<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
	<channel>
		<title>Security on AIBussin — AI applications, systems and books</title>
		<link>https://aibussin.com/categories/security/</link>
		<description>Recent content in Security on AIBussin — AI applications, systems and books</description>
		<generator>Hugo</generator>
		<language>en-US</language>
		
		
		
		
			<lastBuildDate>Mon, 05 Oct 2026 09:00:00 +0100</lastBuildDate>
		
			<atom:link href="https://aibussin.com/categories/security/index.xml" rel="self" type="application/rss+xml" />
			<item>
				<title>The Project .pi Directory and Trust</title>
				<link>https://aibussin.com/books/pi/08-chapter/</link>
				<pubDate>Fri, 02 Oct 2026 09:00:00 +0100</pubDate>
				<guid>https://aibussin.com/books/pi/08-chapter/</guid>
				<description>&lt;p&gt;You clone a repository. It contains &lt;code&gt;.pi/extensions/migrate-schema.ts&lt;/code&gt;, and &lt;code&gt;migrate-schema.ts&lt;/code&gt; calls &lt;code&gt;execSync&lt;/code&gt; with a string built from a prompt. Pi asks whether you trust the folder. You are being asked a real question: does &amp;ldquo;yes&amp;rdquo; merely acknowledge the repository, or does it allow that extension to load and run with your process permissions?&lt;/p&gt;&#xA;&lt;p&gt;&lt;code&gt;security.md&lt;/code&gt; answers it precisely, and the answer has a wrinkle worth knowing before you answer.&lt;/p&gt;&#xA;&lt;h2 id=&#34;what-triggers-the-prompt&#34;&gt;What triggers the prompt&lt;/h2&gt;&#xA;&lt;p&gt;Pi requires a project-trust decision when it finds any of these from the current working directory:&lt;/p&gt;</description>
			</item>
			<item>
				<title>Authority, Isolation, and What Makes an Agent Worth Building</title>
				<link>https://aibussin.com/books/pi/42-chapter/</link>
				<pubDate>Mon, 05 Oct 2026 09:00:00 +0100</pubDate>
				<guid>https://aibussin.com/books/pi/42-chapter/</guid>
				<description>&lt;p&gt;Everything so far has been a mechanism. Tools, extensions, events, sessions, modes, protocols. This chapter is about the one question that turns a set of mechanisms into something you would let near a production repository, and about what you are then able to claim.&lt;/p&gt;&#xA;&lt;p&gt;Two claims carry the chapter. &lt;strong&gt;Capability is not authority:&lt;/strong&gt; Pi can do almost anything the account that started it can do, and several documented mechanisms bear on whether it &lt;em&gt;should&lt;/em&gt; do a particular thing. Those mechanisms matter, but none by itself is a sandbox. And &lt;strong&gt;a prohibition on what the agent can reach needs enforcement outside the agent&amp;rsquo;s ability to override it.&lt;/strong&gt; That may be an operating-system restriction, a credential with limited scope, or a remote service that independently denies an operation. A prompt, an in-process gate and a tool allowlist can still provide useful controls, but they must not be mistaken for containment of the whole process.&lt;/p&gt;</description>
			</item>
	</channel>
</rss>
