<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Condukt blog</title>
    <link>https://condukt.tuist.dev/blog/</link>
    <description>A framework for building AI agents in Elixir.</description>
    <language>en-US</language>
    <atom:link href="https://condukt.tuist.dev/blog/rss.xml" rel="self" type="application/rss+xml" />
    
    <lastBuildDate>Sun, 03 May 2026 00:00:00 GMT</lastBuildDate>
    
    <item>
      <title>Where an agent&#39;s tools should run</title>
      <link>https://condukt.tuist.dev/blog/sandboxes-for-tool-calls/</link>
      <guid isPermaLink="true">https://condukt.tuist.dev/blog/sandboxes-for-tool-calls/</guid>
      <pubDate>Sun, 03 May 2026 00:00:00 GMT</pubDate>
      <author>The Condukt team</author>
      <description>We just landed sandboxes in Condukt. Same agent, swap where the tool calls actually execute. Here is the thinking behind it.</description>
    </item>
    
    <item>
      <title>Secrets belong in the session</title>
      <link>https://condukt.tuist.dev/blog/agent-session-secrets/</link>
      <guid isPermaLink="true">https://condukt.tuist.dev/blog/agent-session-secrets/</guid>
      <pubDate>Sun, 03 May 2026 00:00:00 GMT</pubDate>
      <author>The Condukt team</author>
      <description>Agents need to act against real systems. We added session secrets to Condukt so credentials become part of the execution boundary, not part of the conversation.</description>
    </item>
    
  </channel>
</rss>
