Search engines find a site's pages faster, and notice changes sooner, when the site lists them in a sitemap. The framework generates one at /sitemap.xml from the pages that exist, so it is never out of date and there is nothing to maintain. robots.txt points to it.
What It Lists
Every live page and post, with the file's last-modified time as <lastmod> — the moment the content actually changed, rather than the editorial pagedate. The home page is listed as the site root.
Left out: the 404 page, navigation, the footer columns, drafts (which are .html.draft files and so do not exist as pages), and any page that says it does not want to be listed:
<!-- sitemap: no -->
Add that line to a page's metadata block to keep it out — a thank-you page after a form, say, or a landing page that should only be reached from a link.
Telling Search Engines
robots.txt already carries Sitemap: /sitemap.xml, which is how most crawlers find it. You can also submit https://yoursite.com/sitemap.xml once in Google Search Console and Bing Webmaster Tools, which then shows you what was indexed and when.
How It Works
A rewrite rule in the site's .htaccess maps /sitemap.xml to index.php?sitemap, which hands over to sitemap_generator.php before the cache layer loads. The page list comes from swf_sitemap_pages() in required/feeds.php, the same file that knows which posts and categories the RSS feeds contain.
If you upgraded from a version without a sitemap, make sure your .htaccess has the rule:
RewriteRule ^sitemap\.xml$ index.php?sitemap=1 [L]