Content
50%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-organized reference catalog with genuinely concrete marketing applications, but it is a ~450-line monolith: each model re-explains psychology Claude already knows, and none of the category catalogs are split into reference files. The top-level workflow is present but loose, with no guidance on selecting or combining models.
Suggestions
Trim or drop the generic one-line definitions under each model heading (e.g., 'People prefer things they've seen before') — Claude already knows these concepts; keep only the marketing application and any counterintuitive framing, which would cut the body roughly in half.
Move the per-category model catalogs (Foundational, Buyers, Persuasion, Pricing, Design, Growth) into references/ files (e.g., references/persuasion.md) and keep SKILL.md as an overview with the Quick Reference table linking to each category — one level deep, clearly signaled.
Strengthen 'How to Use This Skill' into a concrete sequence: check product-marketing context → use the Quick Reference table to shortlist models by challenge → apply the Task-Specific Questions as an explicit checkpoint before recommending.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Nearly every one of the 70+ models opens with a generic definition Claude already knows (e.g., 'People prefer things they've seen before. Familiarity breeds liking,' 'Losses feel roughly twice as painful as equivalent gains') before the marketing application. The padding is systematic across ~450 lines, not occasional — matching 'Noticeably verbose; several unnecessary explanations or padded sections' rather than the mostly-efficient anchor 3. | 2 / 5 |
Actionability | The '**Marketing application**' lines are concrete and specific ('$80 product: "20% off" beats "$16 off",' 'Pre-select the plan you want customers to choose'), and the Quick Reference table maps challenges to models. For an instruction-only skill the guidance is mostly actionable; not 5 because the top-level process ('Identify which mental models apply') is direction rather than a concrete procedure. | 4 / 5 |
Workflow Clarity | A loose four-step sequence is present ('Identify… Explain… Provide… Suggest…') plus task questions, but there are no checkpoints for model selection or synthesis — sequence present, checkpoints implicit. Not 4 because the steps are not concretely sequenced with checkpoints; this is a knowledge skill, so the destructive-operation cap does not apply. | 3 / 5 |
Progressive Disclosure | The body has real structure (seven category sections, a Quick Reference table), but the entire 70-model catalog is inlined in SKILL.md with no references/ directory — content that clearly belongs in per-category reference files is inline. Not 4 because nothing is split out into bundle files; not 2 because the structure and navigation aids are genuine. | 3 / 5 |
Total | 12 / 20 Passed |