<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
	<channel>
		<title>Posts on Ertuğ Karamatlı</title>
		<link>https://karamatli.com/posts/</link>
		<description>Recent content in Posts on Ertuğ Karamatlı</description>
		<generator>Hugo</generator>
		<language>en</language>
		
		
		
			<copyright>© Ertuğ Karamatlı</copyright>
		
		
			<lastBuildDate>Tue, 21 Jul 2026 00:00:00 +0000</lastBuildDate>
		
			<atom:link href="https://karamatli.com/posts/index.xml" rel="self" type="application/rss+xml" />
			<item>
				<title>Network-restricted, hardened Linux VMs for AI agents and untrusted code</title>
				<link>https://karamatli.com/posts/network-isolated-kvm-sandbox-ai-agents/</link>
				<pubDate>Tue, 21 Jul 2026 00:00:00 +0000</pubDate>
				<guid>https://karamatli.com/posts/network-isolated-kvm-sandbox-ai-agents/</guid>
				<description>Between supply chain attacks and increasingly capable AI coding agents, you&amp;rsquo;re already running code you&amp;rsquo;ve never read, with your own privileges. Instead of sandboxing individual tools, put the whole development environment (editor, agents, containers, and all) inside a plain KVM VM that can reach the internet but has no route to the host, the LAN, or any other private address. The companion tool kvm-sandbox creates a hardened sandbox VM with one command and verifies the network boundary actually holds. To stress-test that claim, I gave an AI agent root inside one and told it to break out; it couldn&amp;rsquo;t.</description>
			</item>
	</channel>
</rss>
