/creating-data-visualizations

name: creating-data-visualizations
description: |
  Create analytical charts and plots from existing data.
  Use for exploratory or reporting visuals such as bars, lines, scatters, and dashboards; not for publication-grade scientific figures or AI-generated schematics.
allowed-tools: Read, Write, Edit, Grep, Glob, Bash(cmd:*)
version: 1.0.0
author: Jeremy Longshore <jeremy@intentsolutions.io>
license: MIT

Data Visualization Creator

Use this skill when the user needs a chart from data that already exists.

Overview

This skill focuses on everyday analytical visualization choices: bars, lines, scatters, distributions, comparisons, and simple dashboards.

When to Use This Skill

• Plotting trends, distributions, comparisons, or correlations from structured data
• Turning query output or CSV tables into charts for analysis or slides
• Choosing a sensible chart type and axis treatment for a non-specialist audience

Not For / Boundaries

• Journal-ready figures, multi-panel publication layouts, TIFF/600dpi exports: use `scientific-visualization`
• Structural diagrams, flowcharts, and mechanism illustrations: use `scientific-schematics` or `markdown-mermaid-writing`
• Full research-report ownership: use `scientific-reporting`

Integration

• Pair with `exploratory-data-analysis` before plotting questionable data
• Hand off to `scientific-visualization` if the chart must meet publication standards

Example Requests

• "Plot monthly conversion rate with a confidence band"
• "Make a scatter plot of latency vs throughput"
• "Turn this CSV summary into a small dashboard"