Content
82%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 strong instruction-only skill body: lean, free of padding, and highly actionable with named tools, concrete query templates, explicit write-back steps, and firm anti-hallucination guardrails. The remaining gaps are the absence of an example outreach draft and any validate/verify checkpoint before delivering the report, plus optional splitting of longer catalogs into reference files.
Suggestions
Include one short example outreach message per angle (resource inclusion, article update, comparison mention) to make the 'Outreach drafts' deliverable fully concrete.
Add an explicit verification step before delivery, e.g. confirming every listed contact path has a recorded source URL and that no prospect is a direct competitor.
Consider moving the query-pattern catalog and output-format spec into a single reference file to keep SKILL.md a tighter overview.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and operational throughout — it never explains concepts Claude already knows (no definition of backlinks, SERPs, or outreach), and sections like tool lists, query patterns, and guardrails carry only actionable content. This matches 'assumes Claude's competence; every token earns its place' rather than 4, which requires identifiable over-explanation to trim. | 5 / 5 |
Actionability | Guidance is mostly executable: named MCP tools with what each is for, copy-paste-ready query patterns ("<topic> resources", "best <category> tools"), concrete write-back syntax ({ appendResearchLog: { summary: ... } }), and constraints like "at most 10 queries per call". It falls short of 5 because drafting outreach is a core deliverable yet no example message or template is provided — a minor gap matching anchor 4. | 4 / 5 |
Workflow Clarity | The 9-step workflow is clearly sequenced, with conditional handling ("Continue without backlink evidence if it is unavailable", local-SEO branch, prospect filtering criteria) and verification-style guardrails ("Only record contact details that were actually found"). It matches anchor 4 rather than 5 because there is no explicit validate-then-fix feedback loop or final report verification step; it is not capped at 3 since the workflow is neither destructive nor batch-mutation based. | 4 / 5 |
Progressive Disclosure | The skill is a single well-organized file with clear section headers, and the actual bundle contains no reference files (all cross-references point to sibling skills like seo-report, not nested documents), so navigation is easy and nothing is buried. It matches anchor 4 rather than 5 because at ~110 lines, content such as the query-pattern catalog and output-format spec could be split into one-level-deep reference files to keep SKILL.md a purer overview. | 4 / 5 |
Total | 17 / 20 Passed |