<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>Toolhive on thrix&#39;s blog</title>
    <link>http://vadkerti.net/tags/toolhive/</link>
    <description>Recent content in Toolhive on thrix&#39;s blog</description>
    <generator>Hugo -- 0.152.2</generator>
    <language>en-us</language>
    <lastBuildDate>Mon, 12 Oct 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="http://vadkerti.net/tags/toolhive/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Keeping secrets out of my AI agents</title>
      <link>http://vadkerti.net/posts/keeping-secrets-out-of-ai-agents/</link>
      <pubDate>Mon, 12 Oct 2026 00:00:00 +0000</pubDate>
      <guid>http://vadkerti.net/posts/keeping-secrets-out-of-ai-agents/</guid>
      <description>I started with API keys and tokens leaking into agent context, went through CLAUDE.md rules and fetching secrets on demand from 1Password, and ended up at sandboxed agents, MCP servers that hold the credentials, and one-hour GitHub and GitLab tokens on a dedicated identity.</description>
    </item>
  </channel>
</rss>
