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.
The body is a lean, well-organized instruction-only skill that defines its own vocabulary, gives a testable definition of done, and specifies the workspace layout concretely. Its only soft spots are reliance on an only-parenthetically-defined /grilling discipline and a process the reader must assemble from sections rather than follow as steps.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~26-line body is lean and teaches only its own invented vocabulary ("Push right: defer the checkpoint as far as it will go", "The user reads a brief, not a draft"); it explains nothing Claude already knows and every sentence carries instruction. | 5 / 5 |
Actionability | Concrete, specific guidance throughout — "a round of questions at a time, a recommended answer attached to each", "Create, edit, and delete specs as the grilling resolves things", "interview them about their world" when NOTES.md is thin — with one minor gap: the "/grilling" discipline it depends on is only characterized parenthetically rather than spelled out. | 4 / 5 |
Workflow Clarity | The iterative process is legible (grill in rounds, resolve, create/edit/delete specs) with an explicit validation checkpoint — "done when an implementer agent could build it without asking a single question. Grill until then; nothing is done while a question remains" — but the sequence must be assembled across sections rather than laid out as steps. | 4 / 5 |
Progressive Disclosure | Under 50 lines with no need for external reference files; content is split into well-organized sections (The loop lens, Vocabulary, Definition of done, The workspace) and the only cross-skill reference ("/grilling") is clearly signaled one level deep, satisfying the simple-skill exception. | 5 / 5 |
Total | 18 / 20 Passed |