<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>AI Agent on 星核技术社</title><link>https://hyuzz-nuc.github.io/tags/ai-agent/</link><description>Recent content in AI Agent on 星核技术社</description><image><title>星核技术社</title><url>https://hyuzz-nuc.github.io/images/favicon.png</url><link>https://hyuzz-nuc.github.io/images/favicon.png</link></image><generator>Hugo</generator><language>zh-CN</language><lastBuildDate>Mon, 15 Jun 2026 10:00:00 +0800</lastBuildDate><atom:link href="https://hyuzz-nuc.github.io/tags/ai-agent/index.xml" rel="self" type="application/rss+xml"/><item><title>手写一个最小 Agent：50 行代码搞懂工具调用循环</title><link>https://hyuzz-nuc.github.io/posts/build-minimal-agent-from-scratch/</link><pubDate>Mon, 15 Jun 2026 10:00:00 +0800</pubDate><guid>https://hyuzz-nuc.github.io/posts/build-minimal-agent-from-scratch/</guid><description>从零手写一个最小 Agent，包含工具定义、tool call 解析、工具执行、最大步数和错误处理，50 行 Python 代码搞懂 Agent 的核心循环</description></item><item><title>Anthropic 实战解读：构建高效 Agent 的正确姿势</title><link>https://hyuzz-nuc.github.io/posts/anthropic-building-effective-agents/</link><pubDate>Sun, 07 Jun 2026 10:00:00 +0800</pubDate><guid>https://hyuzz-nuc.github.io/posts/anthropic-building-effective-agents/</guid><description>Anthropic 官方工程博客深度解读：从增强型 LLM 到五种 Workflow 模式，再到自主 Agent，一文搞懂 Agentic 系统的设计哲学</description></item></channel></rss>