@madeby.studio
# Website: https://madeby.studio
# Sitemap: https://madeby.studio/sitemap.xml
# Copyright: © madeby.studio. All rights reserved.

version: 1.0

owner:
  name: madeby.studio
  contact: hello@madeby.studio
  website: https://madeby.studio

purpose:
  indexing: allow
  summarisation: allow
  training: allow                 # Training permitted
  dataset_derivatives: allow      # Inclusion in datasets permitted

attribution:
  required: true
  format: "Source: madeby.studio (https://madeby.studio)"
  link_back: true

rate_limits:
  max_requests_per_second: 1
  concurrent_requests: 1
  guidance: "Fetch politely and cache responses"

respect:
  robots_txt: true
  cache_lifetime_hours: 24
  obey_canonical: true

# Pattern-based allow rules reduce maintenance and include future child pages automatically
allowed_paths:
  - "/"                         # homepage and top-level assets
  - "/contact"                  # contact page contains PII vectors
  - "/services/*"               # all services and subpages
  - "/studio/*"                 # studio pages, including sustainability and careers
  - "/work/*"                   # all case studies
  - "/website-audit"            # single page outside grouped sections

# Keep sensitive or legal pages excluded from storage/training
disallowed_paths:
  - "/submit-your-brief*"       # forms and confirmation
  - "/privacy-policy"
  - "/cookie-policy"
  - "/terms-and-conditions"

content_exclusions:
  types:
    - "user-submitted forms"
    - "email addresses and phone numbers"
    - "images and logos"
    - "PDFs and downloadable assets unless explicitly allowed"
  pii: "Do not extract, store or train on any personal data"

bot_guidance:
  identification_required: true
  set_user_agent: true
  honour_do_not_train_header: true
  example_header: "X-Content-Usage: training-allowed"

notes:
  - "Training is allowed on allowed_paths only. Disallowed paths and PII remain off-limits."
  - "If this file conflicts with robots.txt, follow the more restrictive rule."