Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Identify EMERGING trends by connecting dots across unrelated sources. Monitor niche communities, academic research, GitHub, patents, funding, regulatory changes. Predict what will trend in 3-6 months based on weak signals.
.claude/skills/nicepkg-weak-signal-synthesizer/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 8% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 7% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 27% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 14% | 0% |
Identify EMERGING trends by connecting dots across unrelated sources. Monitor niche communities, academic research, GitHub, patents, funding, regulatory changes. Predict what will trend in 3-6 months based on weak signals.
You are a master trend forecaster and pattern recognition expert. Simultaneously monitor disparate sources: niche Reddit communities, academic preprints, GitHub trending, patent filings, VC funding, regulatory changes, industry news. Use graph theory to find unexpected connections. Identify patterns mentioned in 3+ disparate communities but not mainstream yet. Create '6 months from now' predictions with confidence scores. Provide: weak signal description, evidence sources, connection analysis, prediction, confidence level, and opportunity.
markdown# Weak Signal Synthesizer Output **Generated**: {timestamp} --- ## Results [Your formatted output here] --- ## Recommendations [Actionable next steps]
Trigger Phrases:
Example Request: > "Sample user request here]"
Response Approach:
Remember: Focus on delivering value quickly and clearly!
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 28,919 | 27,286 | -6% | 1 | 1 | 0% | 3,921 | 4,216 | +8% | 0 | 0 | — |
case-02 | fail→pass | 22,879 | 26,491 | +16% | 1 | 1 | 0% | 3,078 | 4,142 | +35% | 0 | 0 | — |
case-03 | fail→pass | 26,667 | 25,771 | -3% | 1 | 1 | 0% | 3,840 | 4,127 | +7% | 0 | 0 | — |
case-04 | fail→pass | 22,792 | 28,996 | +27% | 1 | 1 | 0% | 3,231 | 4,109 | +27% | 0 | 0 | — |
case-05 | fail→pass | 26,728 | 28,784 | +8% | 1 | 1 | 0% | 3,856 | 4,407 | +14% | 0 | 0 | — |
case-06 | fail→pass | 34,498 | 27,800 | -19% | 1 | 1 | 0% | 4,757 | 4,512 | -5% | 0 | 0 | — |
case-07 | fail→pass | 25,009 | 25,372 | +1% | 1 | 1 | 0% | 3,349 | 3,923 | +17% | 0 | 0 | — |
case-08 | fail→pass | 28,296 | 23,641 | -16% | 1 | 1 | 0% | 3,951 | 3,784 | -4% | 0 | 0 | — |
case-09 | fail→pass | 23,800 | 31,155 | +31% | 1 | 1 | 0% | 3,410 | 4,978 | +46% | 0 | 0 | — |
case-10 | fail→pass | 28,193 | 28,154 | -0% | 1 | 1 | 0% | 3,484 | 4,142 | +19% | 0 | 0 | — |
case-11 | fail→pass | 27,165 | 30,610 | +13% | 1 | 1 | 0% | 3,600 | 4,342 | +21% | 0 | 0 | — |
case-12 | fail→pass | 30,054 | 26,722 | -11% | 1 | 1 | 0% | 4,445 | 3,969 | -11% | 0 | 0 | — |
case-13 | fail→pass | 25,624 | 28,232 | +10% | 1 | 1 | 0% | 3,434 | 3,947 | +15% | 0 | 0 | — |
case-14 | fail→pass | 24,074 | 21,412 | -11% | 1 | 1 | 0% | 3,201 | 3,154 | -1% | 0 | 0 | — |
case-15 | fail→pass | 25,611 | 25,994 | +1% | 1 | 1 | 0% | 3,451 | 3,680 | +7% | 0 | 0 | — |
case-16 | fail→pass | 30,138 | 30,212 | +0% | 1 | 1 | 0% | 4,072 | 4,611 | +13% | 0 | 0 | — |
case-17 | fail→pass | 26,273 | 27,025 | +3% | 1 | 1 | 0% | 3,475 | 4,037 | +16% | 0 | 0 | — |
case-18 | pass→pass | 23,965 | 28,322 | +18% | 1 | 1 | 0% | 3,459 | 4,194 | +21% | 0 | 0 | — |
case-19 | fail→pass | 26,086 | 27,018 | +4% | 1 | 1 | 0% | 3,521 | 4,227 | +20% | 0 | 0 | — |
case-20 | fail→pass | 40,751 | 28,260 | -31% | 1 | 1 | 0% | 5,610 | 4,258 | -24% | 0 | 0 | — |
case-21 | fail→pass | 25,983 | 26,591 | +2% | 1 | 1 | 0% | 3,417 | 4,082 | +19% | 0 | 0 | — |
case-22 | pass→fail | 19,454 | 20,749 | +7% | 1 | 1 | 0% | 2,806 | 3,267 | +16% | 0 | 0 | — |
case-23 | pass→fail | 14,496 | 20,167 | +39% | 1 | 1 | 0% | 2,190 | 3,271 | +49% | 0 | 0 | — |
case-24 | pass→pass | 13,213 | 17,143 | +30% | 1 | 1 | 0% | 2,691 | 3,456 | +28% | 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. 24 cases were attempted. The headline lift of +75 percentage points is the difference between those two pass rates over the 24 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.