Feeds and Sitemaps
Two machine-readable outputs ship with every build: an Atom feed at
/feed.xml and a sitemap at /sitemap.xml. Both are generated from the
published posts, so a draft or a future-dated post can never leak into
either.
The Atom feed
The feed lists the newest posts — capped by features.feed.limit, which
this site sets to 10 — with absolute URLs derived from site.url. Every
page also advertises it from the document head:
html
<link rel="alternate" type="application/atom+xml" href="/feed.xml">
Point a reader at the feed URL, or let them discover it automatically; both work because the link tag is in the template, not hand-written per page.
The sitemap
sitemap.xml enumerates every emitted URL: the home list and its
overflow pages, each post, each static page, and every tag archive. The
crawler that reads it maps each <loc> straight back to a file inside
dist/.
Why absolute URLs matter
Feeds and sitemaps are consumed off-site, so relative URLs are
meaningless there. Kiln rejects a build whose site.url is not an
absolute http(s) origin — the example site sets
https://example.com in kiln.config.ts.