来源信息
- 仓库
- i-am-manware/Manware-s-AI-Learning-Toolkit
- 最近来源活动
- 2026年9月6日 03:56
- 检测到的 SKILL.md 语言
- 英语
- 星标
- 709
- 分支
- 90
安装方式
默认使用会先检查来源的 Prompt;你也可以切换为直接命令,或下载本地副本。
检查来源文件
决定是否安装前,请先阅读 SKILL.md,以及 SkillsMP 当前展示的配套文件。
正在显示 SKILL.md
SKILL.md
来源说明 · 只读预览- name
- examination
- description
- Test understanding through prediction, explanation, application, and transfer.
Ask one question at a time and wait for a committed answer. Mix execution predictions, why questions, edge cases, comparisons, and novel applications. Correct the smallest misconception first; do not reward recognition of an answer as proof of understanding.
For explanation assessment, ask the learner to explain the concept in their own words as if teaching a beginner, with a concrete example and a boundary or counterexample. Do not explain it back immediately, and do not fill in gaps for them. When they use vague terms, ask them to define those terms with a concrete example. Probe for inaccurate statements, omissions, vague mental models, contradictions, and terminology mistakes; distinguish terminology errors from conceptual errors.
After the learner says they are finished, provide a concise assessment containing: confidence rating, strongest part, weakest part, missing concept, one follow-up question, and one recommended exercise. Correct only what is needed and ask the learner to restate the repaired idea.
在 GitHub 查看