Content
100%Weight 40%Scale 1-3Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A well-crafted, compact skill body: executable commands with flags, a concrete response example, explicit error-handling guidance, and clean section organization without external-reference overhead. It adds only what Claude would not already know.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean body built from a compact primitives table, copy-paste commands, and a JSON example; it assumes Claude's competence and avoids explaining basic concepts, so every section earns its place. | 3 / 3 |
Actionability | Provides fully executable commands with concrete flags ('bestblogs read deep RAW_xxx --json', '--markdown-only', '--no-mark-read') plus a realistic example response, making the guidance copy-paste ready. | 3 / 3 |
Workflow Clarity | The core is a single unambiguous action whose variants are clearly labeled, and error handling explicitly enumerates 404, private-source, and mark-read-failure states with recovery guidance — satisfying the simple-skill workflow-clarity allowance. | 3 / 3 |
Progressive Disclosure | The skill is self-contained with no bundle files and is organized into clearly labeled sections (何时使用, 原语, 动作, 错误处理, 组合建议) with no nested references, fitting the well-organized-sections criterion for compact single-purpose skills. | 3 / 3 |
Total | 12 / 12 Passed |