Conciseness

Individual sentences are lean, but the 'Check', 'Fix', and 'Code Review' sections restate the same parse → verify @context/@type → check required properties procedure nearly verbatim, and the 'Explain' section describes concepts Claude already knows ("enhanced search result formats that include stars, images, expandable Q&s"). Fits anchor 3 (mostly efficient, could be tightened) better than 2 because the fix is merging redundant sections, not cutting padded prose.

Actionability

Concrete, executable guidance throughout: exact numbered steps, per-type required properties ("Article: headline, author… datePublished"), JSON.parse() for syntax checking, and specific validation URLs (search.google.com/test/rich-results, jsonlint.com). As an instruction-only skill the absence of inline code is not penalized; 'Run through Google's Rich Results Test API if available' is the one under-specified step, keeping it below anchor 5.

Workflow Clarity

Both 'Check' (steps 1–5) and 'Fix' (steps 1–6) present clear numbered sequences with validation checkpoints (Rich Results Test before deploying). The fix→re-validate feedback loop is implied rather than explicitly stated as a retry cycle, which matches anchor 4 rather than 5. This is a read/validate task, not a destructive or batch operation, so the cap-at-3 rule does not apply.

Progressive Disclosure

The body is short (~40 lines) and well-sectioned, and closes with a clearly signaled, one-level-deep reference — "For full implementation details, code examples, and framework-specific guidance, see `references/rule.md`" — and the referenced file exists in the bundle (references/rule.md, 159 lines with valid/invalid JSON-LD examples). Matches the clear-overview-with-well-signaled-references anchor.

Specificity

Names the JSON-LD domain and several concrete actions ("generating or auditing any `<script type='application/ld+json'>` blocks", "implementing structured data for articles, products, FAQs, breadcrumbs, or local businesses", "investigating why rich results are not appearing") across five content types. Scored 4 rather than 5 because coverage has minor gaps — e.g., validating/fixing existing blocks is implied by 'auditing' but not explicitly listed.

Completeness

An explicit "Use when…" clause is present with concrete trigger phrases, so completeness is not capped at 3. The 'what' (generate/audit/implement JSON-LD structured data) is stated but woven into the when-clause rather than clearly standalone, so it falls short of anchor 5's 'clearly and explicitly answers both what AND when'.

Trigger Term Quality

Good coverage of natural terms users would say: "structured data", "rich results", "Google Search Console", "JSON-LD", and the literal script tag. A few common synonyms are missing — notably "schema markup" and "schema.org", which users frequently say when they need this skill.

Distinctiveness Conflict Risk

Clear niche (JSON-LD structured data for SEO) with highly distinct triggers — the literal `<script type='application/ld+json'>` tag, 'rich results', and 'Google Search Console' — making conflict with other skills minimal.