Content
72%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 a well-structured overview: efficient tables, Apple-specific detail instead of general knowledge, and an exemplary one-level-deep reference index pointing to real files. Its weaknesses are the implicit, distributed workflow (context check → questions → reference lookup → output format is never stated as an ordered process) and the useless generated boilerplate "When to Use" line at the end.
Suggestions
Replace the vacuous "When to Use" section ("This skill is applicable to execute the workflow or actions described in the overview.") with real trigger conditions for content display components, or delete it.
State the operating procedure explicitly as a numbered sequence — check the context file, ask the four questions, match needs via the Component Selection Guide, consult the referenced file, then produce output in the given format.
Either supply the "key properties and setup" promised by the Output Format section inline for the most common components, or point each row of the Selection Guide at the specific reference section that contains it.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and information-bearing — the Reference Index and Component Selection Guide are compact tables, and the Key Principles consist of Apple-specific facts (tvOS lockups with parallax, audio graph support for charts) rather than concepts Claude already knows. It is not 5 because of minor padding: the closing boilerplate "This skill is applicable to execute the workflow or actions described in the overview." is pure filler, and the Raintree footer link earns no tokens; not 3 because these are the only blemishes in an otherwise efficient body. | 4 / 5 |
Actionability | The Component Selection Guide gives a concrete need→component→platform mapping with version constraints ("Charts (Swift Charts) | iOS 16+, macOS 13+"), and principles name specific mechanisms (compositional layout, diffable data sources, cell reuse, prefetching). Per the scoring notes, absence of code is not penalized in an instruction-only skill. It is not 5 because the promised "Configuration guidance — key properties and setup" is described as output format rather than given, leaving minor gaps between 'pick this component' and 'configure it'. | 4 / 5 |
Workflow Clarity | A workflow exists but is distributed and implicit: the intro says to check `.claude/apple-design-context.md` first, "Questions to Ask" gathers inputs, and "Output Format" shapes the answer — yet nothing orders these as a sequence, and the section layout (Principles → Reference Index → Selection Guide → Output Format → Questions) does not read procedurally. This matches anchor 3 ('sequence present but checkpoints missing or implicit'); it is not 4 because the operational order must be inferred rather than stated, and not lower because the pieces themselves are individually clear and no destructive/batch validation is required. | 3 / 5 |
Progressive Disclosure | The body is a true overview: an indexed table points to 8 reference files (all verified present in ./references/ — charts.md, collections.md, image-views.md, image-wells.md, color-wells.md, web-views.md, activity-views.md, lockups.md), each one level deep, with topic and key-content columns signaling exactly what each file covers. This matches anchor 5 ('Clear overview with well-signaled one-level-deep references; content appropriately split; easy navigation') with no nesting and no bulk detail inlined in SKILL.md. | 5 / 5 |
Total | 16 / 20 Passed |