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 an exceptionally lean, self-contained routing procedure that names concrete tools and skills with explicit routing criteria and a termination loop. Its only weaknesses are interpretive decision criteria without worked examples and no verification step for the final work order, which keep actionability and workflow clarity at 4 rather than 5.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The 16-line body is lean with zero padding: every sentence instructs ("Look up what the record, the repo, or the world can answer", "Route again until nothing is left to invent") and nothing explains concepts Claude already knows. Matches anchor 5 ('Lean and efficient; assumes Claude's competence; every token earns its place') with no over-explanation to trim as anchor 4 would require. | 5 / 5 |
Actionability | The guidance is mostly executable for an instruction-only skill: it names the exact tool ("invoking the named skill with the Skill tool", "AskUserQuestion") and each routing rule maps a specific condition to a named skill (clarify, elicit, explain, interrogate, reframe, deepen, shape). Not 5 because the routing conditions and the cost heuristic ("costs less to undo than the question costs to ask") require interpretation with no worked examples; better than 3 since tool invocations and criteria are concrete, not pseudocode. | 4 / 5 |
Workflow Clarity | A clear sequence is present — look up answerable facts, route the rest by kind, re-route until nothing is left to invent, ask open questions, state the work order — with an explicit termination condition ("Route again until nothing is left to invent") and a decision rule for turning gaps into assumptions. No destructive or batch operations, so the validation cap does not apply; not 5 because there is no verification step for the final work order and some step ordering is implicit. | 4 / 5 |
Progressive Disclosure | The skill is self-contained at ~16 lines with no bundle files (references/, scripts/, assets/ are absent) and no need for external references; the routing bullet list plus short closing procedure is well-organized for this size. Per the rubric's simple-skill note, an under-50-line skill with no need for external references scores 5 with well-organized sections. | 5 / 5 |
Total | 18 / 20 Passed |