<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Git Worktree on Text Matrix</title><link>https://txtmix.com/tags/git-worktree/</link><description>Recent content in Git Worktree on Text Matrix</description><generator>Hugo</generator><language>zh-cn</language><lastBuildDate>Sat, 23 May 2026 08:55:34 +0800</lastBuildDate><atom:link href="https://txtmix.com/tags/git-worktree/index.xml" rel="self" type="application/rss+xml"/><item><title>Superpowers：让 AI 编程助手真正学会软件工程</title><link>https://txtmix.com/posts/tech/superpowers-vscode-ai-pair-programming-guide/</link><pubDate>Sat, 02 May 2026 15:04:24 +0800</pubDate><guid>https://txtmix.com/posts/tech/superpowers-vscode-ai-pair-programming-guide/</guid><description>&lt;h2 id="前言ai-编程助手的工程化困境">前言：AI 编程助手的工程化困境&lt;/h2>
&lt;p>过去两年里，AI 编程助手从简单的补全工具演变成了能独立完成模块开发的「搭档」。以 Claude Code、Cursor、GitHub Copilot 为代表的工具已经能在足够简单的场景下替代初级工程师完成实际工作。然而，当项目复杂度上升、边界条件增多时，AI 的表现往往开始偏离预期：&lt;/p></description></item></channel></rss>