AI Agent Hub
Back to skills
Science Figures Academic Chart Generator icon

Science Figures Academic Chart Generator

Data Analysis Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please install @user_40f1996d/science-figures by following https://skillhub.cn/install/skillhub.md.

About this skill

Problem

Research plotting often stalls after data preparation: JSON or CSV exists, but tuning matplotlib fonts, error bars, significance markers, journal typography, and vector output can still produce inconsistent results. science-figures compresses this repetitive work into one command.

Workflow and Core Capabilities

It runs locally with scienceplots and matplotlib, accepts JSON/CSV, and outputs PNG, SVG, or PDF. A typical flow is to choose --type, point to --data, set --style, then adjust --cmap, --vmin, --vmax, --kde, or --layout as needed. It covers 17 chart types, including bar, heatmap, scatter, histogram, radar, surface, contour, and subplots, plus 8 journal styles such as nature, ieee, springer, lancet, acs, and aps. Use forest for meta-analysis, annotated heatmap for matrices, and histogram with density for distribution comparisons.

Boundaries

It targets static publication figures from structured data, not experimental design, statistical testing, or interactive dashboards. Chinese labels require usable fonts; use --cjk-font when needed. SciencePlots is recommended with LaTeX, and a built-in equivalent style is used when it is unavailable.

Use Cases

  • When preparing a paper, turn CSV experimental results into grouped bars with error lines in Nature style for a figure panel.
  • Before submission, export the same experimental data as SVG and PDF for the journal figure and a report slide deck.
  • For meta-analysis, generate a forest plot from JSON effect sizes and confidence intervals to compare intervention effects clearly across groups.
  • When comparing multivariate metrics, create a radar plot or subplot grid from JSON data for a report appendix panel section.

Best For

  • Graduate student preparing SCI figures: needs to turn JSON/CSV experimental data into Nature- or IEEE-styled bar or heatmap plots.
  • Algorithm engineer writing technical reports: needs journal-styled line plots, subplot grids, or 3D surface plots from metric CSVs.
  • Research assistant doing statistical summaries: needs box, violin, histogram, and density plots to compare distribution differences.
  • PhD researcher conducting meta-analysis: needs to convert JSON effect sizes and confidence intervals into forest plots with vector output.