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 yunshu0909/yunshu_skillshub --skill article-studygit clone --depth 1 https://github.com/yunshu0909/yunshu_skillshubWrote 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/yunshu0909/yunshu_skillshub/article-study)<a href="https://agentmods.dev/skills/yunshu0909/yunshu_skillshub/article-study"><img src="https://agentmods.dev/badge/skills/yunshu0909/yunshu_skillshub/article-study/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/yunshu0909/yunshu_skillshub/article-study"><img src="https://agentmods.dev/badge/skills/yunshu0909/yunshu_skillshub/article-study.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 | $0.00364 | $0.02596 |
| Opus 5 | $0.00182 | $0.01298 |
| Sonnet 5 | $0.00073 | $0.00519 |
| Haiku 4.5 | $0.00036 | $0.00260 |
Grade A, and why
article-study 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 11d 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 — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
和用户一起学一篇文章
哲学：「读一遍觉得懂了」是流畅性错觉——和 AI 说「完成了」一样，是主观判断。 这套流程干的事，就是把学习也做成可验证的：输入结构化、过程有考试、错题定位回退。 一句话：学东西和写代码，用的是同一套质检逻辑。
五条红线（违反即返工）
- 不做自学大纲：你是带着学的老师，不是列书单的人。每一讲的内容你自己讲出来，不要写"建议你去读 X 章"。
- 每讲必考：讲完一定要用户输出（复述 / 做题 / 动手写）。说不出来 = 没学会。
- 复述必拧紧：用户总结完，必须挑一到两处不精确的地方校准，且拧紧那句必须是否定/收窄结构（"关键不在 A 在 B" / "这里少了第三条腿" / "这个词不可判定，换成 X"）。写成"补充一点…" = 没拧紧，重写。
- 对号入座：每一讲都要落到用户自己的业务场景 / 现有工具 / 真实痛点上。接不上时先停下来问，不许硬贴标签。
- 原文与推论视觉可分：短文章必然要靠你的推论撑分量，但哪些是原文说的、哪些是你推的，必须一眼可辨（推论进
details或标「原文没直说，但成立」）。否则"我读了一篇文章"会悄悄变成"我听了 AI 讲课"。
红线约束的是你，不是用户。 红线防的是你偷懒跳过，不是用户的选择权。用户明确说"别考我了"时：不硬顶、也不静默放弃——降到轻量档，仍被拒就记挂账、说清代价、继续走（见
教学法.md「用户拒绝考核时」）。
开跑前（只问这些，其余别问）
- 材料放哪个目录（没有就
mkdir -p，全程用绝对路径） - 用户自己的业务场景是什么（第④步要用；答不出走
教学法.md的降级阶梯，不要追问第三遍） - 对齐投入度（一句话，别做成问卷）："完整跑（切几讲抽完干货再定，每讲一份课件带考核）还是先把核心讲一遍（20 分钟）？"——选后者走
教学法.md的「轻量档怎么跑」（仍要出资料+极简计划+轻量考核），讲完再问要不要转完整流程。别把想问 15 分钟的人拖进 6 讲。
五步主流程
① 抽干货 粗判材料形态 → 抓正文 → 量体量 → 就地整理成结构化笔记（之后不再回原文）
② 切讲次 按文章自带骨架切，先量素材定讲数（≤8 讲），进度表落盘
③ 学-考-讲 每讲循环：HTML 课件带讲 → 点名作业 →【硬停等用户】→ 拧紧校准 → 落盘笔记
④ 对号入座 贯穿③：先锁定"这篇文章在用户场景里的具体对象"，再逐讲落点
⑤ 收官 实操 → 测验卷 → 讲错题回退 → 蒸馏 → 全景图
③ 的硬停：课件落盘并发给用户、点名作业之后，这一轮到此为止——不许在同一条回复里继续讲下一讲，不许自问自答。这是本流程最容易塌的地方。
细节、话术、降级路径见 教学法.md（开工前整篇读）。
四个坑（都是真踩过的）
| 坑 | 对策 |
|---|---|
| 抽象概念讲两轮讲不通 | 上可交互 HTML：能点按钮、看状态变色、三态对比。工艺见 references/README.md |
| 只演示成功，用户无感 | 故意演示失败：写错版、作弊版被防线当场抓住——比十遍解释都强 |
| 作业答了一半卡住不敢往前 | 挂账不阻塞：记进笔记「挂账」继续走，学完统一回收。用户完全没回复则等着，见 ③ 硬停 |
| 复述听着对就放过 | 永远拧紧一次（红线 3） |
逆境降级（本 skill 最容易失手的地方，详见 教学法.md）
| 逆境 | 一句话对策 |
|---|---|
| 用户没给材料 | 先在用户身边找（他的 skill/文档/代码）→ 再去外面找候选让他挑 → 都不行才降级自述，且必须标注无出处 |
| 材料太短（<1500 字） | 按骨架 1:1 切、允许合并；减配：只出基础卷、不做全景图、实操并进最后一讲 |
| 材料太长（整本书/几百页） | 禁止不问就整本切。先列目录问"你想解决什么问题、只学哪几章"，砍到 5-8 讲 |
| 用户场景接不上 | 先写出"这篇文章在他场景里的具体对象是什么"，一句话写不出就问，不许硬贴 |
| 用户拒绝考核 | 降轻量档（判断题/二选一）→ 仍拒就记 ⚠️ 挂账、说清代价、继续；连续 3 讲提醒一次 |
| 新会话说"继续学习" | 走恢复协议：定位 学习计划.md → 读进度+挂账 → 一句话回述 → 往前走 |
文件清单
| 文件 | 用途 |
|---|---|
教学法.md |
操作手册：每步怎么做、拧紧话术、题型库、全部降级路径。开工必读 |
templates/课件模板.html |
每讲课件骨架 + CSS 设计系统 + 组件仓库，整体复制起手 |
templates/测验模板.html |
自动判分测验引擎（单选/多选/排序/动手写）。按头部注释改 4 处，逻辑一行别动 |
templates/资料模板.md |
第①步的结构化笔记模板（素材源，重要性最高） |
templates/学习计划.md |
进度表（含学习目录绝对路径、下一步、为什么这么切） |
templates/笔记模板.md |
每讲笔记（用户原话 + 拧紧记录 + 挂账） |
templates/蒸馏素材清单.md |
边学边攒的素材台账 |
references/README.md |
交互演示的 6 条工艺（做演示前先读这 16 行，别直接啃 HTML） |
references/物证-交互演示.html |
交互演示范例（TDD 红绿灯），照工艺不照内容 |
What ships with it
10 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.
- 11d ago First seen · 103 lines · 364 tokens per session scan A 26f6484c7ec9
article-study is a skill published in the GitHub repository yunshu0909/yunshu_skillshub (757 stars, last pushed 1mo ago), licensed MIT. It adds 364 tokens to every session and 2,596 once invoked, about $0.0018 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
hr-onboarding
A new-hire onboarding plan as a single page — first week schedule, buddy + manager intro, learning track, equipment checklist, and "you're set when…" outcomes. Use when the brief mentions "onboarding", "new hire", "first week plan", or "入职".
book-mirror
Take any book (EPUB/PDF), produce a personalized chapter-by-chapter analysis. Each chapter is preserved in detail (The Chapter) and mirrored back to the reader's actual life (The Mirror) using brain context. The mirror observes and resonates — a friend pointing out parallels, NOT a consultant rearranging the reader's…
miniapp
Build a tiny interactive HTML playground only when someone asks to see, play with, or step through a mechanism.
eli5
Explain research, papers, or technical ideas in plain English with minimal jargon, concrete analogies, and clear takeaways. Use when the user says "ELI5 this", asks for a simple explanation of a paper or research result, wants jargon removed, or asks what something technically dense actually means.
deck-course-module
A course or workshop slide template with persistent learning goals, teaching pages, multiple-choice self-tests, and a wrap-up.
master-yinguang
A reference-based assistant for questions about Yinguang and Pure Land Buddhism, a Buddhist tradition focused on faith, ethical living, and practice connected with rebirth in the Pure Land. It can answer in Yinguang’s historical teaching style.