<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Agent SDK on Bruce on AI Engineering</title><link>http://www.heyuan110.com/tags/agent-sdk/</link><description>Recent content in Agent SDK on Bruce on AI Engineering</description><generator>Hugo</generator><language>en</language><lastBuildDate>Fri, 17 Apr 2026 10:00:00 +0800</lastBuildDate><atom:link href="http://www.heyuan110.com/tags/agent-sdk/index.xml" rel="self" type="application/rss+xml"/><item><title>Claude Agent SDK: Build Production AI Agents in Python (2026 Guide)</title><link>http://www.heyuan110.com/posts/ai/2026-04-17-claude-agent-sdk-guide/</link><pubDate>Fri, 17 Apr 2026 10:00:00 +0800</pubDate><guid>http://www.heyuan110.com/posts/ai/2026-04-17-claude-agent-sdk-guide/</guid><description>&lt;p&gt;&lt;img src="http://www.heyuan110.com/posts/ai/2026-04-17-claude-agent-sdk-guide/cover.webp"
 alt="Claude Agent SDK: build production AI agents with Python toolchain"
 
 loading="lazy"
 decoding="async"
 fetchpriority="low"
 width="1200"
 height="630"
/&gt;
&lt;/p&gt;
&lt;p&gt;Most AI agent frameworks make you implement the boring parts: tool execution loops, file permission handling, command timeout management. You spend more time on plumbing than on the actual agent logic.&lt;/p&gt;
&lt;p&gt;The Claude Agent SDK takes a different approach. Instead of giving you an LLM API and saying &amp;ldquo;build your own tools,&amp;rdquo; it exposes the exact same toolchain that powers Claude Code — Read, Write, Edit, Bash, Glob, Grep, WebSearch — as a Python (and TypeScript) library. Claude decides which tools to use, executes them, observes results, and continues autonomously.&lt;/p&gt;</description></item></channel></rss>