<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
	<channel>
		<title>Jun Suzuki</title>
		<description>Writing about agent engineering, AI, and legal tech.</description>
		<link>https://blog.junsuzuki.xyz</link>
		<atom:link href="https://blog.junsuzuki.xyz/rss.xml" rel="self" type="application/rss+xml"/>
		
		<item>
			<title>Anthropic&apos;s skills playbook vs our custom knowledge layer</title>
			<description>Anthropic packs domain knowledge into Claude Code skills. We built a separate knowledge layer. Turns out it&apos;s the same architecture.</description>
			<link>https://blog.junsuzuki.xyz/blog/anthropic-skills-playbook-custom-knowledge</link>
			<guid isPermaLink="true">https://blog.junsuzuki.xyz/blog/anthropic-skills-playbook-custom-knowledge</guid>
			<pubDate>Sat, 04 Apr 2026 00:00:00 GMT</pubDate>
		</item>
		<item>
			<title>We outgrew CLAUDE.md: building a knowledge layer that compounds</title>
			<description>CLAUDE.md doesn&apos;t scale past a few hundred lines. Here&apos;s how we built a repo-scoped knowledge layer that gives Claude Code the right context at the right time.</description>
			<link>https://blog.junsuzuki.xyz/blog/beyond-claude-md-repo-scoped-knowledge-layer</link>
			<guid isPermaLink="true">https://blog.junsuzuki.xyz/blog/beyond-claude-md-repo-scoped-knowledge-layer</guid>
			<pubDate>Wed, 18 Mar 2026 00:00:00 GMT</pubDate>
		</item>
	</channel>
</rss>