<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
	<channel>
		<title>Ai on Website of Dmitry Kabanov</title>
		<link>https://dmitrykabanov.com/tags/ai/</link>
		<description>Recent content in Ai on Website of Dmitry Kabanov</description>
		<generator>Hugo</generator>
		<language>en-us</language>
		
		
		
		
			<lastBuildDate>Sat, 31 Oct 2026 00:00:00 +0000</lastBuildDate>
		
			<atom:link href="https://dmitrykabanov.com/tags/ai/index.xml" rel="self" type="application/rss+xml" />
			<item>
				<title>Using Docker Sandboxes with Google Antigravity</title>
				<link>https://dmitrykabanov.com/blog/2026/using-docker-sandboxes-with-google-antigravity/</link>
				<pubDate>Sat, 31 Oct 2026 00:00:00 +0000</pubDate>
				<guid>https://dmitrykabanov.com/blog/2026/using-docker-sandboxes-with-google-antigravity/</guid>
				<description>&lt;p&gt;It is 2026, and it is common to run AI agents in a terminal or an integrated development environment for coding tasks.&lt;/p&gt;&#xA;&lt;p&gt;The problem is, AI agents can be insecure to run directly on your machine:&#xA;only in Summer 2026 were several incidents reported.&lt;sup id=&#34;fnref:1&#34;&gt;&lt;a href=&#34;#fn:1&#34; class=&#34;footnote-ref&#34; role=&#34;doc-noteref&#34;&gt;1&lt;/a&gt;&lt;/sup&gt;&#xA;I have recently experienced a strange situation myself&#xA;when &lt;a href=&#34;https://antigravity.google/&#34;&gt;Antigravity&lt;/a&gt;, an AI agent from Google, has asked me&#xA;if it was allowed to run tests from the &lt;code&gt;Tests/&lt;/code&gt; directory&#xA;&lt;em&gt;one level above the current working directory&lt;/em&gt;,&#xA;because it has looked there for unknown reason.&#xA;Of course, many software packages look for different things on the file system,&#xA;for example, for a C compiler, etc.&#xA;However, they usually check some predefined system locations like &lt;code&gt;/usr/bin&lt;/code&gt;&#xA;on Linux.&lt;/p&gt;</description>
			</item>
	</channel>
</rss>
