Source-backed architecture notes on building an agent
# Add to your Claude Code skills
git clone https://github.com/antinomie-lab/pi-bookLast scanned: 8/10/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-08-10T05:25:11.587Z",
"npmAuditRan": true,
"pipAuditRan": true,
"promptInjectionRan": true
}See how pi-book compares with popular alternatives.
pi-book is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by antinomie-lab. Source-backed architecture notes on building an agent. It has 404 GitHub stars.
Yes. pi-book passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
Clone the repository with "git clone https://github.com/antinomie-lab/pi-book" and add it to your Claude Code skills directory (see the Installation section above).
pi-book is primarily written in Vue. It is open-source under antinomie-lab on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh pi-book against similar tools.
No comments yet. Be the first to share your thoughts!
Based on votes and bookmarks from developers who liked this skill
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
一本中文架构书的工作区。对象是 pi 仓库的 packages/agent。
一个 agent 循环，被做成一个库。
这本书从代码出发，讲清楚 @earendil-works/pi-agent-core 是怎么搭起来的：它承诺什么、拒绝什么、支点在哪里。全书对应 pi 仓库 main 分支的 commit cd20a8d2e。代码引用写成 文件:行号，每个引用就地附上引文——读者不需要打开编辑器，就能核对书里任何一处论断。
多语言： 中文原稿在 agent/（唯一内容源头）。英文与西班牙文译本在 agent/en/ 与 agent/es/。Web 阅读器右上角可切换语言。
| 方式 | 入口 | 适合场景 |
|---|---|---|
| 🌐 Web 在线版 | books.antinomie.org/pi | 沉浸式阅读，代码引文高亮；可切换中 / EN / ES |
| 📥 Markdown 版 | agent/ · en · es | 下载到本地，配合 AI（Claude / Cursor 等）边读边问、对照源码 |
你会读 TypeScript，知道 LLM API 的基本概念（messages、tool calls、streaming），不需要事先了解 pi 仓库。
整体 → 局部 → 横切。
目录只列已发布的章节，后续章节随写随添。
agent/ —— 书的原稿（中文），唯一的内容源头。译本在 agent/en/、agent/es/。可见 agent/README.md 与 agent/TRANSLATION.md。
web/ —— 阅读器（Vite + Vue）。渲染 agent/ 下的中文原稿与译本，不存放正文内容。
cd web && npm install && npm run dev
引文规则、三种卡片组件（插叙 / 岔路 / 为什么不去），范围之外，都写在 agent/README.md 的「体例」一节。