<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Devtools on Kuldeep Pisda</title><link>https://kdpisda.in/tag/devtools/</link><description>Recent content in Devtools on Kuldeep Pisda</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Sat, 29 Aug 2026 09:00:00 +0530</lastBuildDate><atom:link href="https://kdpisda.in/tag/devtools/index.xml" rel="self" type="application/rss+xml"/><item><title>Claude Code's Restricted Mode Closes the Config Escape Hatch on Sandboxed Agents</title><link>https://kdpisda.in/claude-code-restricted-mode/</link><pubDate>Sat, 29 Aug 2026 09:00:00 +0530</pubDate><guid>https://kdpisda.in/claude-code-restricted-mode/</guid><description>&lt;p&gt;Anthropic shipped &lt;code&gt;--restricted&lt;/code&gt; in Claude Code version 2.1.248 this week, and it is a smaller-looking change than it actually is. Most permission work in coding agents has been about which mode a session starts in: plan mode, accept-edits mode, full auto with &lt;code&gt;bypassPermissions&lt;/code&gt;. What those modes have in common is that they are settings, and settings live in files, and files can be edited, inherited from a parent directory, or shipped by someone else entirely. &lt;code&gt;--restricted&lt;/code&gt; is the first Claude Code flag that assumes the settings themselves might be the attacker, not just the agent reading them.&lt;/p&gt;</description></item></channel></rss>