Content
65%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is commendably concise and reasonably sequenced for a meta-skill, but its actionability is undercut by deferring the core template/principles to external files that are not bundled, and its references are not clearly signaled or backed by actual bundle files. Workflow steps lack explicit validation checkpoints.
Suggestions
Either inline a minimal SKILL.md template/structure checklist or add a bundled references/ file (e.g. references/skill-template.md) and link to it with a real markdown link so the substantive guidance is available, not just referenced.
Add explicit validation/completion checkpoints to the Iterate workflow, e.g. 'Verify the saved SKILL.md has name, description, and at least one actionable section before finalizing.'
Replace backtick/bold mentions of external resources with clear navigable links (e.g. [skills.md](skills.md)) and confirm each referenced file exists in the bundle.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~28-line body is lean and sectioned (Extract / Clarify / Iterate) with no padding and no explanation of concepts Claude already knows, matching score 3 'Lean and efficient; assumes Claude's competence; every token earns its place'; it is not verbose enough to drop to 2. | 3 / 3 |
Actionability | Concrete guidance exists ("Extract: step-by-step process, decision points, branching logic", specific clarifying questions, numbered iterate steps), but the substantive know-how — the SKILL.md template and principles — is deferred to unbundled external files ("Load and follow skills.md for template and principles"), leaving it incomplete per score 2 'Some concrete guidance but incomplete; missing key details'. | 2 / 3 |
Workflow Clarity | A clear Extract → Clarify → Iterate sequence with a refinement loop ('Identify the most ambiguous or weak parts and ask about those') is present, but there are no explicit validation or completion checkpoints, matching score 2 'Steps listed but validation gaps; sequence present but checkpoints missing or implicit'; a 3 would require explicit verification steps. | 2 / 3 |
Progressive Disclosure | The body is well-organized into sections but references external resources ("skills.md", "agent-customization") via backticks/bold rather than clear navigable links, and no bundle files (references/scripts/assets) exist to back those references, matching score 2 'references present but not clearly signaled'; it is not a monolithic wall (1) nor a clean one-level-deep reference set (3). | 2 / 3 |
Total | 9 / 12 Passed |