<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Loop Engineering on Martin C. Richards</title>
    <link>//www.martinrichards.me/tags/loop-engineering/</link>
    <description>Recent content in Loop Engineering on Martin C. Richards</description>
    <generator>Hugo</generator>
    <language>en</language>
    <lastBuildDate>Fri, 24 Jul 2026 21:44:55 +0200</lastBuildDate>
    <atom:link href="//www.martinrichards.me/tags/loop-engineering/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>What&#39;s Inside the Harness</title>
      <link>//www.martinrichards.me/post/whats_inside_the_harness/</link>
      <pubDate>Fri, 24 Jul 2026 21:44:55 +0200</pubDate>
      <guid>//www.martinrichards.me/post/whats_inside_the_harness/</guid>
      <description>&lt;h1 id=&#34;whats-inside-the-harness&#34;&gt;What&amp;rsquo;s Inside the Harness&lt;/h1&gt;&#xA;&lt;p&gt;In March I published &lt;a href=&#34;https://www.martinrichards.me/post/building_your_own_agent_harness/&#34;&gt;Building Your Own Agent Harness&lt;/a&gt;. A harness, I said, is the skills, workflows, and methodology that teach your agent how you think and how you build. Since then, &lt;em&gt;harness&lt;/em&gt; has taken on a different meaning: Claude Code, Codex, Cursor, the whole runtime wrapped around a model. In my terms, that combination is the agent. You can call it the harness too, but when you&amp;rsquo;re building with an agent, it is the part you have the least control over. I&amp;rsquo;m not going to fight the internet over what to call a harness. I care about what you put inside it.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
