Content
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A concise, well-organized procedural skill with a clear graduated workflow and explicit validation checkpoints; its main gap is actionability, since guidance is conceptual rather than executable.
Suggestions
Add concrete procedural mechanics or worked examples (e.g., how to evaluate each of the five testimony types, or a sample amnesty decision) so officials can act rather than just describe.
Make the fine-schedule units explicit (define '率') and show how contested-fine escalation to 五过 is actually carried out.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and well-structured: short labeled sections, a compact fine-schedule table, and lists without padding or explanation of concepts Claude already knows. | 3 / 3 |
Actionability | It gives concrete decision branches and a specific fine schedule, but offers no executable commands, code, or precise procedural mechanics an official could follow; it describes rather than fully instructs. | 2 / 3 |
Workflow Clarity | The graduated process (五刑 → 五罚 → 五过) is clearly sequenced and the Validation section provides explicit checkpoints verifying party presence, graduated order, doubt-amnesty, and fine-rate matching. | 3 / 3 |
Progressive Disclosure | As a short, single-purpose skill (~70 lines) with no external references needed, the well-organized sections satisfy the simple-skill allowance for a top progressive-disclosure score. | 3 / 3 |
Total | 11 / 12 Passed |