# See https://www.robotstxt.org/robotstxt.html for documentation
# Allow all crawlers.
#
# Content-Signal grants the three uses explicitly rather than leaving them
# unstated. Under contentsignals.org an omitted signal "neither grants nor
# restricts", and some crawlers treat that ambiguity conservatively — so silence
# is not the same as yes. We want the opposite of conservative: the compliance
# guide and the resource articles exist to be found, and AI crawlers already read
# this site around three times more than search engines do.
#
# search building a search index
# ai-input feeding content into a model to answer a live question
# ai-train training or fine-tuning models
# use=full summarise and reproduce, the most permissive of the use extension
#
# Do NOT turn on Cloudflare's managed robots.txt to "tidy" this. Its default is
# search=yes, ai-train=no, which would silently reverse the decision below.
User-agent: *
Content-Signal: search=yes, ai-input=yes, ai-train=yes, use=full
Allow: /
# Disallow authenticated/private areas
Disallow: /my-stuff/
Disallow: /settings/
Disallow: /onboarding/
Disallow: /api/
Disallow: /admin/
# Sitemap location
Sitemap: https://prism.guide/sitemap.xml
# Structured summary for AI agents and LLM crawlers
# https://llmstxt.org — AI crawlers read this site roughly 3x more than search
# engines do, and nothing pointed them at it before.
LLM-Content: https://prism.guide/llms.txt