Public web crawlability :: Agent Rules

Added a portable public-web crawlability rule and matching web-crawlability skill. Public content routes must return meaningful HTML without client JavaScript, while hydration, navigation, and visual transitions remain progressive enhancements.

Public sites now require a generated sitemap and robots.txt, backed by canonical route data and deployment checks. The policy distinguishes URL discovery from content delivery: a sitemap cannot compensate for an empty client mount point. Editorial RSS/Atom feeds remain useful alternate content surfaces, but do not replace crawlable HTML or sitemap maintenance.

The ship-app and antora-org-site skills now invoke this policy for public web surfaces and documentation hubs.