Content
57%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, consistently formatted catalog of ~50 mental models with genuinely useful marketing applications and a helpful challenge-to-model routing table. Its weaknesses are token inefficiency (re-explaining psychology Claude already knows), an inlined monolithic structure with no reference files, and a generic workflow without checkpoints.
Suggestions
Trim or drop the model-definition sentences (Claude already knows anchoring, loss aversion, Pareto, etc.) and keep only the name plus the 'Marketing application' line, cutting the file roughly in half.
Split the catalog into one-level-deep reference files (e.g. references/buyers.md, references/persuasion.md, references/pricing.md, references/growth.md) and leave a lean overview with the Quick Reference table in SKILL.md.
Make the 4-step workflow concrete with checkpoints, e.g. identify the challenge from the Quick Reference table, propose 2-3 applicable models, state the application and its ethical limit, then confirm which the user wants to pursue.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Each of the ~50 entries is compact, but the opening sentence of nearly every model restates standard psychology Claude already knows ('Losses feel roughly twice as painful as equivalent gains feel good', 'Roughly 80% of results come from 20% of efforts') — roughly half the file's tokens re-teach known concepts. Not 4 because the over-explanation is systematic rather than minor; not 2 because the entries are tightly written and the marketing-application lines do add non-obvious value. | 3 / 5 |
Actionability | Many applications are concretely directive ('Pre-select the plan you want customers to choose', "$80 product: '20% off' beats '$16 off'") and the Quick Reference table maps challenges to models. Not 5 because there is no worked end-to-end example showing the process applied to a real scenario, and some entries stay at the level of tactical description. | 4 / 5 |
Workflow Clarity | A 4-step sequence is listed ('1. Identify which mental models apply ... 4. Suggest how to implement ethically') plus diagnostic questions, but the steps are generic with no checkpoints or decision guidance for when to stop or how to validate the fit. Not 4 because the sequence lacks any explicit checkpoints; not 2 because a coherent process and routing table do exist. | 3 / 5 |
Progressive Disclosure | Section headers and the Quick Reference table give good in-file navigation, but this is a ~450-line monolith with zero bundle files — buyers psychology, pricing psychology, and growth models are natural separate reference files a la 'reference/buyers.md'. Not 4 because the SKILL.md is a full catalog rather than an overview pointing to detailed materials; not 2 because structure is far from minimal and nothing is buried. | 3 / 5 |
Total | 13 / 20 Passed |