Content
78%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, concise procedural skill with clear sequencing and a verification section. The main gap is the absence of explicit inter-step validation and error-recovery loops for what is a destructive, batch-scale operation.
Suggestions
Add inter-step validation checkpoints (e.g., 'Confirm the surrendering force is disarmed before entering the camp') rather than only post-hoc verification, to strengthen the destructive-operation feedback loop.
Concretize the credibility-assessment step (what signals distinguish feigned from genuine surrender) to close the main actionability gap.
Tighten the Overview by dropping the Huo Qubing precedent restated from the description to remove minor redundancy.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean with terse bullet steps and no padding of concepts Claude already knows; the Overview's restatement of the Huo Qubing precedent (already in the description) is the main minor redundancy. | 4 / 5 |
Actionability | Steps give concrete domain actions with specifics (e.g., 'Enter the enemy camp with elite cavalry', 'Use official transport (传)', 'Establish as subordinate territories (属国)', '8,000 executed'), leaving only minor gaps such as how to concretely assess credibility. | 4 / 5 |
Workflow Clarity | An explicit eight-step sequence is paired with a Verification section covering credibility, dissenter neutralization, and title conferral; it lacks an explicit error-recovery feedback loop, which keeps it just below a 5. | 4 / 5 |
Progressive Disclosure | At under 50 lines with no need for external references (no bundle files present), the well-organized Overview / Steps / Expected Outcome / Verification sections satisfy the simple-skill exception for a top score. | 5 / 5 |
Total | 17 / 20 Passed |