<?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>Agent Harness on Ethan&#39;s Blog</title>
    <link>https://ethan6188.github.io/tags/agent-harness/</link>
    <description>Recent content in Agent Harness on Ethan&#39;s Blog</description>
    <generator>Hugo</generator>
    <language>zh-cn</language>
    <lastBuildDate>Tue, 07 Jul 2026 01:15:00 +0800</lastBuildDate>
    <atom:link href="https://ethan6188.github.io/tags/agent-harness/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Claude Code、OpenClaw、Hermes、DeepAgents：四种 Agent Harness 的设计对比</title>
      <link>https://ethan6188.github.io/posts/agent-harness-comparison/</link>
      <pubDate>Tue, 07 Jul 2026 01:15:00 +0800</pubDate>
      <guid>https://ethan6188.github.io/posts/agent-harness-comparison/</guid>
      <description>从核心循环、上下文工程、记忆、技能与自我改进、子代理、安全边界到模型耦合，逐个维度拆解四个代表性 Agent Harness 的设计取舍，以及工程上该怎么选。</description>
    </item>
  </channel>
</rss>
