<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
	<channel>
		<title>Agents on Nick Perkins</title>
		<link>https://nickperkins.au/tags/agents/</link>
		<description>Recent content in Agents on Nick Perkins</description>
		<generator>Hugo</generator>
		<language>en</language>
		
		
		
		
			<lastBuildDate>Sun, 19 Jul 2026 10:00:00 +1000</lastBuildDate>
		
			<atom:link href="https://nickperkins.au/tags/agents/index.xml" rel="self" type="application/rss+xml" />
			<item>
				<title>Agent to Agent Communication: MCP Is the Wrong Layer</title>
				<link>https://nickperkins.au/article/agent-communication-protocol/</link>
				<pubDate>Sun, 19 Jul 2026 10:00:00 +1000</pubDate>
				<guid>https://nickperkins.au/article/agent-communication-protocol/</guid>
				<description>&lt;p&gt;A few weeks ago I wrote about &lt;a href=&#34;https://nickperkins.au/article/skills-over-mcps/&#34;&gt;replacing MCPs with skills and scripts&lt;/a&gt;. Someone pushed back in the comments with one exception I couldn&amp;rsquo;t wave away: agent to agent communication has no CLI. I pointed them at Agent Communication Protocol and moved on, but the question stuck, and I&amp;rsquo;ve spent some time since looking at A2A properly.&lt;/p&gt;&#xA;&lt;h2 id=&#34;the-gap-id-papered-over&#34;&gt;The gap I&amp;rsquo;d papered over&lt;/h2&gt;&#xA;&lt;p&gt;In the skills post I argued that for most tool connection, a skill file plus a script beats an MCP server. Less context tax, less cost, lighter workflow. I still think that&amp;rsquo;s true for tools.&lt;/p&gt;</description>
			</item>
	</channel>
</rss>
