Namespace extensions: content:encoded, dc: and Media RSS

RSS 2.0 is small by design and extended by namespaces. Declaring them correctly is what makes full content, metadata and images work.

The extensions you will actually use

<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0"
     xmlns:content="http://purl.org/rss/1.0/modules/content/"
     xmlns:dc="http://purl.org/dc/elements/1.1/"
     xmlns:media="http://search.yahoo.com/mrss/"
     xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Release Notes</title>
    <link>https://example.com/releases</link>
    <description>Product changes, weekly.</description>
    <language>en-gb</language>
    <atom:link href="https://example.com/feed.xml" rel="self" type="application/rss+xml"/>
    <lastBuildDate>Thu, 18 Sep 2026 09:00:00 +0100</lastBuildDate>

    <item>
      <title>Platform 3.2</title>
      <link>https://example.com/releases/2026-09-platform-3-2</link>
      <guid isPermaLink="false">release-2026-09-platform-3-2</guid>
      <pubDate>Thu, 18 Sep 2026 09:00:00 +0100</pubDate>
      <dc:creator>Platform Team</dc:creator>
      <!-- summary for readers that show a teaser -->
      <description>Streaming transforms and faster cold start.</description>
      <!-- full HTML for readers that show the whole article -->
      <content:encoded><![CDATA[<p>Streaming transforms and faster cold start.</p>
<img src="https://example.com/img/3-2.png" alt="Dashboard"/>]]></content:encoded>
      <media:thumbnail url="https://example.com/img/3-2-300.png" width="300" height="169"/>
      <media:content url="https://example.com/img/3-2.png" medium="image" width="1200" height="675"/>
    </item>
  </channel>
</rss>

Pitfalls with extensions

• A prefix that is used but not declared on the rss element makes the document not well formed. Feed validators report this as a namespace error, and readers drop the item.
• The prefix itself is arbitrary — only the URI matters. Copying a feed and renaming the prefix is safe; changing the URI is not.
• Inside content:encoded, use CDATA or escape every <. Mixing both is where most broken feeds come from.
• description and content:encoded serve different readers. Fill both rather than duplicating megabytes into each.
• media:content and enclosure are different mechanisms: an enclosure is a downloadable asset, media content is metadata about a resource.
• Extension order inside item is flexible, but keeping a consistent order makes diffs readable when you debug a feed.

import html

def item_xml(it) -> str:
    """Escape for text content; use CDATA only where the schema expects markup."""
    def esc(s: str) -> str:
        return html.escape(s, quote=True)

    parts = [
        "  <item>",
        "    <title>" + esc(it.title) + "</title>",
        "    <link>" + esc(it.url) + "</link>",
        '    <guid isPermaLink="false">' + esc(it.id) + "</guid>",
        "    <pubDate>" + it.published.strftime("%a, %d %b %Y %H:%M:%S %z") + "</pubDate>",
        "    <description>" + esc(it.summary) + "</description>",
        # Nothing may be literally ]]> inside a CDATA section, so split it
        "    <content:encoded><![CDATA["
        + it.html.replace("]]>", "]]]]><![CDATA[>")
        + "]]></content:encoded>",
        "  </item>",
    ]
    return "\n".join(parts)

FAQ

Related

The RSS 2.0 feed format Full-text versus summary feeds

Last refreshed 2026-09-18.