Content
67%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 review skill: a quick-reference of thresholds, a concrete Check procedure with specific size limits, a prescriptive Fix playbook, and a single clean pointer to references/rule.md. Its main cost is redundancy — the intro and Explain sections restate trade-offs already covered by rule.md — and it lacks runnable commands or before/after code in the overview itself.
Suggestions
Trim the intro paragraph and the 'Explain' section to one or two lines each and defer the full trade-off discussion to references/rule.md, since both currently duplicate that file's content and explain concepts Claude already knows.
Add one runnable anchor to the body, e.g., the SVGO CLI command (`npx svgo icon.svg`) or a minimal bad/good inline-SVG snippet, so the Fix steps are copy-paste executable rather than descriptive.
Close the workflow loop by making the existing 'Show before/after size comparison' an explicit verification step: re-measure the character count after extracting or sprite-ing and confirm the HTML payload actually shrank.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The opening paragraph ("A complex illustration exported from Figma can easily be 50-200KB... balloons every page response") and the "Explain" section both restate the same trade-offs Claude already knows and that also appear in references/rule.md — noticeable duplication that could be trimmed to a pointer. Not a 2 because the Quick Reference, Check, and Fix sections are tight and actionable with no filler. | 3 / 5 |
Actionability | Concrete, specific guidance throughout: "flag any over 1KB (approximately 1000 characters)", "Keep inline SVGs under 1KB (ideally under 500 bytes) after SVGO optimisation", and named tools (SVGO, SVGR, SVG sprites). Not a 5 because there are no runnable commands or code examples in the body (e.g., an actual SVGO CLI invocation or before/after markup), so the guidance is executable in intent but not copy-paste ready. | 4 / 5 |
Workflow Clarity | The Check → Fix → Explain sequence is clearly laid out with numbered sub-steps for scanning, measuring, checking reuse, and reporting, and "Show before/after size comparison" acts as a verification checkpoint. Not a 5 because there is no explicit validation/feedback loop (e.g., re-measure after extraction to confirm the size reduction) and the steps are compact prose rather than a checklist. | 4 / 5 |
Progressive Disclosure | The body is a concise overview with well-organized sections and a single clearly signaled one-level-deep reference ("For full implementation details, code examples, and framework-specific guidance, see `references/rule.md`" — verified to exist, 171 lines). Not a 5 because the intro paragraph duplicates rule.md's "Why It Matters" text inline, so the split between overview and reference isn't perfectly clean. | 4 / 5 |
Total | 15 / 20 Passed |