Original Summary

I turned Andrej Karpathy’s tips on understanding LLM output into an Open Source Agent Skill : [make-it-click](<a href="https:&#x2F;&#x2F;github.com&#x2F;ajithraghavan&#x2F;make-it-click" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;ajithraghavan&#x2F;make-it-click</a>)<p>Inspired by his X post : as Models do more of the work, we spend more time understanding it. So the format of an answer matters<p>The Skill picks the simplest format that makes an idea click, and moves up only when needed :<p><pre><code> Clear writing in ASD-STE100 style (strict, 80%, or light) Diagrams Interactive HTML pages Narrated explainer videos (3b1b style) </code></pre> Bonus : an oversight mode for reviewing Agent work (what changed -&gt; why -&gt; risk -&gt; what to check)<p>MIT licensed. Works with Claude.ai and Claude Code<p><a href="https:&#x2F;&#x2F;github.com&#x2F;ajithraghavan&#x2F;make-it-click" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;ajithraghavan&#x2F;make-it-click</a><p>Inspired by @karpathy: [<a href="https:&#x2F;&#x2F;x.com&#x2F;karpathy&#x2F;status&#x2F;2105819303471976479" rel="nofollow">https:&#x2F;&#x2F;x.com&#x2F;karpathy&#x2F;status&#x2F;2105819303471976479</a>]

  • 情报分类：技术学习与提效
  • 分类依据：内容涉及技术、AI、软件工具或工程实践
  • 信息来源：Hacker News 新项目
  • 发布时间：2026/10/6 22:03:18