<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Memory Poisoning on Vivian＠資安媽媽包</title><link>https://vvnblog.com/tags/memory-poisoning/</link><description>Recent content in Memory Poisoning on Vivian＠資安媽媽包</description><generator>Hugo -- gohugo.io</generator><language>zh-tw</language><copyright>Copyright © 2026 Vivian All Rights Reserved.</copyright><lastBuildDate>Sun, 16 Aug 2026 00:00:00 +0800</lastBuildDate><atom:link href="https://vvnblog.com/tags/memory-poisoning/index.xml" rel="self" type="application/rss+xml"/><item><title>AI 黑魔法（15）：當 AI 會自己動手，拆解 Agent 失守的瞬間</title><link>https://vvnblog.com/posts/ai-dark-arts-15/</link><pubDate>Sun, 16 Aug 2026 00:00:00 +0800</pubDate><guid>https://vvnblog.com/posts/ai-dark-arts-15/</guid><description>AI 一旦有了工具、記憶和自主迴圈，錯誤就不再只停在回答裡，工具參數、記憶裡的一句話、專案裡的一份文字檔，都可能被 Agent 一路升級成已經核准的指令。</description></item></channel></rss>