Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →14 data visualization skills. Trigger: charts, plots, figures, publication-quality graphics. Design: one skill per tool with code templates and academic formatting conventions.
.claude/skills/brycewang-stanford-dataviz-skills/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 90% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 245% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -34% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -70% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -60% | 0% |
Select the skill matching the user's need, then read its SKILL.md.
| Skill | Description | |-------|-------------| | algorithm-visualizer-guide | Guide to Algorithm Visualizer for interactive algorithm exploration | | bokeh-visualization-guide | Guide to Bokeh for interactive browser-based research visualizations | | chart-image-generator | Generate publication-quality chart images from research data | | color-accessibility-guide | Colorblind-friendly palettes and accessible visualization design | | d3-visualization-guide | Guide to D3.js for building custom interactive data visualizations | | echarts-visualization-guide | Guide to Apache ECharts for interactive research data dashboards | | geospatial-viz-guide | Create maps, choropleths, and spatial data visualizations for research | | interactive-viz-guide | Interactive data visualization with Plotly, ECharts, and D3 | | metabase-analytics-guide | Guide to Metabase for open-source research data analytics and dashboards | | network-visualization-guide | Visualize networks, graphs, citation maps, and relational data | | plotly-interactive-guide | Guide to Plotly.py for interactive scientific visualizations in Python | | publication-figures-guide | Create journal-quality scientific figures with proper styling and accessibility | | python-dataviz-guide | Publication-quality data visualization with matplotlib, seaborn, and plotly | | redash-analytics-guide | Guide to Redash for SQL-driven research data dashboards and sharing |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 4,063 | 8,377 | +106% | 1 | 1 | 0% | 638 | 1,213 | +90% | 0 | 0 | — |
case-02 | fail→fail | 5,340 | 4,559 | -15% | 1 | 1 | 0% | 826 | 729 | -12% | 0 | 0 | — |
case-03 | fail→pass | 2,620 | 4,151 | +58% | 1 | 1 | 0% | 376 | 1,299 | +245% | 0 | 0 | — |
case-04 | fail→fail | 14,253 | 5,626 | -61% | 1 | 1 | 0% | 2,601 | 1,077 | -59% | 0 | 0 | — |
case-05 | fail→fail | 14,593 | 9,128 | -37% | 1 | 1 | 0% | 2,347 | 910 | -61% | 0 | 0 | — |
case-06 | fail→fail | 14,994 | 10,664 | -29% | 1 | 1 | 0% | 2,798 | 1,058 | -62% | 0 | 0 | — |
case-07 | fail→fail | 14,398 | 3,270 | -77% | 1 | 1 | 0% | 2,352 | 824 | -65% | 0 | 0 | — |
case-08 | fail→fail | 18,900 | 10,374 | -45% | 1 | 1 | 0% | 3,402 | 859 | -75% | 0 | 0 | — |
case-09 | fail→fail | 11,980 | 3,218 | -73% | 1 | 1 | 0% | 2,267 | 706 | -69% | 0 | 0 | — |
case-10 | fail→pass | 16,421 | 9,816 | -40% | 1 | 1 | 0% | 2,585 | 1,710 | -34% | 0 | 0 | — |
case-11 | fail→pass | 17,870 | 2,573 | -86% | 1 | 1 | 0% | 3,100 | 930 | -70% | 0 | 0 | — |
case-12 | fail→fail | 12,672 | 7,475 | -41% | 1 | 1 | 0% | 2,357 | 862 | -63% | 0 | 0 | — |
case-13 | fail→pass | 14,601 | 3,444 | -76% | 1 | 1 | 0% | 2,424 | 973 | -60% | 0 | 0 | — |
case-14 | fail→pass | 11,539 | 9,241 | -20% | 1 | 1 | 0% | 2,227 | 1,979 | -11% | 0 | 0 | — |
case-15 | pass→pass | 4,671 | 4,833 | +3% | 1 | 1 | 0% | 717 | 1,047 | +46% | 0 | 0 | — |
case-16 | pass→fail | 7,188 | 4,342 | -40% | 1 | 1 | 0% | 1,100 | 851 | -23% | 0 | 0 | — |
case-17 | pass→fail | 6,209 | 4,193 | -32% | 1 | 1 | 0% | 1,015 | 893 | -12% | 0 | 0 | — |
case-18 | pass→pass | 6,787 | 3,529 | -48% | 1 | 1 | 0% | 908 | 1,130 | +24% | 0 | 0 | — |
case-19 | pass→pass | 7,198 | 8,252 | +15% | 1 | 1 | 0% | 1,114 | 1,662 | +49% | 0 | 0 | — |
case-20 | fail→pass | 11,209 | 7,753 | -31% | 1 | 1 | 0% | 1,903 | 1,995 | +5% | 0 | 0 | — |
case-21 | pass→pass | 9,835 | 4,266 | -57% | 1 | 1 | 0% | 1,530 | 1,298 | -15% | 0 | 0 | — |
case-22 | pass→pass | 11,824 | 11,912 | +1% | 1 | 1 | 0% | 2,393 | 2,088 | -13% | 0 | 0 | — |
case-23 | pass→fail | 10,366 | 3,647 | -65% | 1 | 1 | 0% | 1,823 | 842 | -54% | 0 | 0 | — |
case-24 | fail→fail | 11,694 | 5,967 | -49% | 1 | 1 | 0% | 2,030 | 723 | -64% | 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, and 12 counted toward the lift figure. The other 12 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +17 percentage points is the difference between those two pass rates over the 12 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.