<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Step-Functions on Technodrone</title><link>https://blog.technodrone.cloud/tags/step-functions/</link><description>Recent content in Step-Functions on Technodrone</description><generator>Hugo</generator><language>en</language><lastBuildDate>Wed, 15 Jul 2026 07:05:00 +0300</lastBuildDate><atom:link href="https://blog.technodrone.cloud/tags/step-functions/index.xml" rel="self" type="application/rss+xml"/><item><title>Cascading Failures: When One Agent's Mistake Takes Down the Whole System (ASI08)</title><link>https://blog.technodrone.cloud/2026/07/cascading-failures-owasp-agentic-top10.html</link><pubDate>Wed, 15 Jul 2026 07:05:00 +0300</pubDate><guid>https://blog.technodrone.cloud/2026/07/cascading-failures-owasp-agentic-top10.html</guid><description>&lt;p&gt;This is post #8 of the &lt;a href="https://blog.technodrone.cloud/2026/07/aws-intro-owasp-agentic-top10.html"&gt;OWASP Agentic AI Top 10: What Builders on AWS Need to Know&lt;/a&gt; series.&lt;/p&gt;
&lt;p&gt;A search tool starts returning errors. 2% failure rate. Nothing dramatic.&lt;/p&gt;
&lt;p&gt;Your agent retries. Makes sense. But the agent&amp;rsquo;s retry logic lives inside a probabilistic reasoning engine that doesn&amp;rsquo;t understand backpressure. It retries immediately. And again. And again. Each retry burns tokens, adds latency, and puts more load on the already-struggling search service.&lt;/p&gt;</description></item></channel></rss>