<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>Mcp on lgcode</title>
    <link>https://blog.lgcode.me/tags/mcp/</link>
    <description>Recent content in Mcp on lgcode</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Wed, 05 Aug 2026 10:00:00 -0300</lastBuildDate>
    <atom:link href="https://blog.lgcode.me/tags/mcp/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Publishing an MCP Server to the Registry: What the Schema Doesn&#39;t Tell You</title>
      <link>https://blog.lgcode.me/posts/publishing-an-mcp-server-to-the-registry/</link>
      <pubDate>Wed, 05 Aug 2026 10:00:00 -0300</pubDate>
      <guid>https://blog.lgcode.me/posts/publishing-an-mcp-server-to-the-registry/</guid>
      <description>The manifest validates, the publisher accepts it, the entry goes live — and the install is still broken. The convention that isn&amp;#39;t in the schema, and why it cost me a second PyPI package.</description>
    </item>
    <item>
      <title>What Dogfooding an MCP Server Taught Me About Tool Output</title>
      <link>https://blog.lgcode.me/posts/mcp-tool-output-design/</link>
      <pubDate>Mon, 03 Aug 2026 10:00:00 -0300</pubDate>
      <guid>https://blog.lgcode.me/posts/mcp-tool-output-design/</guid>
      <description>A filter that discarded the answer, a retry that couldn&amp;#39;t work, a partial result posing as a total, and cached prices with no freshness signal. All four found in one live session, all four the same class of mistake.</description>
    </item>
  </channel>
</rss>
