<?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>LLM on James&#39;s Blog</title>
    <link>https://james-farrell.com/tags/llm/</link>
    <description>Recent content in LLM on James&#39;s Blog</description>
    <image>
      <url>https://james-farrell.com/favicon.ico</url>
      <link>https://james-farrell.com/favicon.ico</link>
    </image>
    <generator>Hugo -- gohugo.io</generator>
    <lastBuildDate>Tue, 10 Nov 2026 01:46:41 +0000</lastBuildDate><atom:link href="https://james-farrell.com/tags/llm/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Securing a Self-Hosted AI Agent</title>
      <link>https://james-farrell.com/posts/securing-hermes/</link>
      <pubDate>Tue, 10 Nov 2026 01:46:41 +0000</pubDate>
      
      <guid>https://james-farrell.com/posts/securing-hermes/</guid>
      <description>I&amp;rsquo;ve been experimenting with Hermes Agent, an open-source AI agent that can execute commands, browse the web, manage scheduled tasks and interact with external services.
The idea is pretty simple: have an assistant running 24/7 that can handle some of the boring administrative tasks I would otherwise have to remember to do myself.
The problem is that an AI agent with shell access and the ability to execute arbitrary code isn&amp;rsquo;t something I&amp;rsquo;d be particularly comfortable running alongside my personal files, banking credentials or other services.</description>
    </item>
    
  </channel>
</rss>
