<?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>安全评估 on 硅基 AGI · 智能体学习与测评</title><link>https://guijiagi.com/tags/%E5%AE%89%E5%85%A8%E8%AF%84%E4%BC%B0/</link><description>Recent content in 安全评估 on 硅基 AGI · 智能体学习与测评</description><generator>Hugo</generator><language>zh-cn</language><copyright>本站内容采用 CC BY-NC-SA 4.0 国际许可协议授权</copyright><lastBuildDate>Tue, 30 Jun 2026 17:20:00 +0800</lastBuildDate><atom:link href="https://guijiagi.com/tags/%E5%AE%89%E5%85%A8%E8%AF%84%E4%BC%B0/index.xml" rel="self" type="application/rss+xml"/><item><title>AI Agent 安全攻防 2026：从越狱到权限管理</title><link>https://guijiagi.com/posts/agent-security-defense-2026/</link><pubDate>Tue, 30 Jun 2026 17:20:00 +0800</pubDate><guid>https://guijiagi.com/posts/agent-security-defense-2026/</guid><description>2026年AI Agent安全全景：常见攻击向量、防御策略、权限管理框架与安全评估方法</description></item><item><title>LLM 红队测试实践：攻击即防御</title><link>https://guijiagi.com/posts/red-teaming-llm/</link><pubDate>Wed, 24 Jun 2026 14:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/red-teaming-llm/</guid><description>系统性介绍 LLM 红队测试方法论，涵盖攻击向量分析、自动化工具链、修复流程与真实案例</description></item></channel></rss>