Content
65%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 clean, well-structured overview that uses progressive disclosure effectively to push implementation detail into references/rule.md. Its weaknesses are the in-body side: no executable examples and no validation checkpoints, with the actionable depth living entirely in the reference file.
Suggestions
Inline one small executable snippet (e.g. a 3-line oklch token definition) so the body is actionable on its own without requiring the reference.
Add an explicit verification checkpoint to the Fix section (e.g. 'Verify perceived lightness is equal across shade steps 300/500/700') rather than leaving validation only in rule.md.
Trim the intro paragraph's explanation of perceptual uniformity, which repeats content already in references/rule.md and restates a concept Claude already knows.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~30-line body is mostly lean with useful Check/Fix/Explain/Code Review sections, but the intro paragraph re-explains perceptual uniformity — a concept Claude already knows — and duplicates rule.md's 'Why It Matters', so a little could be trimmed. | 4 / 5 |
Actionability | Gives actionable directives ('Convert the colour token palette from hsl/hex to oklch', 'Use @media (color-gamut: p3)') but no executable code or concrete token examples in the body itself — all working examples live in references/rule.md. | 3 / 5 |
Workflow Clarity | Check → Fix → Explain → Code Review forms a loose checklist rather than a sequenced process, and the validation/verification steps (which exist in rule.md) are absent from the body, leaving checkpoints implicit. | 3 / 5 |
Progressive Disclosure | A concise overview body with a single, clearly-signaled one-level-deep reference ('see references/rule.md') that is verified to exist, splitting detailed implementation out appropriately. | 5 / 5 |
Total | 15 / 20 Passed |