Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →社媒矩阵数据追踪面板模板（Social Media Matrix Tracker）。 Use when users ask for a cinematic, data-dense social media analytics dashboard with multi-platform metrics, interactive charts, hover insights, range compare, and dark/light theme switching in a single HTML artifact.
.claude/skills/nexu-io-social-media-matrix-tracker-template/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 10% | 0% |
| case-04 | ✓→✗ | ▼ Worse | 123% | 0% |
| case-05 | ✓→✗ | ▼ Worse | 93% | 0% |
| case-06 | ✓→✗ | ▼ Worse | 205% | 0% |
| case-01 | ✗→✗ | ＝ Same ✗ | 10% | 0% |
Ship a premium, cinematic social-media analytics template with high data density and production-grade micro-interactions.
textsocial-media-matrix-tracker-template/ ├── SKILL.md ├── assets/ │ └── template.html ├── references/ │ └── checklist.md └── example.html
DESIGN.md first, map tokens to CSS variables, then adapt assets/template.html.references/checklist.md before emitting the artifact.assets/template.html ships with hand-tuned sample arrays. When swapping in real data, follow this contract so charts and insights stay coherent:
values[] and xLabels[] are the same length and aligned by index.hourChart (engagements / decimals: 0), slaChart (minutes / decimals: 0), roiChart (multiplier x / decimals: 1), cohortChart (% / decimals: 0). Mismatched units break tooltip + A/B insight copy.mixChart, every bars[i].stack must remain a 3-element array ([Video, Thread, Carousel]). For sentimentChart, arcs[] percentages should sum to 100.Emit one short sentence before the artifact (e.g. "Cinematic social media matrix tracker — dark theme by default, all interactions live.") and then a single self-contained HTML artifact.
<artifact> wrapper exactly as shown below — the daemon parser keys on the wrapper element.identifier="social-media-matrix-tracker" is the canonical id (don't suffix with project name).type="text/html" and title are required.template.html already declares.template.html (with DESIGN.md tokens applied), not a verbatim copy and not a separate index.html. Do not emit both.xml<artifact identifier="social-media-matrix-tracker" type="text/html" title="Social Media Matrix Tracker"> <!doctype html> <html>...</html> </artifact>
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 41,129 | 37,066 | -10% | 1 | 1 | 0% | 8,263 | 9,068 | +10% | 0 | 0 | — |
case-02 | fail→fail | 37,989 | 43,213 | +14% | 1 | 1 | 0% | 8,260 | 9,065 | +10% | 0 | 0 | — |
case-03 | fail→pass | 42,089 | 36,233 | -14% | 1 | 1 | 0% | 8,255 | 9,060 | +10% | 0 | 0 | — |
case-04 | pass→fail | 20,199 | 40,703 | +102% | 1 | 1 | 0% | 4,044 | 9,022 | +123% | 0 | 0 | — |
case-05 | pass→fail | 23,586 | 37,793 | +60% | 1 | 1 | 0% | 4,684 | 9,022 | +93% | 0 | 0 | — |
case-06 | pass→fail | 15,172 | 37,853 | +149% | 1 | 1 | 0% | 2,956 | 9,020 | +205% | 0 | 0 | — |
case-07 | fail→fail | 21,140 | 36,234 | +71% | 1 | 1 | 0% | 3,887 | 9,038 | +133% | 0 | 0 | — |
case-08 | fail→fail | 37,253 | 39,189 | +5% | 1 | 1 | 0% | 8,231 | 9,036 | +10% | 0 | 0 | — |
case-09 | fail→fail | 38,662 | 39,582 | +2% | 1 | 1 | 0% | 8,229 | 9,034 | +10% | 0 | 0 | — |
case-10 | fail→fail | 14,933 | 39,929 | +167% | 1 | 1 | 0% | 2,553 | 9,036 | +254% | 0 | 0 | — |
case-11 | fail→fail | 37,343 | 38,586 | +3% | 1 | 1 | 0% | 8,228 | 9,033 | +10% | 0 | 0 | — |
case-12 | fail→fail | 41,893 | 39,777 | -5% | 1 | 1 | 0% | 8,222 | 9,027 | +10% | 0 | 0 | — |
case-13 | fail→fail | 16,316 | 37,475 | +130% | 1 | 1 | 0% | 3,107 | 9,026 | +191% | 0 | 0 | — |
case-14 | fail→fail | 38,662 | 42,248 | +9% | 1 | 1 | 0% | 8,227 | 9,032 | +10% | 0 | 0 | — |
case-15 | fail→fail | 33,614 | 42,595 | +27% | 1 | 1 | 0% | 7,028 | 9,029 | +28% | 0 | 0 | — |
case-16 | fail→fail | 38,952 | 43,302 | +11% | 1 | 1 | 0% | 8,222 | 9,027 | +10% | 0 | 0 | — |
case-17 | fail→fail | 35,779 | 43,404 | +21% | 1 | 1 | 0% | 7,788 | 9,028 | +16% | 0 | 0 | — |
case-18 | fail→fail | 39,391 | 40,510 | +3% | 1 | 1 | 0% | 8,232 | 9,037 | +10% | 0 | 0 | — |
case-19 | fail→fail | 39,668 | 39,120 | -1% | 1 | 1 | 0% | 8,223 | 9,028 | +10% | 0 | 0 | — |
case-20 | fail→fail | 37,130 | 39,318 | +6% | 1 | 1 | 0% | 8,219 | 9,024 | +10% | 0 | 0 | — |
case-21 | fail→fail | 39,404 | 37,487 | -5% | 1 | 1 | 0% | 8,223 | 9,028 | +10% | 0 | 0 | — |
case-22 | fail→fail | 39,110 | 37,679 | -4% | 1 | 1 | 0% | 8,222 | 9,027 | +10% | 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 -9 percentage points is the difference between those two pass rates over the 22 comparable cases. 7 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.