Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate 5 creative, cost-effective marketing ideas with channels, messaging, and engagement rationale. Use when brainstorming marketing campaigns, planning product promotion, or looking for creative marketing tactics.
.claude/skills/phuryn-marketing-ideas/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-12 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-05 | ✓→✗ | ▼ Worse | 1% | 0% |
| case-20 | ✓→✓ | ＝ Same ✓ | -4% | 0% |
| case-21 | ✓→✓ | ＝ Same ✓ | 5% | 0% |
| case-23 | ✓→✓ | ＝ Same ✓ | 38% | 0% |
Generate 5 creative, cost-effective marketing ideas with channels, messaging, and engagement rationale. Use when brainstorming marketing campaigns, planning product promotion, or exploring creative marketing approaches.
You are an experienced product marketer specializing in cost-effective growth strategies and creative campaign development.
Analyze the following product and market context: $ARGUMENTS
Generate 5 creative marketing ideas for promoting this product to the target market segment. For each idea:
Prioritize strategies that deliver high impact with limited budget. Consider unconventional approaches and leverage emerging trends where applicable.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 16,698 | 16,721 | +0% | 1 | 1 | 0% | 2,668 | 3,002 | +13% | 0 | 0 | — |
case-02 | fail→fail | 17,179 | 14,010 | -18% | 1 | 1 | 0% | 2,893 | 2,600 | -10% | 0 | 0 | — |
case-03 | fail→fail | 15,436 | 14,088 | -9% | 1 | 1 | 0% | 2,516 | 2,707 | +8% | 0 | 0 | — |
case-04 | fail→fail | 16,837 | 15,433 | -8% | 1 | 1 | 0% | 2,596 | 2,489 | -4% | 0 | 0 | — |
case-05 | pass→fail | 16,171 | 14,200 | -12% | 1 | 1 | 0% | 2,589 | 2,603 | +1% | 0 | 0 | — |
case-06 | fail→fail | 15,110 | 16,040 | +6% | 1 | 1 | 0% | 2,548 | 3,007 | +18% | 0 | 0 | — |
case-22 | fail→fail | 17,038 | 24,863 | +46% | 1 | 1 | 0% | 3,902 | 4,400 | +13% | 0 | 0 | — |
case-07 | fail→fail | 17,123 | 15,264 | -11% | 1 | 1 | 0% | 2,451 | 2,851 | +16% | 0 | 0 | — |
case-08 | fail→fail | 17,275 | 14,840 | -14% | 1 | 1 | 0% | 2,864 | 2,879 | +1% | 0 | 0 | — |
case-09 | fail→fail | 22,268 | 15,258 | -31% | 1 | 1 | 0% | 2,826 | 2,792 | -1% | 0 | 0 | — |
case-10 | fail→fail | 14,849 | 13,460 | -9% | 1 | 1 | 0% | 2,368 | 2,522 | +7% | 0 | 0 | — |
case-11 | fail→fail | 18,509 | 17,919 | -3% | 1 | 1 | 0% | 3,001 | 2,913 | -3% | 0 | 0 | — |
case-12 | fail→pass | 19,810 | 19,143 | -3% | 1 | 1 | 0% | 2,575 | 3,480 | +35% | 0 | 0 | — |
case-13 | fail→fail | 14,557 | 14,340 | -1% | 1 | 1 | 0% | 2,292 | 2,323 | +1% | 0 | 0 | — |
case-14 | fail→fail | 16,444 | 14,779 | -10% | 1 | 1 | 0% | 2,540 | 2,307 | -9% | 0 | 0 | — |
case-15 | fail→fail | 19,335 | 14,787 | -24% | 1 | 1 | 0% | 2,402 | 2,799 | +17% | 0 | 0 | — |
case-16 | fail→fail | 17,004 | 13,532 | -20% | 1 | 1 | 0% | 2,198 | 2,532 | +15% | 0 | 0 | — |
case-17 | fail→fail | 12,682 | 15,757 | +24% | 1 | 1 | 0% | 2,203 | 2,948 | +34% | 0 | 0 | — |
case-18 | fail→fail | 18,031 | 14,300 | -21% | 1 | 1 | 0% | 2,721 | 2,503 | -8% | 0 | 0 | — |
case-19 | fail→fail | 14,653 | 13,792 | -6% | 1 | 1 | 0% | 2,299 | 2,470 | +7% | 0 | 0 | — |
case-20 | pass→pass | 14,771 | 11,679 | -21% | 1 | 1 | 0% | 2,448 | 2,361 | -4% | 0 | 0 | — |
case-21 | pass→pass | 8,025 | 7,335 | -9% | 1 | 1 | 0% | 1,945 | 2,045 | +5% | 0 | 0 | — |
case-23 | pass→pass | 14,840 | 14,910 | +0% | 1 | 1 | 0% | 2,380 | 3,284 | +38% | 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. 3 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.