Cangjie Skill is a system that turns methods from books, long videos, podcasts, and other source material into executable skills for AI agents. It helps users package knowledge into callable workflows, using the repository's code, methods, and templates; catalogue add-ons relate to these agent skills.
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 kangarooking/cangjie-skill --skill reading-metaskillgit clone --depth 1 https://github.com/kangarooking/cangjie-skillWrote 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/kangarooking/cangjie-skill/reading-metaskill)<a href="https://agentmods.dev/skills/kangarooking/cangjie-skill/reading-metaskill"><img src="https://agentmods.dev/badge/skills/kangarooking/cangjie-skill/reading-metaskill/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/kangarooking/cangjie-skill/reading-metaskill"><img src="https://agentmods.dev/badge/skills/kangarooking/cangjie-skill/reading-metaskill.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00139 | $0.01571 |
| Opus 5 | $0.00069 | $0.00785 |
| Sonnet 5 | $0.00028 | $0.00314 |
| Haiku 4.5 | $0.00014 | $0.00157 |
Grade A, and why
reading-metaskill 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 10d 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 — 122 lines — stays where its author put it; the contents beside it link to each section on GitHub.
阅读元技能
R — 原文 (Reading)
读你所爱读的直到你爱上阅读。……我大概每天读一到两小时。那个使我进入0.00001%的行列。……真正的人不会每天读一小时。真正的人，每天读一分钟或更少。使它成为实际的习惯才是最重要的事情。
— 纳瓦尔·拉维坎特, 《纳瓦尔宝典》 第一章·财富
I — 方法论骨架 (Interpretation)
阅读是可以用它换来任何其他技能的元技能。培养它的关键是降低门槛、保持愉悦： ① 读你所爱（包括「精神垃圾食品」）直到爱上阅读——没有垃圾这种东西； ② 没有读完义务——跳读、从中间读、同时穿插 10–20 本都合法，把书当博客/推文； ③ 质量靠「原著优先」——先读达尔文再读道金斯、先读亚当·斯密再读当代经济学家，解读本会灌输立场； ④ 每天 1–2 小时、让它成为实际习惯，比「读很多」更重要（数量是虚荣指标）； ⑤ 以教促学——向别人解释你学到的东西，能讲明白才是真掌握。
A1 — 书中的应用 (Past Application)
案例 1: 同时读 10–20 本
- 问题: 被「必须读完」训练毁了阅读习惯
- 方法论的使用: 把书当博客，无读完义务，穿插跳读
- 结论: 「我觉得没有任何义务去读完这本书。突然间，书籍又回到了我的阅读库。」
- 结果: 从 Twitter 多巴胺时代回到每天 1–2 小时深度阅读
案例 2: 读书数十年后重读
- 问题: 如何真正掌握一本书
- 方法论的使用: 「越好的书，就越要慢慢地理解和吸收」；重读好书
- 结论: 「我不想去读完所有书，我只想反复读那100本伟大的书」
- 结果: 少而精的重读取代数量竞赛
A2 — 触发场景 (Future Trigger) ★
用户会在什么情境下需要这个 skill?
- 想读书但读不进去/总卡在某页
- 问「入门某领域该读什么」
- 想提升学习能力：「怎么学得又快又牢」
- 被书单焦虑：「别人一年读 100 本我好焦虑」
语言信号
- "推荐几本书入门/怎么开始读书"
- "我读不进去/总半途而废"
- "怎么学习新领域"
- "how to read more / learn anything / what to read first"
与相邻 skill 的区分
- 与
judgment-training的区别: 本 skill 是输入管线（怎么读），判断力是输出能力（怎么想） - 与
screen-detox的区别: 深度阅读替代屏幕多巴胺，但本 skill 不负责戒断
E — 可执行步骤 (Execution)
-
降低门槛，随手开读
- 完成标准: 挑一本「读起来有意思」的书（不挑「应该读」的），从最吸引你的章节开始
- 判停条件: 若读得痛苦超过 20 分钟，换一本或跳读，不做完读义务
-
建立每日最小习惯
- 完成标准: 每天固定 1–2 小时（或固定 15 分钟起步），连续 7 天不断
-
按「原著优先」补基础
- 完成标准: 对想入门的领域，先找到该领域 1–2 本奠基原著加入队列
-
以教促学
- 完成标准: 每周写/讲一次「我最近学到的东西」，直到能向小孩讲明白
B — 边界 (Boundary) ★
不要在以下情况使用此 skill
- 用户要具体书评/摘录（这不是阅读习惯训练）
- 备考场景（应试策略与培养习惯不同）
作者在书中警告的失败模式
- 读错次序的博学: 「一开始读的是一系列虚假的或部分真实的东西，这些东西组成他们世界观的基本公理」
- 为社会认可而读: 读大家都在读的书=融入羊群，生活回报在脱离群体的一侧
作者的盲点 / 时代局限
- 每天 1–2 小时对高压人群不现实；「读原著」假设读者有耐心与语言门槛（译本也是折中）
- 作者强调非虚构/经典，未覆盖小说与专业实操书的阅读价值
容易混淆的邻近方法论
judgment-training: 阅读供给基础，判断力消费基础
相关 skills (阶段 3 定稿)
- composes-with:
judgment-training、productize-yourself（终身学习=特殊知识来源） - contrasts-with:
screen-detox（深度阅读 vs 多巴胺零食）
审计信息
- 验证通过: V1 ✓ / V2 ✓ / V3 ✓ (v08)
- 测试通过率: 见 test-results.md
- 蒸馏时间: 2026-08-01
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 10d ago First seen · 122 lines · 139 tokens per session scan A 00bb3092ea7a
reading-metaskill is a skill published in the GitHub repository kangarooking/cangjie-skill (9,743 stars, last pushed 3d ago), licensed MIT. It adds 139 tokens to every session and 1,571 once invoked, about $0.0007 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 2026-08-30.
Other skills, from other repositories
skill-builder
Use when automatically detect source types and build AI skills using Skill Seekers. Use when the user wants to create skills from documentation, repos, PDFs, videos, or other knowledge sources.
us-program-research
This skill should be used when the user needs structured research and ranking of US academic programs (PhD, Master.
understand
Teach, explain, or reshape an answer so the user genuinely comprehends and can use it — leading with the governing idea, showing the mechanism, using worked examples, and checking understanding rather than just simplifying. Use when the user invokes $understand, says they don't follow, asks to learn or understand…
comprehend-languages
Use this skill when the words are legible but the meaning is trapped inside an unfamiliar dialect, format, or community.
legend-lore
Use this spell when you need everything knowable about a person, company, library, framework, concept, or system — not a quick answer, but a deep dossier.
layered-first-principles-teaching
Use when explaining complex concepts to others, designing training materials, or preparing technical presentations with progressive disclosure.