Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →When the user wants to plan a product launch, execute launch channels, or create a launch checklist. Also use when the user mentions "product launch," "launch strategy," "product announcement," "launch channels," or "market launch." For GTM motion and positioning, use gtm-strategy. For cold start and first users, use cold-start-strategy. For Product Hunt day-of, use product-hunt-launch.
.claude/skills/mkurman-product-launch/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 23% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 44% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 38% | 0% |
---|-----|--------| | PR | Press release, media relations | public-relations | | Paid ads | Scale acquisition post-PMF | paid-ads-strategy | | Organic | SEO, content, community | seo-strategy, content-marketing | | Email | Announcement to existing users | email-marketing | | Product Hunt / Directories | Launch-day buzz; early adopters | cold-start-strategy, directory-submission |
| Factor | Guideline | |--------|-----------| | PMF first | Validate product-market fit before scaling; see pmf-strategy | | GTM alignment | One clear story; all teams use same messaging; see gtm-strategy | | Avoid rush | Most failures = scale before PMF |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | fail→pass | 22,522 | 27,229 | +21% | 1 | 1 | 0% | 3,690 | 4,532 | +23% | 0 | 0 | — |
case-01 | fail→fail | 27,660 | 19,118 | -31% | 1 | 1 | 0% | 4,367 | 3,510 | -20% | 0 | 0 | — |
case-03 | fail→pass | 30,716 | 29,018 | -6% | 1 | 1 | 0% | 4,723 | 4,600 | -3% | 0 | 0 | — |
case-04 | pass→pass | 20,002 | 22,301 | +11% | 1 | 1 | 0% | 2,922 | 3,661 | +25% | 0 | 0 | — |
case-05 | pass→pass | 19,090 | 21,390 | +12% | 1 | 1 | 0% | 2,787 | 3,435 | +23% | 0 | 0 | — |
case-06 | pass→pass | 16,435 | 20,150 | +23% | 1 | 1 | 0% | 2,430 | 3,437 | +41% | 0 | 0 | — |
case-07 | fail→pass | 15,166 | 14,206 | -6% | 1 | 1 | 0% | 2,276 | 2,543 | +12% | 0 | 0 | — |
case-08 | fail→fail | 14,751 | 16,839 | +14% | 1 | 1 | 0% | 2,281 | 3,007 | +32% | 0 | 0 | — |
case-09 | fail→pass | 15,956 | 20,183 | +26% | 1 | 1 | 0% | 2,311 | 3,334 | +44% | 0 | 0 | — |
case-10 | fail→fail | 17,407 | 23,857 | +37% | 1 | 1 | 0% | 2,664 | 4,152 | +56% | 0 | 0 | — |
case-11 | fail→pass | 12,461 | 14,322 | +15% | 1 | 1 | 0% | 1,775 | 2,441 | +38% | 0 | 0 | — |
case-12 | pass→pass | 13,200 | 15,117 | +15% | 1 | 1 | 0% | 1,974 | 2,607 | +32% | 0 | 0 | — |
case-13 | fail→fail | 12,303 | 13,807 | +12% | 1 | 1 | 0% | 1,807 | 2,459 | +36% | 0 | 0 | — |
case-14 | fail→fail | 13,341 | 17,923 | +34% | 1 | 1 | 0% | 2,113 | 3,146 | +49% | 0 | 0 | — |
case-15 | fail→pass | 16,197 | 19,560 | +21% | 1 | 1 | 0% | 2,352 | 3,422 | +45% | 0 | 0 | — |
case-16 | fail→fail | 17,945 | 37,542 | +109% | 1 | 1 | 0% | 2,584 | 3,464 | +34% | 0 | 0 | — |
case-17 | fail→pass | 14,978 | 20,405 | +36% | 1 | 1 | 0% | 2,283 | 3,309 | +45% | 0 | 0 | — |
case-18 | pass→pass | 8,439 | 7,201 | -15% | 1 | 1 | 0% | 1,300 | 1,476 | +14% | 0 | 0 | — |
case-19 | fail→fail | 17,011 | 19,587 | +15% | 1 | 1 | 0% | 2,480 | 3,327 | +34% | 0 | 0 | — |
case-20 | pass→fail | 17,089 | 19,942 | +17% | 1 | 1 | 0% | 2,518 | 3,172 | +26% | 0 | 0 | — |
case-21 | pass→fail | 14,396 | 12,940 | -10% | 1 | 1 | 0% | 2,035 | 2,289 | +12% | 0 | 0 | — |
case-22 | fail→fail | 19,334 | 18,540 | -4% | 1 | 1 | 0% | 2,796 | 3,271 | +17% | 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. 22 cases were attempted. The headline lift of +23 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
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.