# https://www.robotstxt.org/robotstxt.html
User-agent: *
# Content signals (https://contentsignals.org): AI and other automated clients
# MAY index/search and reference this site, but should NOT use it as training
# data. This is a preference, not a block — crawlers (ClaudeBot, GPTBot, etc.)
# are allowed so the site stays legible to AI answer engines and agents.
# NOTE: Cloudflare's managed robots.txt used to PREPEND hard `Disallow: /`
# blocks for AI bots at the edge (which fails the a14y allows-ai-bots check).
# This line is the origin's own, softer stance — keep Cloudflare's managed
# robots.txt / AI-bot block DISABLED so this file is served unmodified.
Content-Signal: search=yes,ai-train=no,use=reference
Allow: /
# Stakeholder-review routes — not for indexing
Disallow: /preview/
# Authed app routes — never index
Disallow: /worlds/
Disallow: /profile
Disallow: /first-run
Disallow: /billing/
# Raw prerendered snapshots are served at canonical routes (e.g. /pricing),
# not these internal paths. They carry canonical tags, but disallow anyway
# to keep the duplicate copies out of the index.
Disallow: /prerendered/
Disallow: /static/prerendered/

Sitemap: https://storywright.io/sitemap.xml

# LLM-readable site summary (https://llmstxt.org)
# llms.txt: https://storywright.io/llms.txt
# llms-full.txt: https://storywright.io/llms-full.txt