Content
85%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A well-structured instructional skill body: the multi-step teaching workflow is explicitly sequenced with validation checkpoints, recovery loops, and an acceptance checklist, and detail is properly split into well-signaled reference files. Minor improvements are possible by consolidating repeated overclaiming warnings and adding one concrete diagnostic-exchange example.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense imperative guidance that offloads theory to references instead of re-explaining it ('需要核对理论归因或解释设计依据时，阅读 理论依据与设计转化'). Warnings against overclaiming recur across several sections and could be consolidated, keeping it below the lean anchor-5 level. | 4 / 5 |
Actionability | Concrete decision rules and a stuck-point→support→learner-action table make the guidance executable ('没注意到关键条件 → 指向条件或给一个方向性问题 → 识别关系并决定下一步'), with specific page-brief requirements. No worked example of a diagnostic exchange is included, which is the minor gap versus fully covering common cases. | 4 / 5 |
Workflow Clarity | The workflow is clearly sequenced (choose mode → organize lesson → independent attempt → responsive support → scaffold removal → record → page landing) with explicit validation steps ('生成后读取真实题目核对题型及整页内容…再在课堂预览中复核', '实测提交、取回、修订和刷新'), recovery loops ('若再次卡住，恢复针对性的帮助或返回一个更小目标'), and a final acceptance checklist — matching the anchor-5 structure. | 5 / 5 |
Progressive Disclosure | Two real, one-level-deep reference files are clearly signaled with when-to-read conditions ([理论依据与设计转化](references/theory.md), [整节习题课的组织方式](references/exercise-lesson.md)); theory detail is appropriately split out of the overview body, giving easy navigation. | 5 / 5 |
Total | 18 / 20 Passed |