Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Senior PM agent with 6 knowledge domains, 30+ frameworks, 12 templates, and 32 SaaS metrics with formulas. Pure Markdown, zero scripts.
.claude/skills/sickn33-product-manager/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -49% | 0% |
| case-12 | ✓→✗ | ▼ Worse | 41% | 0% |
| case-11 | ✓→✓ | ＝ Same ✓ | 71% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 31% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 46% | 0% |
You are a Senior Product Manager agent with deep expertise across 6 knowledge domains. You apply 30+ proven PM frameworks, use 12 ready-made templates, and calculate 32 SaaS metrics with exact formulas.
Apply frameworks including RICE scoring, MoSCoW prioritization, Jobs-to-be-Done, Kano Model, Opportunity Solution Trees, North Star Metric, Impact Mapping, Story Mapping, and 20+ more.
Use 12 built-in templates for PRDs, one-pagers, retrospectives, competitive analysis, launch checklists, and more.
Calculate 32 SaaS metrics with exact formulas: MRR, ARR, Churn Rate, LTV, CAC, LTV:CAC Ratio, Net Revenue Retention, Quick Ratio, Rule of 40, Magic Number, and more.
Works with Claude Code, Cursor, Windsurf, OpenAI Codex, Gemini CLI, GitHub Copilot, Antigravity, and 14+ AI coding tools.
GitHub: https://github.com/Digidai/product-manager-skills
User request:
> Prioritize these roadmap candidates with an explicit framework, assumptions, evidence, and a recommended next decision.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-11 | pass→pass | 8,728 | 11,223 | +29% | 1 | 1 | 0% | 1,458 | 2,500 | +71% | 0 | 0 | — |
case-12 | pass→fail | 9,054 | 10,846 | +20% | 1 | 1 | 0% | 1,569 | 2,216 | +41% | 0 | 0 | — |
case-01 | fail→pass | 12,589 | 5,176 | -59% | 1 | 1 | 0% | 2,706 | 1,390 | -49% | 0 | 0 | — |
case-13 | fail→fail | 19,469 | 15,605 | -20% | 1 | 1 | 0% | 3,527 | 3,353 | -5% | 0 | 0 | — |
case-02 | pass→pass | 4,623 | 3,750 | -19% | 1 | 1 | 0% | 924 | 1,211 | +31% | 0 | 0 | — |
case-03 | pass→pass | 4,008 | 4,375 | +9% | 1 | 1 | 0% | 968 | 1,413 | +46% | 0 | 0 | — |
case-04 | pass→pass | 4,663 | 6,424 | +38% | 1 | 1 | 0% | 994 | 1,792 | +80% | 0 | 0 | — |
case-05 | pass→pass | 4,439 | 6,120 | +38% | 1 | 1 | 0% | 944 | 1,717 | +82% | 0 | 0 | — |
case-06 | pass→pass | 5,107 | 3,985 | -22% | 1 | 1 | 0% | 1,072 | 1,341 | +25% | 0 | 0 | — |
case-07 | pass→pass | 3,658 | 6,010 | +64% | 1 | 1 | 0% | 803 | 1,594 | +99% | 0 | 0 | — |
case-08 | pass→pass | 3,986 | 5,063 | +27% | 1 | 1 | 0% | 690 | 1,327 | +92% | 0 | 0 | — |
case-09 | pass→pass | 3,458 | 4,488 | +30% | 1 | 1 | 0% | 649 | 1,226 | +89% | 0 | 0 | — |
case-10 | pass→pass | 11,044 | 9,356 | -15% | 1 | 1 | 0% | 1,930 | 2,116 | +10% | 0 | 0 | — |
case-14 | fail→fail | 10,809 | 10,560 | -2% | 1 | 1 | 0% | 2,344 | 2,497 | +7% | 0 | 0 | — |
case-15 | fail→fail | 22,042 | 14,177 | -36% | 1 | 1 | 0% | 4,002 | 2,929 | -27% | 0 | 0 | — |
case-16 | fail→fail | 26,956 | 21,376 | -21% | 1 | 1 | 0% | 6,162 | 5,443 | -12% | 0 | 0 | — |
case-17 | pass→pass | 11,528 | 11,758 | +2% | 1 | 1 | 0% | 1,990 | 2,798 | +41% | 0 | 0 | — |
case-18 | pass→pass | 10,493 | 9,402 | -10% | 1 | 1 | 0% | 1,822 | 2,126 | +17% | 0 | 0 | — |
case-19 | pass→pass | 3,649 | 5,501 | +51% | 1 | 1 | 0% | 819 | 1,581 | +93% | 0 | 0 | — |
case-20 | pass→pass | 9,927 | 7,537 | -24% | 1 | 1 | 0% | 1,929 | 1,824 | -5% | 0 | 0 | — |
case-21 | pass→pass | 9,496 | 12,135 | +28% | 1 | 1 | 0% | 1,841 | 2,731 | +48% | 0 | 0 | — |
case-22 | pass→pass | 18,648 | 17,682 | -5% | 1 | 1 | 0% | 3,253 | 3,809 | +17% | 0 | 0 | — |
case-23 | pass→pass | 11,257 | 11,633 | +3% | 1 | 1 | 0% | 1,854 | 2,454 | +32% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of 0 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
The publisher has shipped newer versions since this run, so these numbers describe v1, not the version currently listed.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.