# infographics

> Create professional infographics using AI with smart iterative refinement and quality review — supporting 10 infographic types, 8 industry styles, colorblind-safe palettes, and integrated web research for data-accurate outputs.

**Use case**: Create polished data-driven infographics with AI refinement

**Canonical URL**: https://agentcookbooks.com/skills/infographics/

**Topics**: claude-code, skills, science, science

**Trigger phrases**: "create an infographic", "visualize this data as infographic", "make a timeline infographic", "comparison infographic", "process infographic"

**Source**: [K-Dense AI](https://github.com/K-Dense-AI/scientific-agent-skills/tree/main/scientific-skills/infographics)

**License**: MIT

---

## What it does

`infographics` is a Claude Code skill from K-Dense AI's [scientific-agent-skills repo](https://github.com/K-Dense-AI/scientific-agent-skills). It turns Claude into an infographic designer using Nano Banana Pro AI with iterative quality refinement reviewed by Gemini 3 Pro. The skill supports 10 infographic types (comparison, timeline, process, statistical, hierarchical, geographic, relationship, list, mixed media, data visualization), 8 industry styles, and colorblind-safe palettes, with `research-lookup` integration for sourcing accurate data.

A session produces a polished infographic image: content is researched or supplied, the layout and visual design are generated, quality is reviewed against the target type's design criteria, and the output is refined until it meets the threshold.

## When to use it

Reach for it when:

- You need a data-driven infographic for a report, presentation, or communication material — not just a chart, but a designed visual that tells a story
- You want timeline, process flow, or comparison infographics where the visual layout conveys relationships that tables or charts don't
- You need a professional-looking output without manual design work in Figma or Illustrator

When *not* to reach for it:

- Publication-quality data plots for journal figures — use `scientific-visualization`
- Technical scientific diagrams (pathways, architectures, flowcharts) — use `scientific-schematics`

## Install

Copy the `SKILL.md` from K-Dense AI's [infographics folder](https://github.com/K-Dense-AI/scientific-agent-skills/tree/main/scientific-skills/infographics) into `.claude/skills/infographics/` in your project. Requires an OpenRouter API key for the Nano Banana Pro generation backend.

Trigger phrases: "create an infographic", "visualize this data as infographic", "make a timeline infographic", "comparison infographic".

## What a session looks like

A typical session has three phases:

1. **Type and content specification.** Specify the infographic type, the data or story to visualize, the target audience and style, and any brand colors or constraints. Claude selects the appropriate layout template and proposes a content structure.
2. **Generation and quality review.** Nano Banana Pro generates the infographic; Gemini 3 Pro reviews it against the type-specific quality criteria (visual hierarchy, data accuracy, readability, color accessibility). If below threshold, Claude iterates with refined parameters.
3. **Output.** The accepted infographic is saved to the specified path. Claude notes any data elements that were sourced from `research-lookup` and should be verified before publication.

## Receipts

**Where it works well:**
- Timeline infographics for historical or process narratives — the layout generation handles chronological visual flow well and the output is cleaner than typical slide-based alternatives
- Colorblind-safe palette enforcement — the skill applies Okabe-Ito or similar accessible palettes automatically when accessibility is specified

**Where it backfires:**
- Infographics requiring precise, branded typography or pixel-perfect layout for print production — the AI generation produces good-enough output for digital/screen use but not print-ready files
- Rapidly changing data that needs frequent updates — infographics generated by this skill are static images; dynamic data needs a different approach

**Pattern that works:** supply the data directly (as numbers and categories) rather than asking Claude to research it; `research-lookup` integration helps when data isn't available, but supplied data produces more accurate and trustworthy infographics.

## Source and attribution

The canonical SKILL.md lives in the [`infographics` folder](https://github.com/K-Dense-AI/scientific-agent-skills/tree/main/scientific-skills/infographics) of K-Dense AI's scientific-agent-skills repository.

License: MIT. Install, adapt, and redistribute with attribution preserved.

This page documents the skill from a practitioner's perspective. For the formal spec and any updates, defer to the source repo.