Content
62%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured strategic/analytical skill with a clear 5-step workflow and good anti-pattern guidance. Its main weaknesses are verbosity (some explanatory framing that could be cut) and lack of concrete worked examples with actual numbers — the skill describes what analysis to do but doesn't show a complete example with sample data. The cross-references to related skills are helpful but the content itself is monolithic.
Suggestions
Add a concrete worked example showing a sample activation funnel with actual numbers (e.g., 10,000 signups → 7,200 complete step 1 → ... → 1,800 activated) to make the sizing and math steps immediately actionable.
Trim the introductory paragraph and in-prompt explanations of what activation is — Claude and experienced PMs already know this; focus on the novel framework and methodology.
Consider splitting the detailed blocker categorization framework and impact modeling templates into a separate reference file to keep the main SKILL.md as a concise overview.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably well-written but includes some unnecessary framing ('You are getting signups but too few become real users') and explanatory text that a PM-focused Claude doesn't need. The prompt template itself is lengthy with explanations of concepts (e.g., explaining what activation is vs. a funnel step) that could be tightened. However, most content does earn its place by providing specific frameworks and math-based reasoning. | 2 / 3 |
Actionability | The skill provides a structured framework with specific categories (value gap, effort gap, etc.) and scenario modeling, which is useful. However, it lacks concrete executable examples — no sample data, no example funnel with actual numbers worked through, no specific SQL/Amplitude queries. The guidance is detailed but remains at the 'describe what to do' level rather than 'here is exactly how to do it with copy-paste artifacts.' | 2 / 3 |
Workflow Clarity | The 5-step workflow is clearly sequenced and logically ordered: define activation → measure funnel → identify blockers → model impact → recommend. Each step has explicit sub-steps and the output format section serves as a validation checklist. The anti-plays section provides guardrails against common mistakes, and the skill includes feedback loops (stress-test the activation definition, validate hypotheses with session replays, check with build-metric-tree first). | 3 / 3 |
Progressive Disclosure | The skill references related skills (build-metric-tree, craft-experiment-design, create-chart) which is good for navigation. However, all content is in a single monolithic file with no bundle files to offload detail into. The prompt template section is very long and could benefit from splitting detailed frameworks (e.g., blocker categorization, impact modeling) into separate reference files. | 2 / 3 |
Total | 9 / 12 Passed |