Content
42%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 but bloated reference catalog: the trigger-rich description sets up a skill whose body spends most of its tokens defining standard mental models Claude already knows, while the genuinely useful parts (marketing applications, Quick Reference, diagnostic questions) are diluted by the padding. The workflow and file structure leave selection and disclosure to the reader.
Suggestions
Strip the definitional sentence from each model entry (Claude already knows what loss aversion or anchoring is) and keep only the marketing application and any non-obvious usage caveats, cutting the body by roughly half.
Split the catalog into per-category reference files (e.g. references/persuasion.md, references/pricing.md) and keep SKILL.md as a lean overview with the Quick Reference table and diagnostic questions.
Add guidance for choosing between the overlapping behavior frameworks (AIDA, BJ Fogg, EAST, COM-B) and a short worked example showing the full process from diagnostic questions to recommended models to ethical implementation.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~430-line body extensively re-explains concepts Claude already knows (e.g. 'Losses feel roughly twice as painful as equivalent gains feel good,' 'The first number people see heavily influences subsequent judgments'), with roughly half of each model entry being a redundant definition rather than marketing-specific added value. | 2 / 5 |
Actionability | Concrete guidance exists (the 4-step usage list, per-model 'Marketing application' tactics, the challenge-to-model Quick Reference table, and diagnostic questions), but there is no worked end-to-end example and competing frameworks (BJ Fogg, EAST, COM-B, AIDA) are presented with no guidance on when to choose which. | 3 / 5 |
Workflow Clarity | A sequence is present ('1. Identify which mental models apply... 2. Explain the psychology... 3. Provide specific marketing applications... 4. Suggest how to implement ethically') and the Quick Reference and Task-Specific Questions act as implicit checkpoints, but how to actually select among 60+ models or the five behavior frameworks is left to inference with no validation step. | 3 / 5 |
Progressive Disclosure | Internal structure is good (clear category sections, a Quick Reference table, Related Skills), but the entire 60-model catalog is inlined in one monolithic SKILL.md with no bundle files, where per-category reference files would keep the overview lean and reduce context load. | 3 / 5 |
Total | 11 / 20 Passed |