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, feature announcement, or release strategy. Also use when the user mentions 'launch,' 'Product Hunt,' 'feature release,' 'announcement,' 'go-to-market,' 'beta launch,' 'early access,' 'waitlist,' or 'product update.' This skill covers phased launches, channel strategy, and ongoing launch momentum.
.claude/skills/dokhacgiakhoa-launch-strategy/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 40% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 32% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 40% | 0% |
You are an expert in SaaS product launches and feature announcements. Your goal is to help users plan launches that build momentum, capture attention, and convert interest into users.
The best companies don't just launch once—they launch again and again. Every new feature, improvement, and update is an opportunity to capture attention and engage your audience.
A strong launch isn't about a single moment. It's about:
Structure your launch marketing across three channel types. Everything should ultimately lead back to owned channels.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 27,447 | 31,989 | +17% | 1 | 1 | 0% | 4,590 | 4,868 | +6% | 0 | 0 | — |
case-02 | fail→pass | 14,499 | 23,133 | +60% | 1 | 1 | 0% | 2,790 | 3,904 | +40% | 0 | 0 | — |
case-03 | fail→pass | 21,735 | 21,925 | +1% | 1 | 1 | 0% | 3,106 | 4,087 | +32% | 0 | 0 | — |
case-04 | fail→pass | 16,589 | 12,303 | -26% | 1 | 1 | 0% | 2,629 | 2,558 | -3% | 0 | 0 | — |
case-05 | fail→fail | 15,243 | 15,410 | +1% | 1 | 1 | 0% | 2,356 | 3,052 | +30% | 0 | 0 | — |
case-06 | pass→pass | 16,667 | 15,391 | -8% | 1 | 1 | 0% | 2,645 | 3,153 | +19% | 0 | 0 | — |
case-07 | fail→pass | 14,258 | 16,326 | +15% | 1 | 1 | 0% | 2,291 | 3,206 | +40% | 0 | 0 | — |
case-08 | pass→pass | 13,268 | 13,853 | +4% | 1 | 1 | 0% | 2,094 | 3,032 | +45% | 0 | 0 | — |
case-09 | fail→pass | 17,111 | 18,791 | +10% | 1 | 1 | 0% | 2,795 | 3,233 | +16% | 0 | 0 | — |
case-10 | pass→pass | 14,867 | 13,119 | -12% | 1 | 1 | 0% | 2,092 | 2,368 | +13% | 0 | 0 | — |
case-11 | pass→pass | 18,580 | 18,614 | +0% | 1 | 1 | 0% | 2,614 | 3,600 | +38% | 0 | 0 | — |
case-12 | fail→pass | 16,613 | 16,416 | -1% | 1 | 1 | 0% | 2,485 | 3,181 | +28% | 0 | 0 | — |
case-13 | pass→pass | 16,289 | 9,475 | -42% | 1 | 1 | 0% | 2,137 | 2,105 | -1% | 0 | 0 | — |
case-14 | fail→pass | 14,704 | 19,339 | +32% | 1 | 1 | 0% | 2,021 | 2,239 | +11% | 0 | 0 | — |
case-15 | pass→pass | 15,084 | 14,419 | -4% | 1 | 1 | 0% | 2,071 | 2,449 | +18% | 0 | 0 | — |
case-16 | fail→pass | 15,652 | 12,446 | -20% | 1 | 1 | 0% | 2,336 | 2,402 | +3% | 0 | 0 | — |
case-17 | pass→pass | 11,427 | 9,837 | -14% | 1 | 1 | 0% | 1,902 | 1,996 | +5% | 0 | 0 | — |
case-18 | pass→pass | 9,169 | 9,614 | +5% | 1 | 1 | 0% | 1,426 | 2,057 | +44% | 0 | 0 | — |
case-19 | pass→pass | 16,155 | 14,839 | -8% | 1 | 1 | 0% | 2,298 | 2,639 | +15% | 0 | 0 | — |
case-20 | pass→pass | 19,764 | 21,447 | +9% | 1 | 1 | 0% | 2,782 | 3,629 | +30% | 0 | 0 | — |
case-21 | pass→pass | 15,967 | 16,021 | +0% | 1 | 1 | 0% | 2,870 | 3,161 | +10% | 0 | 0 | — |
case-22 | pass→pass | 18,479 | 21,743 | +18% | 1 | 1 | 0% | 3,304 | 3,983 | +21% | 0 | 0 | — |
case-23 | pass→pass | 19,162 | 19,561 | +2% | 1 | 1 | 0% | 2,732 | 3,268 | +20% | 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 +39 percentage points is the difference between those two pass rates over the 23 comparable cases.
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.