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.
A tightly organized, instruction-only skill that names exact SwiftUI window/scene APIs for each scenario and offloads executable examples to a single well-signaled reference file. Its workflow includes a build-and-verify checkpoint and a review checklist, with only minor redundancy and missing error-recovery feedback.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is directive and assumes SwiftUI/window competence without explaining basics, but the Guardrails section restates several rules already covered in the topical sections (e.g. drag affordance, restoration), which is minor redundancy that could be trimmed. | 4 / 5 |
Actionability | It names exact, copy-ready API signatures per situation (e.g. '.toolbarBackgroundVisibility(.hidden, for: .windowToolbar)', '.defaultWindowPlacement { content, context in ... }', 'content.sizeThatFits(.unspecified)') and routes to references/api-snippets.md for full executable examples; the only gap is that complete code lives one hop away in the reference rather than inline. | 4 / 5 |
Workflow Clarity | A clear 7-step workflow is sequenced with an explicit verify checkpoint (build and launch with macos-build-run-debug) plus a Review Checklist, but there is no explicit if-fails-then-fix feedback loop, leaving a minor validation gap below the 5 anchor. | 4 / 5 |
Progressive Disclosure | SKILL.md is a lean, well-sectioned overview and the single concrete-example bundle is split out behind one clearly signaled one-level-deep reference ('For concrete window modifier examples, read references/api-snippets.md'), which exists and is well organized, matching the 5 anchor. | 5 / 5 |
Total | 17 / 20 Passed |