Content
75%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 well-executed instruction-only skill: terse, specific, rule-driven guidance with a priority system and a real verification gate (pnpm commands plus GO/NO-GO reporting). The main gaps are unquantified thresholds in a few rules, a duplicated tool-boundary constraint, and the absence of a single explicit step-by-step workflow section.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is a lean checklist that assumes Claude's competence — no explaining what SEO or Open Graph is — with only minor trims available: the framework/library constraint is stated twice ("Do not introduce new frameworks or SEO libraries unless explicitly requested" and "do not migrate frameworks or SEO libraries unless requested") and the "Source: ui-skills.com." line adds no guidance. Not 5 because those redundant tokens exist. | 4 / 5 |
Actionability | Rules are concrete and checkable ("og:url must match the canonical URL", "set twitter:card appropriately, summary_large_image by default") and the Verify section gives executable commands (pnpm lint/test/build/verify) with a defined GO/NO-GO outcome. Not 5 because several directives are unquantified — "keep titles short and readable", "prefer correct image dimensions" with no pixel values, "a sensible og:type". | 4 / 5 |
Workflow Clarity | A coherent review sequence is present: a defined report format (violation snippet + why + fix), explicit prioritization ("fix critical issues first (duplicates, canonical, indexing)"), cross-field consistency checks, and a final validation step with commands and a GO/NO-GO gate. Not 5 because the sequence is distributed across sections rather than presented as one explicit ordered workflow. | 4 / 5 |
Progressive Disclosure | No bundle files exist, and the flat SKILL.md is well organized: a priority table serves as an overview, category sections mirror it, and everything sits at one navigable level. Not 5 because at roughly 115 lines it exceeds the under-50-line simple-skill exception, and no quick-start/summary distills the highest-priority rules. | 4 / 5 |
Total | 16 / 20 Passed |