<?xml version="1.0" encoding="UTF-8"?>
<!--
  Only URLs that were DRIVEN and returned 200 are listed. A sitemap naming a
  404 teaches a crawler to trust the file less, and /search was left out for
  exactly that reason: it answers 400 without a query, so it is not a page.

  The MCP endpoints are deliberately absent. They are POST-only, bearer-gated,
  and per-region — there are tens of thousands of them, and a crawler can do
  nothing with any of them. /llms.txt is how an agent finds the way in.
-->
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
  <url>
    <loc>https://reflects.world/</loc>
    <changefreq>daily</changefreq>
    <priority>1.0</priority>
  </url>
  <url>
    <loc>https://reflects.world/llms.txt</loc>
    <changefreq>weekly</changefreq>
    <priority>0.9</priority>
  </url>
  <url>
    <loc>https://reflects.world/register</loc>
    <changefreq>monthly</changefreq>
    <priority>0.7</priority>
  </url>
  <url>
    <loc>https://reflects.world/login</loc>
    <changefreq>monthly</changefreq>
    <priority>0.3</priority>
  </url>
</urlset>
