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 agentmods add skills/guoqiaozhou/study-with-claude-code/stopnpx skills add guoqiaoZhou/study-with-claude-code --skill stopgit clone --depth 1 https://github.com/guoqiaoZhou/study-with-claude-codeWrote 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/guoqiaozhou/study-with-claude-code/stop)<a href="https://agentmods.dev/skills/guoqiaozhou/study-with-claude-code/stop"><img src="https://agentmods.dev/badge/skills/guoqiaozhou/study-with-claude-code/stop.svg" alt="Measured on agentmods" 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 | $0.00080 | $0.03693 |
| Opus 5 | $0.00040 | $0.01847 |
| Sonnet 5 | $0.00016 | $0.00739 |
| Haiku 4.5 | $0.00008 | $0.00369 |
Grade A, and why
swcc-stop 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 5d 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 — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.
swcc · stop — 结束复习并归档
归档当前这轮对话里刚发生的复习:独立评分、记薄弱点、更新进度、排下次复习。
动手前先读两份契约:
${CLAUDE_PLUGIN_ROOT}/skills/_shared/data-contract.md—— 落盘格式与艾宾浩斯计算。${CLAUDE_PLUGIN_ROOT}/skills/_shared/mastery-rubric.md—— 掌握度评分两维与档位。 本技能会写:新增一份 review-session 记录，并更新 progress.json 与 knowledge-tree.md。
核心原则
- 掌握度由系统独立评估，绝不让用户自评。 用一个没参与对话的子智能体打分，避开「考官心软」偏差。
- 覆盖度 + 深度两维并重。 会背（覆盖广但浅）和钻牛角尖（深但窄）都不算掌握，见 mastery-rubric。
- 进度只增不乱。 严格按 data-contract 更新 progress，艾宾浩斯算 nextReview，状态机别跳。
- 评分要有理由。 review-session 必须写清「覆盖了什么、深度如何、缺什么」。
subagent 纪律
阶段 2 的评估子智能体必须遵守以下纪律，主对话也要同步约束：
- 工具权限：评估子智能体禁止使用 WebSearch / WebFetch；主对话在
stop执行期间也默认不调用 WebSearch / WebFetch。 - 输入来源：只能读取阶段 1 整理的事实清单 +
${CLAUDE_PLUGIN_ROOT}/skills/_shared/mastery-rubric.md；不允许临时联网查询，也不得引用网页、文章或在线资料。 - 输出契约：严格按
mastery-rubric.md第五节返回 JSON，不要附加解释、Markdown 代码块标记之外的内容或外部链接。 - 失败回退：若 Agent 调用失败、超时、返回非 JSON，或输出中出现
http(s)://URL、未提供的网页/文章引用、"according to web/article/online" 等短语，主对话必须丢弃该输出并视为失败；暂停归档，绝不自己打分，提示用户「评估暂不可用，本次未写入，请稍后重试/swcc-stop」，然后停止。
借口对照表(出现这些念头=正在偷懒,照右栏纠正)
| 借口 | 现实 |
|---|---|
| 「用户自评 8 分,就记 8 分」 | 用户自评天然偏高;只作事实输入,分数不采信。掌握度由子智能体按 rubric 客观给。 |
| 「会话很短,我自己打分就行,不必派子智能体」 | 越短越主观。无论长短,阶段 2 必须走 Agent 工具评估。 |
| 「Agent 调用失败,那我自己打吧」 | 禁止回退自评。失败则暂停归档、提示用户重试 /swcc-stop。 |
| 「这概念讲解时用户反应不错,算掌握」 | 只有考核阶段算数;只讲没考的概念不评分、不记薄弱。 |
| 「答错但意思差不多,算对」 | 按 rubric 硬标准,不给人情分。 |
| 「薄弱点抄上次的,不重新评」 | 每次都重评并更新 weakPoints 的 mastery/consecutivePass/nextReview。 |
Red Flags(自查,出现即停下)
- 正准备采信用户报的分数 → 必须派子智能体。
- 正准备把讲解阶段表现计入评分/薄弱点 → 只看考核阶段。
- 正因「会话短」或「Agent 失败」想自己打分 → 派子智能体;失败则暂停归档。
- 给某维度 ≤3 却让 mastery >5 → 违反 rubric 约束。
- 改了 progress.json 却没同步 knowledge-tree 的勾选/🔴 → 必须一致。
常见错误
- 直接采信 go 收尾小结里的分数:小结是 go 自写、带主观。→ 子智能体要基于原始问答重评。
- 两个计数混用:节点
reviewCount(复习次数,用于艾宾浩斯)≠ 薄弱点consecutivePass(连续达标,用于移除)。 - nextReview 基准算错:节点和薄弱点要各用自己的 reviewCount 取间隔。
- 忘记同步 tree:progress 改了,knowledge-tree 的
[x]/🔴 没跟上。
执行流程
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.
- 5d ago First seen · 171 lines · 80 tokens per session scan A 87cca8b1c0cc
swcc-stop is a skill published in the GitHub repository guoqiaoZhou/study-with-claude-code (2 stars, last pushed 2mo ago), licensed MIT. It adds 80 tokens to every session and 3,693 once invoked, about $0.0004 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
textbook-distillation
Turn a textbook or long-form source into a self-paced learning track: intake the material, build a chapter map, draft a lesson plan, then generate self-contained HTML lecture notes in a style the human specifies (layout, palette, emphasis), each lesson carrying worked examples, exercises, and checkpoint questions.…
researchers-historical
Researches archives, contemporary accounts, and timeline reconstruction. Use when the album subject involves historical events that need primary source verification.
Interview Prep Generator
Generate STAR stories, practice questions, and talking points from resume.
obsidian-to-clew-import
Convert an Obsidian vault or wiki-linked markdown graph into a validated structured-learning graph package for Clew. Use when the user wants to inspect a vault, preview whether it imports cleanly, preserve explicit relation markers, choose only the few import settings that matter, and produce a fail-closed package…
interview-simulator
Simulate role-specific mock interviews, score each answer, and provide concrete feedback and model responses for interview preparation.
Automation Interview Prep
Prepare for SDET and automation interviews round by round, covering coding screens, framework design, API testing tasks, scenario questions, and STAR stories built from real testing work.