Content
86%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 lean, well-structured instruction-only skill: a clear role, explicit inputs, and a fully specified output template with exact table columns. Its only weaknesses are the absence of a worked example or decision criteria (e.g., for Build vs Buy) and no edge-case handling for missing inputs, which leave minor actionability and workflow gaps.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is ~58 lean lines: a one-sentence role ('스타트업 CTO처럼 본다. 기능 욕심을 줄이고, 검증에 필요한 최소 제품을 정의한다.'), a five-item input list, and an output skeleton, with no explanation of concepts Claude already knows and nothing to trim. | 5 / 5 |
Actionability | The output template is concrete and copy-paste ready — exact sections ('MVP 범위', '핵심 사용자 흐름', '12주 개발 마일스톤') and fully specified table columns ('레이어 | 선택 | 이유 | 대안', '주차 | 마일스톤 | 담당 | 검증 기준'). It falls short of 5 because it gives no worked example of a filled roadmap and no decision criteria for entries like 'Build vs Buy:' or what makes a good '검증 기준'. | 4 / 5 |
Workflow Clarity | This is a single, unambiguous generation task (inputs → output format), matching the simple-skill exception, and the output skeleton is internally ordered. Not 5 because there is no guidance for handling missing inputs or for validating the roadmap against the '반드시 검증해야 할 행동' input, leaving minor checkpoint gaps. | 4 / 5 |
Progressive Disclosure | A short, single-purpose skill with no need for external references; the three well-organized sections (역할/입력/출력 형식) carry all content inline with no buried or nested references, which satisfies the under-50-line guideline for a top score. | 5 / 5 |
Total | 18 / 20 Passed |