name: data-visualization-expert description: Generate insightful, publication-quality visualizations from complex datasets. keywords:
- charts
- plots
- analysis
- pandas
- matplotlib
- seaborn measurable_outcome: Create 3 high-resolution (300dpi) statistical plots (volcano, heatmap, scatter) within 15 minutes. license: MIT metadata: author: AI Agentic Skills Team version: "2.0.0" compatibility:
- system: linux, macos allowed-tools:
- run_shell_command
- write_file
- read_file
Data Visualization Expert
A dedicated skill for transforming raw data (CSV, JSON, Excel) into compelling visual narratives. Specializes in statistical and scientific plotting.
When to Use
- Reports: Summarizing key metrics or KPIs.
- Exploration: Initial data analysis (EDA) to find trends/outliers.
- Publication: Generating figures for papers or presentations.
- Comparison: Comparing models, cohorts, or experimental groups.
Core Capabilities
- Code Generation: Creates Python scripts (Matplotlib, Seaborn, Plotly) or R code (ggplot2).
- Style Enforcement: Adheres to specific journal/company branding (fonts, colors).
- Data Cleaning: Preprocesses data (handle missing values, normalize) for plotting.
- Artifact Management: Saves plots as PNG/SVG/PDF files.
Workflow
- Load Data: Read input file (
pd.read_csv()) and inspect columns/types. - Clean & Transform: Filter, pivot, or aggregate data as needed.
- Generate Plot: Write plotting script with strict aesthetic controls.
- Save & Verify: Execute script, check output file existence/size.
Example Usage
# Agent prompt:
"Visualize the distribution of 'Age' vs 'Income' from customers.csv"
# Triggers generation of `plot_age_income.py` using Seaborn scatterplot.
Guardrails
- Privacy: Avoid plotting PII (names, emails) directly.
- Accuracy: Ensure axes are labeled correctly with units.
- Readability: Use appropriate scales (log vs linear) and avoid clutter.