<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>系统安全 on ByF (BaiFan) · 每日深度思考</title>
    <link>https://blog.baifan.site/tags/%E7%B3%BB%E7%BB%9F%E5%AE%89%E5%85%A8/</link>
    <description>Recent content in 系统安全 on ByF (BaiFan) · 每日深度思考</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>zh-cn</language>
    <managingEditor>blog@baifan.site (ByF (BaiFan))</managingEditor>
    <webMaster>blog@baifan.site (ByF (BaiFan))</webMaster>
    <copyright>CC BY-NC 4.0</copyright>
    <lastBuildDate>Wed, 20 May 2026 09:00:00 +0800</lastBuildDate><atom:link href="https://blog.baifan.site/tags/%E7%B3%BB%E7%BB%9F%E5%AE%89%E5%85%A8/index.xml" rel="self" type="application/rss+xml" />
    
    <item>
      <title>权限系统与真实安全边界：沙箱、系统调用与审批逃逸</title>
      <link>https://blog.baifan.site/claude-code-permissions-security/</link>
      <pubDate>Wed, 20 May 2026 09:00:00 +0800</pubDate>
      <author>blog@baifan.site (ByF (BaiFan))</author>
      <guid>https://blog.baifan.site/claude-code-permissions-security/</guid>
      <description>&lt;p&gt;任何运行在宿主机用户态、依赖模型自律与应用层正则拦截的 AI Agent 权限系统，在操作系统内核视角下都不构成强制安全边界。Claude Code 设计的五级权限模型与交互式审批机制，是一套降低开发误操作概率的人机协作流控协议，在恶意对抗场景下缺乏沙箱约束力。&lt;/p&gt;</description>
      
    </item>
    
  </channel>
</rss>
