dataviz

Critiques and advises on data visualization design using Tufte, Wong, and Franchi principles.

Updated Aug 1, 2026
One-click install
npx skills add https://github.com/siegenthalerroger/.llmctl-marketplace --skill dataviz-siegenthalerroger
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: dataviz
Source: https://github.com/siegenthalerroger/.llmctl-marketplace/tree/main/plugins/llmctl-design-0.1.0/skills/dataviz
Command: npx skills add https://github.com/siegenthalerroger/.llmctl-marketplace --skill dataviz-siegenthalerroger

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Choosing the wrong chart type, cluttered dashboards, and misleading axes undermine how audiences understand quantitative data. This Skill provides a rigorous design framework for evaluating and improving charts, graphs, infographics, and dashboards. ## Core Features & Use Cases - Chart Type Selection: Rules for choosing between bar, line, pie, table, sparkline, and small multiples based on data structure and communication goals. - Integrity Auditing: Evaluate visualizations against Tufte's Lie Factor, data-ink ratio, and chartjunk elimination principles. - Editorial Integration: Apply Franchi's narrative approach to combine data, text, and imagery in editorial spreads and dashboards. - Use Case: When reviewing a financial dashboard, use this Skill to check whether the Y-axis starts at zero, whether color encodes information or decorates, and whether the chart type matches the data story. ## Quick Start Ask the assistant to critique this chart and recommend whether a bar chart or line chart better fits the data.

Frequently Asked Questions about dataviz

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I choose the right chart type for my data?▼

Match the chart to the data relationship: bar charts for comparing discrete categories, line charts for trends over time, pie charts for part-to-whole with 2-3 segments, tables for precise lookup under 20 values, and small multiples for comparing many series across one variable.

What is the data-ink ratio in data visualization?▼

The data-ink ratio is the proportion of ink devoted to non-redundant data display divided by total ink used, from Edward Tufte. Maximize it by erasing non-data-ink like heavy grids, 3D effects, and decorative elements, then erasing redundant data-ink.

When should I use a pie chart vs a bar chart?▼

Use a pie chart only for part-to-whole comparisons with 2-3 segments, starting the largest slice at 12 o'clock and never exceeding 5-7 segments. For more categories or precise comparison, use a bar chart with the Y-axis starting at zero.

Does this Skill generate charts or visualizations?▼

No, it never generates designs. It informs, critiques, and advises on visualization choices, providing evaluation checklists, construction rules, and tool-specific guidance for SVG, D3.js, Figma, InDesign, Illustrator, and Affinity.

Why does my chart look cluttered and how do I fix it?▼

Clutter usually comes from chartjunk: heavy grid lines, 3D effects, redundant labels, and decorative color. Remove everything that does not represent data variation, use gray for structural elements, and label data directly instead of using legends.