Content
92%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 an exemplary lean design spec: dense, token-efficient, and fully concrete on dimensions, palette, typography, and card anatomy. The only gap is a handful of underspecified visual details (shadow, avatar gradient, inline SVG icon paths) and the absence of any reference markup the implementer could copy.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is a lean ~28-line spec where every line carries concrete design tokens (hex colors, px radii, font stacks, metric formatting); it never explains concepts Claude already knows and contains no padding. Even anchor 4's 'minor instances of over-explanation' do not apply here. | 5 / 5 |
Actionability | The spec is highly concrete and executable as an instruction-only skill (exact dimensions, hex palette, typography, card anatomy, number formatting like '1.2K / 4.5M'), but a few details remain underspecified: '阴影柔和' (soft shadow) has no value, the avatar 'CSS gradient 占位' gives no gradient, and 'icon 用 inline SVG (X 官方风格)' provides no SVG paths or markup example. Not anchor 5 because there is no copy-paste-ready snippet covering the common case. | 4 / 5 |
Workflow Clarity | This is a simple, single-purpose skill (produce one single-file HTML card) and the single action is unambiguous: canvas, structure, fonts, and color rules fully determine the output, with a recommended default ('配色 dark 推荐, 视频叠加用') resolving the one major branch. Per the simple-skill scoring note, this warrants 5. | 5 / 5 |
Progressive Disclosure | The skill is under 50 lines with no external references needed; content is organized into clearly labeled sections (【意图】【画布】【卡片结构】【字体】【设计细节】) that are easy to navigate. Per the scoring note, well-organized sections alone justify 5 here. | 5 / 5 |
Total | 19 / 20 Passed |