<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Development on Percona Community</title><link>https://percona.community/tags/development/</link><description>Recent content in Development on Percona Community</description><generator>Hugo</generator><language>en-us</language><copyright>© Percona Community. MySQL, InnoDB, MariaDB and MongoDB are trademarks of their respective owners.</copyright><lastBuildDate>Tue, 05 May 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://percona.community/tags/development/index.xml" rel="self" type="application/rss+xml"/><item><title>How I Stopped Babysitting My Coding Agent (With Dotfiles)</title><link>https://percona.community/blog/2026/05/05/how-i-stopped-babysitting-my-coding-agent-with-dotfiles/</link><pubDate>Tue, 05 May 2026 00:00:00 +0000</pubDate><guid>https://percona.community/blog/2026/05/05/how-i-stopped-babysitting-my-coding-agent-with-dotfiles/</guid><description>&lt;p>Most developers at least try to use coding agents for development-related tasks, but babysitting LLMs and managing their permissions is no fun.
Completely skipping permission checks is a dangerous idea on your main machine, and setting up containers or VMs for sandboxing is a pain.
Can we do better?&lt;/p>
&lt;h3 id="the-autonomy-problem">The autonomy problem&lt;/h3>
&lt;p>If you work in software development, you have most certainly heard the phrase:&lt;/p>
&lt;blockquote>
&lt;p>Let&amp;rsquo;s just use an LLM to solve it!&lt;/p></description></item></channel></rss>