# officina.it — crawler policy
# Agentic resource catalog:     https://officina.it/.well-known/ai-catalog.json
# Machine-readable site index:  https://officina.it/llms.txt
# Public API contract:          https://officina.it/openapi.json
# Agent skill for the API:      https://officina.it/api/skill.md

Sitemap: https://officina.it/sitemap.xml

# Content Signals (https://contentsignals.org) — what may be done with what is
# crawled here. Being the answer an assistant gives is the point of publishing
# the guide, the tools and the API docs, so all three signals are yes.
User-agent: *
Content-Signal: search=yes, ai-input=yes, ai-train=yes
Disallow: /cdn-cgi/l/email-protection
Disallow: /lp/
Disallow: /dl/
Disallow: /app/

# Answer engines, assistant fetchers and the crawlers that feed them. Allowed
# by name rather than by silence: a policy that has to be inferred from the
# absence of a rule is one a scanner reads as "no policy", and an assistant
# deciding whether it may quote us should not have to infer.
#
# The training crawlers are here deliberately. A workshop asking an assistant
# which gestionale to use is the traffic this site exists to win, and that
# answer comes from what the model was trained on as much as from what it
# fetches live.
User-agent: GPTBot
User-agent: OAI-SearchBot
User-agent: ChatGPT-User
User-agent: ClaudeBot
User-agent: Claude-User
User-agent: Claude-SearchBot
User-agent: PerplexityBot
User-agent: Perplexity-User
User-agent: Google-Extended
User-agent: Applebot-Extended
User-agent: Meta-ExternalAgent
User-agent: Amazonbot
User-agent: MistralAI-User
User-agent: cohere-ai
User-agent: CCBot
Allow: /
Disallow: /cdn-cgi/l/email-protection
Disallow: /lp/
Disallow: /dl/
Disallow: /app/

# ByteSpider is the one exception: it feeds no answer engine our customers ask,
# and it is the crawler with a reputation for taking a site's bandwidth without
# sending anything back.
User-agent: ByteSpider
Disallow: /