Content
82%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 tight, information-dense skill body: executable CLI commands, explicit workflow rules with user-confirmation checkpoints, and concrete per-medium analysis dimension lists. The only gaps are the missing `update` command example, no sample output format, and the absence of section headers.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and lean — every sentence carries operational rules ("统计 scaffold 不能替代 Agent 回读素材后的定性结论", "少量样本不得称为稳定风格") with zero padding and no explanation of concepts Claude already knows. | 5 / 5 |
Actionability | It opens with copy-paste executable commands with real arguments (`tiktok-style-profiler report --input ... --sample-id ...`) and gives concrete dimension checklists for video and image/text, but the `update` command is never shown and report/DNA output formats are described only abstractly, keeping it short of fully executable. | 4 / 5 |
Workflow Clarity | The report → build → update pipeline is clearly sequenced in prose with real checkpoints ("DNA 调整经用户确认后写入", "update 合并报告但仍需重新检查结论与模板"), but steps are not enumerated and error-recovery guidance is absent, so it sits at 'clear sequence with most checkpoints' rather than 5. | 4 / 5 |
Progressive Disclosure | No bundle files exist and the body is under 20 lines, so no external references are needed; delegating full parameter details to `--help` is appropriate. Content is logically ordered (commands → workflow → dimension lists) but the single H1 has no section headers to separate these blocks, a minor organization gap. | 4 / 5 |
Total | 17 / 20 Passed |