Content
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a highly actionable, well-sequenced audit skill with concrete patterns, fixes, and verification checkpoints; its main weaknesses are some preamble padding and a monolithic structure with no progressive disclosure.
Suggestions
Trim the "Tool Use Is Mandatory" section to a one-line directive and consolidate the repeated per-pattern "verify near `var body`" instruction into a single Phase 2 note.
Move the 10 known anti-pattern definitions and/or the Phase 3/4 cross-reference tables into a `references/` file (e.g. `anti-patterns.md`), keeping SKILL.md as an overview that links one level deep.
Hoist the reusable exclude-path glob list into a short variable-style block referenced once instead of restating exclusions inline.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient and well-organized, but the "Tool Use Is Mandatory" preamble and the repeated "verify near `var body`" phrasings across patterns add padding that could be tightened. | 2 / 3 |
Actionability | It gives concrete, copy-paste-ready Grep patterns and specific fix snippets (e.g. `ForEach(items, id: \.id)`, `.onDisappear { timer?.invalidate() }`, `static let`), leaving little ambiguity about what to run or change. | 3 / 3 |
Workflow Clarity | A clearly sequenced five-phase pipeline with an explicit validation checkpoint ("use Read to verify the surrounding context before reporting") and a fully specified output format; being a read-only audit, the destructive-operation cap does not apply. | 3 / 3 |
Progressive Disclosure | Content is well-sectioned into phases but lives entirely in one ~250-line monolithic file with no bundle files or one-level-deep references to split the 10 detection patterns or Phase 3/4 tables; the under-50-line simple-skill exception does not apply. | 2 / 3 |
Total | 10 / 12 Passed |