Content
87%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.
The body is concise, highly actionable, and well-structured for a single-file instruction skill. Its only gap is that fact-verification and exit signals are stated as rules rather than as explicit ordered checkpoints in the writing workflow.
Suggestions
Promote the fact-verification rule ('只用真实信息。查不到的直接标注') into an explicit ordered step within the report-generation sequence, e.g. a '验证' checkpoint before the '换不换' conclusion, to make the validation loop a first-class workflow step rather than a standing rule.
Add a brief inline example of a finished report skeleton (one short filled-in instance of the 一–五 sections) so the abstract table schema becomes a concrete, copy-ready template.
Make the per-assumption exit-signal mechanism explicit as a step: after writing assumptions, attach each to a named '退出信号' field with an example, ensuring the verification feedback loop is unmissable.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean instruction prose that assumes competence — it never explains what an investment report is, and every line instructs (sections, banned words, naming convention, output tool) without padding. | 3 / 3 |
Actionability | Provides concrete, specific guidance: a five-section schema with named table fields, a banned-word list, a literal file-naming example, and an explicit output tool/path. For an instruction-only skill this is copy-ready and actionable. | 3 / 3 |
Workflow Clarity | The five-section sequence (一–五) plus 最后一句 and 输出/生成规则 is clearly ordered, but verification is present as a standing rule ('查不到的直接标注，不硬撑，不编') rather than as an explicit checkpoint embedded in the sequence. | 2 / 3 |
Progressive Disclosure | A single self-contained file with no bundle references, organized into clearly labeled sections — appropriate for a standalone instruction skill that needs no external files. | 3 / 3 |
Total | 11 / 12 Passed |