Is the Nice Figures skill free?

Yes. The skill itself is free from Mapika/nice-figures. SkillProof publishes the install command and an independent test verdict at no cost.

What is the Nice Figures SkillProof Score?

9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 5/5.

How do I install Nice Figures?

Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.

Can I use Nice Figures with Cursor, Copilot, Gemini CLI, Codex or other AI tools?

The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.

Nice Figures

Matplotlib figures in the soft-pastel alignment-research register, 16 recipes

by Mapika · Mapika/nice-figures

Tested · Works

Install — copy the command Download source ↗

Test report

Verdict: Tested · Works
Score: 9.2/10
Tested: Aug 10, 2026
Environment: Claude Code 2.x (agent harness)

Built the same grouped bar chart twice — once with plain matplotlib, once through recipe 4 — and the difference is visible at a glance: the skill version has rounded bar tops, a coherent coral/peach/graphite palette, a top legend sitting clear of the tall bars, muted grey ticks with no gridlines, and both a 300 dpi PNG and a PDF with pdf.fonttype 42 set, versus the default blue/orange/green bars with a legend parked over the data. The README claim that every recipe is executed in CI holds: all 36 tests pass on a clean checkout. Its mandatory look-at-the-PNG step earned its keep immediately — the rendered "down better" badge came out as a tofu box because the arrow glyph is missing from the fallback font when Inter is not installed, which no error message would have caught. Nothing in the skill touches the network, the filesystem outside the working directory, or any credential.

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

Installs cleanly 5/5
Triggers reliably 5/5
Output vs. baseline 8/10
Docs & honesty 5/5

What Nice Figures does

A matplotlib style module plus 16 copy-runnable chart recipes that produce research-blog figures: bold display titles, warm coral/peach/sage palettes, rounded bar tops, smoothed trends with shaded bands, minimal axes and white conference-ready backgrounds. Triggers on requests for training curves, eval bar charts, scaling-law scatters, poster or appendix figures, or anything asked for in an Anthropic-style or soft-pastel style. Exports PDF and PNG at 300 dpi and instructs the agent to open the PNG and fix what it sees before delivering.

How to install Nice Figures

git clone https://github.com/Mapika/nice-figures.git
mkdir -p ~/.claude/skills
cd nice-figures && cp -r plugins/nice-figures/skills/nice-figures ~/.claude/skills/nice-figures

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger Nice Figures

/nice-figures Matplotlib figures in the soft-pastel alignment-research register, 16 recipes

It also activates on plain-language prompts like these:

Make a research-blog style bar chart of these eval results
Plot training curves with smoothed trends and shaded bands
Create a polished matplotlib figure comparing 4 eval scenarios

Frequently asked questions

Is the Nice Figures skill free?: Yes. The skill itself is free from Mapika/nice-figures. SkillProof publishes the install command and an independent test verdict at no cost.
Does Nice Figures work with Claude Code?: We tested it with Claude Code 2.x (agent harness) on Aug 10, 2026. Verdict: Tested · Works. Built the same grouped bar chart twice — once with plain matplotlib, once through recipe 4 — and the difference is visible at a glance: the skill version has rounded bar tops, a coherent coral/peach/graphite palette, a top legend sitting clear of the tall bars, muted grey ticks with no gridlines, and both a 300 dpi PNG and a PDF with pdf.fonttype 42 set, versus the default blue/orange/green bars with a legend parked over the data. The README claim that every recipe is executed in CI holds: all 36 tests pass on a clean checkout. Its mandatory look-at-the-PNG step earned its keep immediately — the rendered "down better" badge came out as a tofu box because the arrow glyph is missing from the fallback font when Inter is not installed, which no error message would have caught. Nothing in the skill touches the network, the filesystem outside the working directory, or any credential.
What is the Nice Figures SkillProof Score?: 9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 5/5.
How do I install Nice Figures?: Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
Can I use Nice Figures with Cursor, Copilot, Gemini CLI, Codex or other AI tools?: The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.