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.
The body is a lean, well-organized overview with exemplary progressive disclosure, correctly pushing the heavy code examples into references/rule.md. Its weaknesses are vague Check/Fix/Explain sections that instruct without specifics and a Quick Reference that lists tips rather than a sequenced testing workflow.
Suggestions
Replace the abstract Check/Fix text with the concrete essentials inline, e.g. the WebPageTest URL, a minimal test configuration (location, browser, connection, 3+ runs), and the Core Web Vitals thresholds (LCP < 2.5s, CLS < 0.1) currently buried in references/rule.md.
Order the Quick Reference into an explicit sequence: configure the test → run 3+ tests from a user-proximate location → review filmstrip for visual progress → inspect the waterfall for blocking resources and slow third-parties.
Drop or merge the redundant Check section (it restates the title/intro) to tighten token usage.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~30-line body is lean, assumes Claude's competence, and contains no concept tutorials; the intro sentence and the Check section both restate "analyze page performance", which is minor trimmable redundancy. Efficient with minor over-explanation fits anchor 4 rather than 5, where every token earns its place. | 4 / 5 |
Actionability | Quick Reference gives concrete hints ("Run 3+ tests", "Use filmstrip view to identify visual progress", "Check waterfall for blocking resources and slow third-parties") but the Check/Fix/Explain sections are abstract ("Implement recommended optimizations from WebPageTest results") with no executable commands or metric thresholds in the body. Some concrete but incomplete guidance matches anchor 3, not 4 (no executable code/commands) and not 2 (the hints exceed minimal). | 3 / 5 |
Workflow Clarity | The Quick Reference bullets are unordered tips rather than a sequenced procedure; there is no explicit run-test → read-metrics → flag-issues order or checkpoints, though as a read-only analysis skill the destructive-operation cap does not apply. Steps listed with implicit sequence and no checkpoints fits anchor 3 rather than 4 (no clear sequence present). | 3 / 5 |
Progressive Disclosure | The body is a short, well-sectioned overview that defers the 1073-line bulk of code examples and framework guidance to a clearly signaled, verified one-level-deep reference ("For full implementation details, code examples, and framework-specific guidance, see `references/rule.md`"). This matches anchor 5: clear overview, well-signaled single-level reference, easy navigation. | 5 / 5 |
Total | 15 / 20 Passed |