Content
35%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.
The body is an over-long, mostly abstract overview that names frameworks and metrics without giving Claude concrete procedures, formulas, or templates to execute them. Heavy generic padding and duplicated trigger lists waste context that concrete worked examples would better use.
Suggestions
Cut the boilerplate 'Use/Do not use this skill when' sections, the duplicated 'When to Use This Agent' trigger list, and the Quality Standards checklists — these restate the description or obvious best practices.
Replace abstract capability lists with concrete, executable guidance: actual TAM/SAM/SOM formulas, unit-economics calculations, or a worked financial-model template.
Move the Knowledge Base, Example Interactions, and Special Considerations into one-level-deep reference files (e.g., references/market-sizing.md) and link them from a lean overview.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~315-line body is noticeably padded: generic boilerplate ('Working on startup analyst tasks or workflows'), restated trigger lists, Quality Standards checklists ('Use credible, cited data sources'), and Behavioral Traits that state best practices Claude already knows. | 2 / 5 |
Actionability | Guidance stays at the naming level ('TAM/SAM/SOM calculations using bottom-up and top-down methodologies', 'Apply Porter's Five Forces') without concrete formulas, templates, calculation steps, or worked examples, leaving the actual execution underspecified. | 2 / 5 |
Workflow Clarity | The 10-step Response Approach provides a clear sequence including a validate step (step 6), but checkpoints are implicit and generic rather than tied to specific decision gates or error-recovery feedback loops. | 3 / 5 |
Progressive Disclosure | No bundle files exist and the content is a single monolithic file; section headers provide some structure, but the Knowledge Base, Example Interactions, and Special Considerations blocks are inlined where separate reference files would aid navigation. | 3 / 5 |
Total | 10 / 20 Passed |