<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Linux - Hugo FixIt Blog</title><link>https://pullsec.io/writeups/hackthebox/linux/</link><description>This is my new Hugo FixIt site</description><generator>Hugo 0.164.0 &amp; FixIt v1.0.0-alpha-mszlt87u</generator><language>en</language><lastBuildDate>Sat, 18 Apr 2026 05:12:22 +0200</lastBuildDate><atom:link href="https://pullsec.io/writeups/hackthebox/linux/index.xml" rel="self" type="application/rss+xml"/><item><title>Artificial</title><link>https://pullsec.io/writeups/hackthebox/artificial/</link><pubDate>Sat, 28 Jun 2025 17:15:55 +0200</pubDate><guid>https://pullsec.io/writeups/hackthebox/artificial/</guid><category domain="https://pullsec.io/categories/writeups/">Writeups</category><description>&lt;p&gt;Artificial was a really interesting one. At first glance, it looked like a simple web application, but the use of TensorFlow models immediately caught my attention. I hadn’t worked much with machine learning attack surfaces before, so digging into model deserialization vulnerabilities was both challenging and rewarding. Reproducing the environment locally with Docker helped me understand the behavior much better, and it felt great to turn that into a working RCE. Overall, this box was a solid reminder that modern technologies often introduce new and less obvious attack vectors.&lt;/p&gt;</description></item></channel></rss>