Search, feeds and the sitemap
Three surfaces nobody maintains β all derived from the tree.
Search, feeds and the sitemap
Press / β or βK β anywhere on this site. The search overlay reads the same tree the navigation does, so a page is findable the instant it exists. Nothing indexes, nothing rebuilds, and there is no step anybody can forget.
Hidden is not private
meta.hidden means "not in the navigation". Hidden pages stay searchable on purpose β it is a tidiness setting, not a permission. Owner-only pages are a separate predicate, and the Settings document is the example: it is a real page, it never appears in navigation or search, and asking the server for it as a visitor returns a blank.
Feeds
/feed.xml and /feed.json carry the pages whose TYPE opts into syndication β Docs entries and Live posts here, not project cards. That is editorial intent, not layout: deriving it from how a page renders once put an entire portfolio into the RSS feed, because an icon grid happens to be full-width.
Sitemap
/sitemap.xml lists every reachable page. Redirects are excluded by type β a page that bounces wastes crawl budget β and so is anything owner-only. A renamed page keeps answering on its old URL with a real 301, so nothing you have linked goes stale.