data-storyteller

Convert experimental metrics into publication-ready figures and tables.

Updated Feb 2, 2026
One-click install
npx skills add https://github.com/SALTYf1SH/md-sci-skill --skill data-storyteller
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: data-storyteller
Source: https://github.com/SALTYf1SH/md-sci-skill/tree/main/.claude/skills/data-storyteller
Command: npx skills add https://github.com/SALTYf1SH/md-sci-skill --skill data-storyteller

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pandas, matplotlib, seaborn, numpy, and includes scripts (resource) components.

What problem does it solve?

Turn raw experimental data into publication-ready figures and tables, enabling researchers to quickly translate logs, metrics, and results into polished visuals and concise summaries.

Core Features & Use Cases

  • Generate publication-quality figures from training curves, metrics, and latent-space visualizations.
  • Create markdown or LaTeX tables with significance testing for results.
  • Produce a concise results_summary.md with actionable insights and conclusions.

Quick Start

Point the workflow at your latest metrics.csv and run it to generate publication-ready figures and a results_summary.

Frequently Asked Questions about data-storyteller

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

FAQPage Schema
How do I turn raw metrics CSVs into publication-ready figures?▼

To turn raw metrics CSVs into publication-ready figures, you can point the workflow at your metrics file to automatically generate polished visuals, clear axis labeling with units, and a concise results summary.

Can I visualize TensorBoard logs and JSON metrics using pandas and matplotlib?▼

Yes, you can visualize TensorBoard logs and JSON metrics using pandas and matplotlib. The workflow processes these formats to create deterministic plots and publication-quality training curves.

What is the best way to create LaTeX tables with significance testing from experimental data?▼

The best way to create LaTeX tables with significance testing from experimental data is to use a workflow that automatically formats your results and summarizes metrics into markdown or LaTeX outputs.

Does this data visualization approach enforce reproducibility for machine learning experiments?▼

Yes, this data visualization approach enforces reproducibility for machine learning experiments by utilizing deterministic plotting scripts to ensure consistent results across multiple runs.

Do I need numpy and seaborn installed to generate research study plots?▼

Yes, you need numpy and seaborn installed to generate research study plots. These dependencies are required to support the underlying data manipulation and visualization functions.