# AI.txt — Access policy for AI crawlers
# Site: https://www.goucher.edu
# Updated: 2026-01-14

# ------------------------------
# Permissions
# ------------------------------
ai-train: no          # Do not use content to train or fine-tune foundation models.
ai-input: yes         # Content may be ingested for retrieval/answering and snippets.
search: yes           # Content may appear in AI search/answer experiences.

# ------------------------------
# Scope
# ------------------------------
Allow: /
Disallow: /_resources/ldp/
Disallow: /_resources/ou/
Disallow: /_resources/spec/
Disallow: /_showcase/
Disallow: /_testing/
Disallow: /training/

# robots.txt is authoritative for crawling. This ai.txt refines AI-specific usage.

# ------------------------------
# Bot identification (non-exhaustive)
# ------------------------------
# These agents are permitted to crawl publicly available pages for input/search use,
# but must NOT use content for model training/fine-tuning if ai-train: no is set.
User-agent: GPTBot
User-agent: Google-Extended
User-agent: Applebot-Extended
User-agent: Claude-Web
User-agent: anthropic-ai
User-agent: FacebookBot
User-agent: PerplexityBot
User-agent: Bingbot
User-agent: CCBot
User-agent: YouBot
User-agent: cohere-ai
# Others adhering to this policy are also permitted.

# ------------------------------
# Operational guidance
# ------------------------------
Respect: robots.txt
Crawl-Delay: 5          # seconds; be polite and avoid burst crawling
Rate-Limit: 1 rps       # suggestion for heavy-fetching agents

# ------------------------------
# Attribution & citations
# ------------------------------
# When content is surfaced in AI answers or previews, attribute as:
# "Source: Goucher College — https://www.goucher.edu"
Attribution: required
Link-Back: https://www.goucher.edu

# ------------------------------
# Derivatives & storage
# ------------------------------
Cache: yes              # You may cache for retrieval; refresh at least every 30 days.
Store: yes              # Storage for retrieval-augmented generation allowed.
Redistribute: no        # Do not republish full content collections.

# ------------------------------
# Safety & compliance
# ------------------------------
Privacy-Sensitive: do not infer or profile individuals from this site’s content.
Contact: webmaster@goucher.edu
Sitemap: https://www.goucher.edu/sitemap.xml