<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Agentskills.io on Text Matrix</title><link>https://txtmix.com/tags/agentskills.io/</link><description>Recent content in Agentskills.io on Text Matrix</description><generator>Hugo</generator><language>zh-cn</language><lastBuildDate>Tue, 21 Jul 2026 20:06:14 +0800</lastBuildDate><atom:link href="https://txtmix.com/tags/agentskills.io/index.xml" rel="self" type="application/rss+xml"/><item><title>Anthropic Cybersecurity Skills：把 817 个安全分析任务打包成 AI Agent 可直接调用的标准技能库</title><link>https://txtmix.com/posts/tech/mukul975-anthropic-cybersecurity-skills-guide/</link><pubDate>Fri, 26 Jun 2026 18:07:28 +0800</pubDate><guid>https://txtmix.com/posts/tech/mukul975-anthropic-cybersecurity-skills-guide/</guid><description>&lt;h2 id="一句话判断">一句话判断&lt;/h2>
&lt;p>&lt;code>mukul975/Anthropic-Cybersecurity-Skills&lt;/code> 不是漏洞扫描器、不是 payload 仓库、也不是写给人类看的安全手册。它把 817 个&amp;quot;资深安全分析师在特定场景下会做的标准操作&amp;quot;按 &lt;a href="https://agentskills.io" target="_blank" rel="noopener noreffer ">agentskills.io&lt;/a> 规范打包成结构化技能单元,每个技能都映射到 6 套主流安全框架(MITRE ATT&amp;amp;CK v19.1、NIST CSF 2.0、MITRE ATLAS v5.4、MITRE D3FEND v1.3、NIST AI RMF 1.0、MITRE F3 v1.1),目的是让 LLM(大语言模型)驱动的 Agent(自主智能体)在执行真实安全任务时,有一套和人类专家同构的决策脚手架。&lt;/p></description></item></channel></rss>