analysis-report

Render quantitative results as plots derived directly from data.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/databio/ai-sandbox --skill analysis-report
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: analysis-report
Source: https://github.com/databio/ai-sandbox/tree/main/workspaces/.claude/skills/analysis-report
Command: npx skills add https://github.com/databio/ai-sandbox --skill analysis-report

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Render all quantitative results as plots derived directly from data to prevent numeric fabrication.

Core Features & Use Cases

  • All quantitative information is represented exclusively as plots derived from data.
  • Supports workflows that produce summaries, correlations, and comparisons where AI could otherwise invent figures.
  • Provides a simple, repeatable process to generate shareable plots and reports.

Quick Start

Run the analysis workflow to generate a plots-only report from your data.

Frequently Asked Questions about analysis-report

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

FAQPage Schema
How do I prevent AI from fabricating numbers in data analysis reports?

To prevent numeric fabrication in data analysis, use a plots-only reporting approach that renders all quantitative results directly from data. This forces the pipeline to compute statistics and save visualizations to a plots directory instead of generating text-based figures.

How do I generate a plots-only report from summary statistics?

Generating a plots-only report requires a pipeline that computes statistics, saves plots to a dedicated directory, and assembles a report that references only those plots. This ensures all quantitative information is represented exclusively as visualizations.

Does plots-only reporting work for correlation and comparison workflows?

Plots-only reporting works for data analysis workflows involving summaries, correlations, and comparisons. It is designed for scenarios where AI could otherwise invent figures, ensuring all quantitative information is derived directly from the source data.

What is the best way to ensure data integrity in automated analytics reports?

The best way to ensure data integrity in automated analytics is to eliminate text-based numbers and render all quantitative results as plots derived directly from data. This prevents the AI from inventing figures and provides a repeatable process for generating reports.

When should I not use plots-only reporting for data visualization?

You should not use plots-only reporting if your workflow requires exact numeric values in text, as this approach represents all quantitative information exclusively as plots. It is designed to prevent fabrication by omitting raw figures from the final report.