Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add R2h1/my-book-skills --skill v13-20git clone --depth 1 https://github.com/R2h1/my-book-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/r2h1/my-book-skills/v13-20)<a href="https://agentmods.dev/skills/r2h1/my-book-skills/v13-20"><img src="https://agentmods.dev/badge/skills/r2h1/my-book-skills/v13-20/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/r2h1/my-book-skills/v13-20"><img src="https://agentmods.dev/badge/skills/r2h1/my-book-skills/v13-20.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00108 | $0.01727 |
| Opus 5 | $0.00054 | $0.00864 |
| Sonnet 5 | $0.00022 | $0.00345 |
| Haiku 4.5 | $0.00011 | $0.00173 |
Grade A, and why
20本法则 scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.
20本法则
R — 原文
"如果你要开咖啡厅,那就买20本关于咖啡厅的书来读吧。几乎所有的咖啡厅的老板都不读任何书,只是凭自己的直觉或头脑来开咖啡厅。"
— 金承焕, 第七章
I — 方法论骨架
20本法则是在最短时间内成为某个领域'内行'的方法:
- 20本足够建立认知框架 — 20本书能覆盖该领域90%以上的核心概念、方法论和常见错误。即使每本只理解了20%,也足以超过90%不读书的从业者
- 大脑自动转换视角 — 连续读同一领域的书,大脑会开始用该领域的视角看世界——'拿着锤子看到的所有东西都是钉子'
- 身份认同开始转变 — 读20本后,你开始觉得自己'是做XX的人',身份认同自然转变
- 选书标准 — 不是随便20本,而是该领域公认的经典和畅销书
核心洞察:从业者90%不读书,所以即使只读了最基础的20本书,你就已经比90%的人更懂这个领域。
A1 — 书中的应用
案例 1: 作者自己用20本法则
- 问题: 要进入一个新领域(从商业到自我开发到YouTube)
- 方法论的使用: 每次进入新领域都先集中买20本书阅读
- 结论: 即使只读了20本,也比90%凭直觉做的人强
- 结果: 在多个领域快速建立专业度
案例 2: 亲戚弟弟学习博客营销
- 问题: 手机店需要网络营销能力
- 方法论的使用: 作者让他读相关书籍,用20本法则快速入门
- 结论: 从零开始通过阅读建立营销知识
- 结果: 1年内月收入从600万升至4000万韩元
案例 3: 建议开咖啡厅前读20本书
- 问题: 大多数人凭直觉开咖啡厅,3个月内倒闭
- 方法论的使用: 先读20本关于咖啡厅的书再行动
- 结论: 读过书的老板会在那附近排进前两名
A2 — 触发场景
用户会在什么情境下需要这个 skill?
- 用户想进入新领域但不知从何下手
- 用户觉得某领域知识太难或太专业
- 用户已经在该领域工作但发现自己在凭直觉
语言信号
- '我对XX感兴趣但完全不懂'
- 'XX好难,不知道从哪里开始'
- '我都是靠自己摸索'
- '我想学XX但不知道读什么书'
与相邻 skill 的区分
- 与
22战略的关系: 20本法则是'选书方法'(横向快速入门),22战略是'坚持读写的方法'(纵向深度训练)',两者互补 - 与
提坦道具组合策略的关系: 20本法则是将某个技能提升到B级的最佳路径
E — 可执行步骤
当 skill 被激活后,agent 应按以下步骤执行:
-
列出20本书单
- 确定目标领域
- 搜索该领域的经典和畅销书籍
- 选书标准:亚马逊/豆瓣评分高、作者权威、内容全面
- 完成标准: 列出至少20本目标领域书籍
-
快速阅读战略
- 不要求精读,不求每页都懂
- 遇到读不懂的跳过,不影响整体理解
- 目标是建立认知框架,不是成为专家
- 完成标准: 制定了'每天读XX页/每周读XX本'的计划
- 判停条件: 如果用户卡在某本书读不下去,建议换一本而不是死磕
-
输出验证
- 每读完一本写200字总结:这本书的核心观点,你学到了什么
- 完成标准: 完成20本书的阅读和输出
B — 边界
不要在以下情况使用此 skill
- 用户需要的是实践技能(如游泳/开车/烹饪)——这些领域需要动手而非阅读
- 目标领域好书不足20本(过于小众或新兴领域)
- 用户已经在该领域有扎实基础——20本法则是'入门'而非'深造'工具
作者在书中警告的失败模式
- ce26: 读书不按顺序——直接读难书导致放弃。建议从简单的书开始逐渐增加难度
- ce27: 用批判代替学习——初学者的傲慢,挑作者毛病而不是学习
- ce14: 只读书不实践——读完20本后必须行动,否则知识只是概念
作者的盲点 / 时代局限
- 某些领域(如AI、区块链)变化极快,书籍出版可能已经落后于实践——需要配合博客/论文更新
- 作者假设'能找到20本好书'——部分领域中文好书可能不够20本
- 20本需要一定的阅读速度和理解力,对阅读基础差的人可能负担过重
容易混淆的邻近方法论
- '精读一两本经典'(Less is More)——20本法则主张广度优先(先建立框架再深挖),与精读法不同
- 在线课程入门——书籍的信息密度通常高于视频课程,20本法则是更高效率的入门方式
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 144 lines · 108 tokens per session scan A d6be27423590
20本法则 is a skill published in the GitHub repository R2h1/my-book-skills (2 stars, last pushed 27d ago), licensed MIT. It adds 108 tokens to every session and 1,727 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
ai-learning-coach
An AI study coach that guides someone through setting a learning goal, understanding material, recalling it, practising, correcting mistakes, reviewing later, and completing a project. It is for learning a subject rather than simply receiving an answer.
book-digest-cn
A Chinese book-summary and reading-notes guide. It focuses on the question a book answers, the author's claims and evidence, and how the ideas can be applied, rather than simply repeating the table of contents.
homework-tutor-cn
A Chinese-language homework coaching guide for parents of primary and secondary school students.
chinese-lesson-plan
A Chinese-language lesson-planning guide for primary and secondary school teachers. It helps structure lessons, materials, learning goals, activities, and assessment for subjects such as Chinese, maths, English, and science.
book-distill
A method for turning a long book, transcript, podcast, course, interview, or article collection into small, reusable skills for an agent. It extracts principles, checklists, decision procedures, limits, and a glossary.
book-distiller
A process for turning a book into a practical, reusable method guide rather than a simple summary. It researches the book’s methods, decision rules, examples, limits, and supporting criticism.