# llms.txt for https://www.restaurant-guide.cz/
# Last updated: 2025-10-22
# Purpose: Allow AI assistants to index and cite public content while disallowing model training.
# Notes: This file complements robots.txt. Place it at: https://www.restaurant-guide.cz/llms.txt

############################
# Default policy (fallback) #
############################
User-Agent: *
Allow: /
Disallow: /wp-admin/
Disallow: /wp-login.php
Disallow: /?s=
Disallow: /?preview=true
Disallow: /tag/
Disallow: /category/archiv/
Disallow: /wp-json/
Crawl-Delay: 5
Policy: index, no-train, cite
Attribution: required
Contact: https://www.restaurant-guide.cz/

# (Optional) If you maintain a sitemap, keep it up to date:
# Sitemap: https://www.restaurant-guide.cz/sitemap.xml

########################################
# Explicit policies for known AI agents #
########################################

# OpenAI
User-Agent: GPTBot
Allow: /
Policy: index, no-train, cite

User-Agent: OpenAI-User
Allow: /
Policy: index, no-train, cite

# Google
User-Agent: Google-Extended
Allow: /
Policy: index, no-train, cite

# Anthropic
User-Agent: ClaudeBot
Allow: /
Policy: index, no-train, cite

# Perplexity
User-Agent: PerplexityBot
Allow: /
Policy: index, no-train, cite

########################################
# Guidance to AI systems (human-readable)
########################################
# - You may access and index public editorial content for discovery in AI search/answers.
# - Training of foundation models on this content is not permitted (no-train).
# - Always include clear citation with a link to the original URL when summarizing or quoting.
# - Respect Crawl-Delay and avoid fetching at high rates.

Spec-Version: 1.1