<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Agentic Workflows on Nick Perkins</title>
    <link>https://nickperkins.au/tags/agentic-workflows/</link>
    <description>Recent content in Agentic Workflows on Nick Perkins</description>
    <generator>Hugo</generator>
    <language>en</language>
    <lastBuildDate>Sat, 20 Jun 2026 10:00:00 +1000</lastBuildDate>
    <atom:link href="https://nickperkins.au/tags/agentic-workflows/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>I Stopped Using MCPs and Replaced Them with Skills and Scripts</title>
      <link>https://nickperkins.au/article/skills-over-mcps/</link>
      <pubDate>Sat, 20 Jun 2026 10:00:00 +1000</pubDate>
      <guid>https://nickperkins.au/article/skills-over-mcps/</guid>
      <description>Everyone&amp;rsquo;s excited about MCPs. The Model Context Protocol is supposed to be the standard way AI agents talk to external services. And it is good, for what it is. But I&amp;rsquo;ve been replacing MCPs with skills and scripts in my own workflows, and honestly, I&amp;rsquo;m not going back.&#xA;The problem with MCPs When you load an MCP server, all of its tool definitions get injected into the agent&amp;rsquo;s context. Every tool name, every parameter schema, every description.</description>
    </item>
  </channel>
</rss>
