creating-data-visualizations_skill

Overview

This skill automates generation of plots, charts, and graphs from raw data, choosing visualization types that best reveal patterns and relationships. It speeds analysis by inspecting data structure, recommending chart types, and producing ready-to-use visual outputs. Use it to turn tables, time series, or categorical summaries into clear visualizations quickly.

How this skill works

The skill analyzes input data to infer types, distributions, and key variables. It selects an appropriate visualization (bar, line, scatter, histogram, heatmap, etc.) and builds the plot using standard plotting libraries and visualization best practices. It can accept dataframes, CSVs, or direct arrays and returns the generated chart plus suggestions for labels, color schemes, and refinements.

When to use it

• You need a quick, sensible visualization from a dataset or table.
• You want the system to pick an effective chart type automatically.
• You need exploratory visuals to reveal trends, outliers, or correlations.
• You want code or image output for reports, dashboards, or presentations.
• You’re prototyping visualizations and need scaffolding or alternatives.

Best practices

• Provide clean, well-labeled data: columns should have descriptive names and consistent types.
• Specify the goal (compare categories, show trend, explain distribution, reveal correlation).
• Include any filters, grouping, or aggregation requirements up front.
• Request desired output format (PNG, SVG, notebook code, or interactive plot) if you have a preference.
• Review axis labels, units, and legends for clarity and adjust color palettes for accessibility.

Example use cases

• Generate a bar chart of sales by region from a sales CSV for a weekly report.
• Create a line chart of daily stock prices for the past year to visualize trends.
• Produce a scatter plot with regression fit to examine correlation between marketing spend and conversions.
• Draw a histogram of response times to identify performance bottlenecks.
• Build a heatmap of feature correlations to guide feature selection for modeling.