Content
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, actionable skill with a clear workflow and excellent examples. Its main weakness is length — several reference sections (anti-patterns, inclusive language, startup framing) could be extracted into separate files to reduce token cost. The content is well-organized but slightly verbose in areas where Claude's existing knowledge could be leveraged more.
Suggestions
Extract the Frameworks & Best Practices subsections (anti-patterns, inclusive language, compensation transparency, startup framing) into a separate reference file and link to it, keeping only the HERO structure and a brief summary in the main skill.
Remove explanatory rationale that Claude already knows (e.g., 'research shows this matters for women and minority applicants', 'Listings with ranges get significantly more applicants') and keep only the actionable instruction.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably well-written but includes some content Claude already knows (e.g., explaining why gendered language is bad, why salary ranges matter). The anti-patterns section and inclusive language guidelines, while useful, could be tightened. The HERO framework acronym explanation is somewhat redundant given the workflow already covers the same ground. | 2 / 3 |
Actionability | The skill provides highly concrete, actionable guidance: specific section structures with bullet counts (6-8 bullets for What You'll Do), word count targets (400-700 words), specific anti-patterns with concrete fixes, and two complete example prompt/output pairs showing exactly what good output looks like. | 3 / 3 |
Workflow Clarity | The 7-step workflow is clearly sequenced from role clarification through final polish, with an explicit anti-pattern check step (step 4) serving as a validation checkpoint. The process logically flows from understanding the need to producing polished output, and includes a review step for comp and inclusivity before final polish. | 3 / 3 |
Progressive Disclosure | The content is well-structured with clear sections and headers, but it's quite long for a single file. The frameworks, anti-patterns, inclusive language guidelines, compensation transparency, and startup-specific framing sections could be split into referenced files. The related skills section is a nice touch but the main file itself is monolithic at ~150+ lines. | 2 / 3 |
Total | 10 / 12 Passed |