Content
78%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 actionable and well-structured with executable bash and a clear sequenced workflow plus a validation checkpoint. Its main weakness is moderate redundancy in presenting the invocation fallback three times.
Suggestions
Collapse the three presentations of the executable/clone invocation into a single if/elif/else block to remove redundancy and tighten conciseness.
Replace the comment-only else branch with a concrete echo/print command so the failure path is fully copy-paste executable.
Add an explicit validation step after the CLI runs (e.g., checking that stdout is non-empty and budget/citations are present) to strengthen the workflow's feedback loop.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and assumes Claude's competence with no conceptual padding, but the executable/clone invocation is presented three times (separate routes then a combined if/elif/else), a minor redundancy that could be tightened. | 4 / 5 |
Actionability | Concrete, copy-paste-ready bash commands cover topic and recent modes with an explicit executable-or-clone fallback; minor gaps are the `<topic>` placeholder requiring substitution and the comment-only else branch lacking an echo command. | 4 / 5 |
Workflow Clarity | A numbered 'Before You Start' sequence, 'Execute', and 'Return' give a clear order with a path-canonicalization validation checkpoint and fallback error guidance, though validation is not exhaustive across every step. | 4 / 5 |
Progressive Disclosure | Well-organized into clear sections with a single well-signaled one-level-deep reference (the Config Resolution Protocol in llm-wiki/SKILL.md) and no nested references; no bundle files are needed for this single-purpose skill. | 5 / 5 |
Total | 17 / 20 Passed |