<?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>后端开发 on 智识脉搏</title>
    <link>https://cuigh.com/zh/tags/%E5%90%8E%E7%AB%AF%E5%BC%80%E5%8F%91/</link>
    <description>Recent content in 后端开发 on 智识脉搏</description>
    <generator>Hugo</generator>
    <language>zh-CN</language>
    <lastBuildDate>Mon, 25 May 2026 16:25:44 +0800</lastBuildDate>
    <atom:link href="https://cuigh.com/zh/tags/%E5%90%8E%E7%AB%AF%E5%BC%80%E5%8F%91/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>LLM Agent 写后端为什么容易翻车？问题不在不会写，而在忘了约束</title>
      <link>https://cuigh.com/zh/posts/why-llm-agents-fail-backend-constraints/</link>
      <pubDate>Mon, 25 May 2026 16:25:44 +0800</pubDate>
      <guid>https://cuigh.com/zh/posts/why-llm-agents-fail-backend-constraints/</guid>
      <description>从 Constraint Decay 这篇论文和今天的开发者社区讨论出发，聊聊为什么 LLM Agent 写后端代码时最危险的地方不是语法，而是权限、状态、数据一致性、架构约束和隐含需求被一点点忘掉。</description>
    </item>
  </channel>
</rss>
