<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Abhishek</title>
    <description>The latest articles on DEV Community by Abhishek (@b_abhishek).</description>
    <link>https://dev.to/b_abhishek</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3916457%2F1ace4a23-0bb5-49dd-9999-58990c72313d.jpeg</url>
      <title>DEV Community: Abhishek</title>
      <link>https://dev.to/b_abhishek</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/b_abhishek"/>
    <language>en</language>
    <item>
      <title>Loop-Watchdog</title>
      <dc:creator>Abhishek</dc:creator>
      <pubDate>Wed, 06 May 2026 18:06:47 +0000</pubDate>
      <link>https://dev.to/b_abhishek/loop-watchdog-491p</link>
      <guid>https://dev.to/b_abhishek/loop-watchdog-491p</guid>
      <description>&lt;p&gt;built a kill switch for looping AI coding agents&lt;/p&gt;

&lt;p&gt;AI coding agents kept:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;retrying the same fixes&lt;/li&gt;
&lt;li&gt;failing the same tests&lt;/li&gt;
&lt;li&gt;burning tokens endlessly&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;So I built Loop Watchdog.&lt;/p&gt;

&lt;p&gt;It sits between coding agents and OpenAI-compatible APIs and detects:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;repeated fix-break loops&lt;/li&gt;
&lt;li&gt;file churn&lt;/li&gt;
&lt;li&gt;retry spam&lt;/li&gt;
&lt;li&gt;repeating error patterns&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;When the loop score gets too high:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;the next model call is blocked&lt;/li&gt;
&lt;li&gt;the session is paused&lt;/li&gt;
&lt;li&gt;alerts are sent to dashboard/Slack/email&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Built with:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Python + FastAPI&lt;/li&gt;
&lt;li&gt;Cloudflare Workers&lt;/li&gt;
&lt;li&gt;D1&lt;/li&gt;
&lt;li&gt;local-first architecture&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Run it with:&lt;br&gt;
loop-watchdog start codex &lt;/p&gt;

&lt;p&gt;Then watch looping sessions get paused before more credits burn.&lt;/p&gt;

&lt;p&gt;Github: &lt;a href="https://github.com/bevinkatti/Loop-Watchdog" rel="noopener noreferrer"&gt;https://github.com/bevinkatti/Loop-Watchdog&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Would love feedback from people building with AI coding agents.&lt;/p&gt;

</description>
      <category>ai</category>
      <category>opensource</category>
      <category>python</category>
      <category>automation</category>
    </item>
  </channel>
</rss>
