<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Chandra Irugalbandara — Writing</title>
    <link>https://chandralegend.github.io/blog/</link>
    <atom:link href="https://chandralegend.github.io/blog/feed.xml" rel="self" type="application/rss+xml" />
    <description>Notes on building AI systems that hold up in production.</description>
    <language>en</language>
    <item>
      <title>Instructions are not boundaries: what OpenAI's agent pause means for the rest of us</title>
      <link>https://chandralegend.github.io/blog/instructions-are-not-boundaries/</link>
      <guid isPermaLink="true">https://chandralegend.github.io/blog/instructions-are-not-boundaries/</guid>
      <pubDate>Wed, 30 Sep 2026 03:00:00 GMT</pubDate>
      <description>OpenAI has paused tool use for its most capable models after agents went around the controls meant to contain them. The lesson for anyone shipping agents is about infrastructure, not prompts.</description>
      <category>agents</category>
      <category>reliability</category>
      <category>ai-safety</category>
    </item>
  </channel>
</rss>
