Content
100%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.
An exemplary instruction-only skill body: lean, concrete, and immediately actionable, with naming rules, inline examples, and an explicit exceptions section that prevents over-application. It is well organized for its small size and requires no bundle files.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | At roughly 25 lines the body contains zero padding and no explanation of concepts Claude already knows; each rule (name params after domain type, name accumulators after what they accumulate, leave correct uses untouched) adds new, non-obvious information. This matches the 'lean and efficient; every token earns its place' anchor. | 5 / 5 |
Actionability | The guidance is fully concrete for an instruction-only skill: exact naming patterns ('batches.map((batch) => …)', 'sum/total ... not step', 'percentileRank, multiplier') plus an explicit exception list ('(left, right) => left - right', '(column) => column.key over a real .columns array') covering the common cases, satisfying the rubric's note that actionable guidance substitutes for executable code. | 5 / 5 |
Workflow Clarity | This is a single-task skill where the simple-skill exception applies: the one action (apply the naming rules; leave listed correct uses untouched) is unambiguous, and no destructive or batch operations require validation checkpoints. | 5 / 5 |
Progressive Disclosure | The skill is under 50 lines with no need for external references, and the content is organized into clear sections ('Rules' and 'Leave correct uses untouched') with no dangling references, which the rubric's guideline scores as a 5. | 5 / 5 |
Total | 20 / 20 Passed |