# ==============================================================
# Resch Verlag – llms.txt
# Controls access for LLM crawlers (Google-Extended, Claude,
# GPTBot, etc.).  Edit the URLs and rules to fit your own site.
# ==============================================================

# --------------------------------------------------------------
# 1️⃣  General policy – allow everything that is public
# --------------------------------------------------------------
# By default we allow all LLMs to read & index **public** pages.
# (You can change this to `Disallow` any section you never want
#  the model to see.)

# Google‑Extended (the “search‑engine‑only” crawler)
User-agent: Google-Extended
Allow: /
# Optional: limit to a specific sub‑directory, e.g.
# Allow: /blog/
# Allow: /products/
# Disallow: /private/

# Open‑AI (GPT‑Bot) – allow all public URLs
User-agent: GPTBot
Allow: /

# Anthropic (Claude) – allow all public URLs
User-agent: ClaudeBot
Allow: /

# Meta‑Massachusetts (LLaMA) – allow all public URLs
User-agent: Meta‑Massachusetts
Allow: /

# --------------------------------------------------------------
# 2️⃣  Sections you want to **exclude**
# --------------------------------------------------------------
# If you have admin panels, member‑only areas, or test pages that
# you never want a model to ingest, block them here.

# Example: block the whole /admin/ folder
Disallow: /admin/
Disallow: /cgi-bin/
Disallow: /private/

# Block the /wp‑login.php and /wp‑admin/ if you run WordPress
# Disallow: /wp-login.php
# Disallow: /wp-admin/

# Block any PDF or DOCX you don’t want crawled (optional)
Disallow: /downloads/*.pdf
Disallow: /downloads/*.docx

# --------------------------------------------------------------
# 3️⃣  Crawl‑rate limits (optional but recommended)
# --------------------------------------------------------------
# Some models respect the `Crawl‑rate` directive.  Keep it modest
# so you don’t overload your server.

# 1 request per second for Google‑Extended is a safe default
Crawl‑rate: Google-Extended 1

# 1 request per second for GPTBot
Crawl‑rate: GPTBot 1

# 0.5 requests per second for ClaudeBot (a bit gentler)
Crawl‑rate: ClaudeBot 0.5

# --------------------------------------------------------------
# 4️⃣  Sitemap / index reference (helps LLMs discover URLs)
# --------------------------------------------------------------
# If you maintain an XML sitemap, point LLMs to it.
Sitemap: https://www.resch-verlag.com/sitemap.xml

# --------------------------------------------------------------
# 5️⃣  End of file
# --------------------------------------------------------------
# Anything not mentioned above follows the default robot policy.
# By keeping this file tiny and explicit you protect your
# proprietary content while still giving LLMs a clear map of what
# is safe to consume.