Content
93%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, highly actionable instruction skill: concrete inputs, named sources, an explicit output template, and sharp quality rules with no wasted tokens. Its only gap is that the multi-step gather→filter→format→output workflow is implied by section ordering rather than stated as an explicit sequence.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Every section earns its place with dense, directive prose ("Suppresses seed noise", "Show 'Filtered out' footer") and zero explanation of concepts Claude already knows; no padding anywhere in the 47-line body. | 5 / 5 |
Actionability | As an instruction-only skill it provides copy-paste-ready guidance: parameter table with defaults and an example value, named sources (SEC Form D, Crunchbase, TechCrunch funding tag), a concrete per-round output template, and rules with a negative example ("'AI is hot' is not a catalyst"). | 5 / 5 |
Workflow Clarity | The sequence (gather from sources → apply filters → fill per-round fields → group by vertical with filtered-out footer) is implied by section order and validation rules exist (the primary-source gate and the "Filtered out" footer), but the steps are never made explicit and there is no error-recovery guidance — matching anchor 4 rather than 5. | 4 / 5 |
Progressive Disclosure | Under 50 lines, single-purpose, no external bundle files and no referenced paths needing verification; the well-organized section headers (Inputs, Sources, Filters, Per-round fields, Output, Rules) satisfy the simple-skill exception for a top score. | 5 / 5 |
Total | 19 / 20 Passed |