CLI Data Visualization Quickstart Skill

What This Skill Enables

Claude can create charts and visualizations from your data (CSV, JSON, Excel) using matplotlib, seaborn, plotly, or other visualization libraries. Generate publication-ready charts, dashboards, and data visualizations with custom styling.

Compatibility

Native

• Claude Code / Claude: native skill usage via SKILL.md.
• Codex/OpenAI workflows: compatible with Agent Skills-style SKILL.md content as reusable workflow instructions.

Manual Adaptation

• Gemini CLI: native skill usage via .gemini/skills/<skill-name>/SKILL.md or .agents/skills/<skill-name>/SKILL.md where supported.
• Cursor: use the generated .cursor/rules/*.mdc adapter for project rules.
• OpenClaw and similar agents: use the same skill content as a reusable prompt/workflow file when native skill import is unavailable.

Prerequisites

Required:

• Claude Pro subscription
• Code Interpreter feature enabled
• Data file uploaded (CSV, JSON, Excel)

What Claude handles:

• Installing visualization libraries (matplotlib, seaborn, plotly)
• Data loading and preprocessing
• Chart generation with customization
• Exporting to PNG, SVG, or interactive HTML
• Multi-chart layouts and dashboards

How to Use This Skill

Quick Chart from Data

Prompt: "Create a bar chart from this CSV showing sales by category. Make it professional-looking with labels and save as chart.png"

Claude will:

1. Load and analyze the CSV
2. Generate bar chart
3. Add labels, title, legend
4. Apply professional styling
5. Export high-resolution PNG

Time Series Plot

Prompt: "Plot this time series data: dates on x-axis, values on y-axis. Show trend line and save as SVG."

Claude will:

1. Parse date column
2. Create line plot
3. Add trend line (regression)
4. Format dates nicely
5. Export as scalable SVG

Multiple Subplots

Prompt: "Create a 2x2 grid of charts from this data:

• Top left: revenue by month
• Top right: customer distribution
• Bottom left: product performance
• Bottom right: regional breakdown Use consistent colors and save as dashboard.png"

Claude will:

1. Create subplot layout
2. Generate each chart
3. Apply consistent styling
4. Add overall title
5. Export combined visualization

Interactive Chart

Prompt: "Create an interactive plotly chart with hover tooltips and zoom. Save as HTML."

Claude will:

1. Use plotly library
2. Create interactive visualization
3. Add hover information
4. Enable zoom/pan
5. Export as standalone HTML file

Common Workflows

Sales Dashboard

"Create a sales dashboard from this data:
1. Line chart: monthly revenue trend
2. Bar chart: top 10 products by sales
3. Pie chart: sales by region
4. Table: key metrics summary
Use a professional color scheme and save as sales_dashboard.png"

Statistical Visualization

"Visualize this dataset statistically:
1. Histogram with distribution curve
2. Box plot showing quartiles
3. Scatter plot with correlation
4. Heatmap of correlations between variables
Add statistical annotations and save as analysis.png"

Comparative Analysis

"Compare Year 2024 vs 2025 data:
1. Side-by-side bar charts
2. Percentage change annotations
3. Highlight positive/negative changes with colors
4. Add summary statistics
Make it presentation-ready"

Custom Styled Chart

"Create a chart matching our brand:
- Primary color: #FF6B35
- Font: Arial
- Style: minimalist, no grid lines
- Background: white
- High DPI for print (300 dpi)
Show monthly data as area chart"

Chart Types Available

Basic Charts

• Line plots (single/multiple series)
• Bar charts (vertical/horizontal)
• Scatter plots (with trend lines)
• Pie charts (with percentages)
• Area charts (stacked/unstacked)

Statistical Charts

• Histograms (with KDE)
• Box plots (with outliers)
• Violin plots
• Heatmaps (correlation matrices)
• Distribution plots

Advanced Charts

• Multi-axis plots
• Subplots and grids
• 3D visualizations
• Animated charts
• Interactive dashboards

Plotting libraries compared

This skill leans on the Python plotting stack; pick the library by output type:

Use Matplotlib for precise static figures, Seaborn for quick statistical charts, or Plotly when you need interactivity in a browser.

Tips for Best Results

1. Describe Your Data: Tell Claude what each column represents
2. Specify Chart Type: Be clear about visualization type (bar, line, scatter, etc.)
3. Styling Preferences: Mention colors, fonts, size, DPI
4. Labels Matter: Ask for clear titles, axis labels, legends
5. Export Format: PNG for presentations, SVG for web, HTML for interactive
6. Size/Resolution: Specify dimensions ("800x600 pixels" or "10x6 inches at 300 dpi")
7. Multiple Charts: Describe layout ("2x2 grid" or "side by side")

Customization Options

Colors & Themes

• Built-in themes (seaborn, ggplot, bmh)
• Custom color palettes
• Brand color matching
• Color-blind friendly palettes

Annotations

• Data labels on points/bars
• Trend lines and statistics
• Reference lines
• Text annotations
• Arrows and callouts

Export Options

• PNG (raster, high DPI)
• SVG (vector, scalable)
• PDF (print-ready)
• HTML (interactive)
• Multiple formats at once

Troubleshooting

Issue: Charts look cluttered Solution: "Simplify the chart: remove grid, use fewer colors, increase spacing"

Issue: Text too small or overlapping Solution: "Increase font size to 12pt and rotate x-axis labels 45 degrees"

Issue: Colors don't match brand Solution: Provide hex codes: "Use #FF6B35 for primary, #4ECDC4 for secondary"

Issue: Export quality is poor Solution: "Export at 300 DPI for print quality" or "Use vector format (SVG/PDF)"

Issue: Legend blocks data Solution: "Move legend outside plot area to the right" or "Use smaller legend with abbreviations"

Issue: Date axis formatting is wrong Solution: "Format x-axis dates as 'MMM YYYY' with one tick per month"

Learn More

• Matplotlib Gallery - Chart examples and code
• Seaborn Tutorial - Statistical visualization guide
• Plotly Documentation - Interactive charts
• Data Viz Best Practices - Choosing the right chart
• Color Brewer - Color scheme picker

Features

• CLI-first workflows
• Common chart templates (bar/line/scatter)
• Headless export
• Reproducible configurations
• Multiple export formats (PNG, SVG, PDF, HTML)
• Statistical visualization support (histograms, box plots, heatmaps)
• Custom styling and branding options
• Interactive terminal charts and graphs with real-time data updates, animated visualizations, and keyboard navigation for exploring data in terminal environments

Use Cases

• Exploratory analysis
• Executive snapshots
• CI artifact generation
• Automated report generation in CI/CD pipelines
• Data exploration and analysis workflows
• Presentation-ready visualizations with custom branding

• Source repositorygithub.com 2025-10-15
• Documentationmatplotlib.org
• Package (SHA-256 pinned)/downloads/skills/cli-data-viz-quickstart.zip