<?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>Ghostwriter on The Engine Room</title><link>/en/tags/ghostwriter/</link><description>Recent content in Ghostwriter on The Engine Room</description><generator>Hugo -- 0.155.3</generator><language>en-us</language><lastBuildDate>Sun, 19 Jul 2026 12:00:00 +0300</lastBuildDate><atom:link href="/en/tags/ghostwriter/index.xml" rel="self" type="application/rss+xml"/><item><title>GhostWriter: Poisoning Long-Term Memory in LLM Agents</title><link>/en/notes/ghostwriter_memory_poisoning/</link><pubDate>Sun, 19 Jul 2026 12:00:00 +0300</pubDate><guid>/en/notes/ghostwriter_memory_poisoning/</guid><description>Why a personal AI agent&amp;rsquo;s memory is a separate attack surface and how to protect it</description></item></channel></rss>