Content
30%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 reads as a product README rather than an operational skill: it catalogs capabilities by count and name but delivers none of the promised substance — no formulas, no templates, no process. An agent invoking this gains a persona label and little actionable guidance.
Suggestions
Deliver the promised substance: include the actual metric formulas (or RICE/MoSCoW scoring procedures) inline, or move them to reference files (e.g., references/metrics.md, references/templates.md) with clearly signaled links from SKILL.md.
Add a basic workflow for handling a request, e.g., identify the relevant knowledge domain -> select an appropriate framework -> apply the matching template -> compute/verify metrics -> produce the output artifact.
Remove the Compatibility section and the description-restating intro; both are marketing content that consumes context without improving task execution.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is short and sectioned, but the Compatibility section ('Works with Claude Code, Cursor, Windsurf... and 14+ AI coding tools') is marketing filler and the intro restates the frontmatter description, so it could be tightened. | 3 / 5 |
Actionability | It names concrete frameworks (RICE, MoSCoW, Kano) and metrics (MRR, ARR, LTV, CAC) but provides no formulas, template contents, or application steps — 'Calculate 32 SaaS metrics with exact formulas' delivers zero formulas. This is high-level hinting without executable guidance, matching anchor 2 rather than 3 (no concrete instruction at all) or 1 (specific artifacts are at least named). | 2 / 5 |
Workflow Clarity | No sequence exists anywhere in the body; the numbered knowledge-domain list is a taxonomy, not a process, and there is no guidance for how to approach a request despite the skill claiming multi-step processes (roadmapping, launch planning). Anchor 1 ('steps missing; no sequence') fits best since not even a rough sequence is present. | 1 / 5 |
Progressive Disclosure | Sections are clearly organized, but the advertised materials (12 templates, exact metric formulas, framework guides) exist in no bundle file — there is no references/ directory and no internal links, only an external GitHub URL. The simple-skill exception does not apply because the skill claims detailed reference content it does not carry. | 3 / 5 |
Total | 9 / 20 Passed |