<?xml version="1.0" encoding="UTF-8"?>
<!--
  The three pages this repository publishes. The equipment library has its own
  sitemap, generated by Hugo in the repository that builds it, and both are
  named in /robots.txt.

  Hand-written, because the site is hand-written: three files, no build step.
  tools/test_sitemap.py fails if this list and the files in site/ disagree,
  which is the failure a hand-written sitemap actually has.

  No lastmod. It is optional, and the honest value is not available here: a date
  typed by hand is a guess, and a date from file mtime is wrong in CI, where a
  fresh shallow checkout gives every file the same timestamp. A wrong lastmod is
  worse than none, because it teaches a crawler to ignore the field.

  URLs match the canonical link on each page and the site's own internal links:
  no trailing slash on /privacy or /support. Both spellings resolve, since
  router.js rewrites any final segment without a dot, so the canonical is what
  keeps them from being counted as two pages.

  Em dashes and commas above rather than a double hyphen: the two-hyphen
  sequence is illegal inside an XML comment, and a sitemap that will not parse
  is a sitemap a crawler discards in silence. This file was written twice.
-->
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
  <url><loc>https://anchorframe.app/</loc></url>
  <url><loc>https://anchorframe.app/privacy</loc></url>
  <url><loc>https://anchorframe.app/support</loc></url>
</urlset>
