<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Agent Toolkit on The road</title><link>https://kane.mx/tags/agent-toolkit/</link><description>Recent content in Agent Toolkit on The road</description><generator>Hugo -- gohugo.io</generator><language>en</language><lastBuildDate>Tue, 12 May 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://kane.mx/tags/agent-toolkit/index.xml" rel="self" type="application/rss+xml"/><item><title>Agent Toolkit for AWS: What It Changes for Claude Code</title><link>https://kane.mx/posts/2026/agent-toolkit-for-aws-claude-code/</link><pubDate>Tue, 12 May 2026 00:00:00 +0000</pubDate><guid>https://kane.mx/posts/2026/agent-toolkit-for-aws-claude-code/</guid><description>
&lt;p>If you've been using Claude Code for AWS development, you've probably seen the pattern: you paste a CloudFormation snippet into your session, Claude suggests something plausible, you deploy it, and the stack events stream lights up with &lt;code>CREATE_FAILED&lt;/code> on a property the model couldn't have known about — because its training data stopped months ago.&lt;/p>
&lt;p>The usual workaround has been hand-rolling context into &lt;code>CLAUDE.md&lt;/code>: copying service endpoint quirks, IAM condition key syntax, and PrivateLink DNS formats that the model gets wrong. It works, but it's manual, fragile, and grows without bound.&lt;/p>
&lt;p>&lt;a href="https://kane.mx/posts/2026/agent-toolkit-for-aws-claude-code/">Read More&lt;/a>&lt;/p></description></item></channel></rss>