review-facts

Audit completed analysis directories for data integrity and reproducibility.

3|Updated May 7, 2026
One-click install
npx skills add https://github.com/bestdan/workflow-skills --skill review-facts
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-facts
Source: https://github.com/bestdan/workflow-skills/tree/main/skills/review-facts
Command: npx skills add https://github.com/bestdan/workflow-skills --skill review-facts

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the problem of confirmation bias and human error in complex data analysis by enforcing an independent, objective audit of completed work.

Core Features & Use Cases

  • Independent Verification: Spawns a dedicated subagent to audit analysis artifacts without inheriting the author's original reasoning.
  • Comprehensive Checklist: Validates links, cited values, reproducibility, number-tracing, units, and formula correctness.
  • Use Case: Before shipping a high-stakes decision memo or financial report, use this skill to ensure all underlying data and calculations are accurate and reproducible.

Quick Start

Invoke the review-facts skill by pointing it to the directory containing your completed analysis artifacts.

Frequently Asked Questions about review-facts

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

FAQPage Schema
How do I audit a data analysis pipeline for reproducibility and fact-checking?

To audit a data analysis pipeline for reproducibility, you can use an automated subagent to independently verify model files, outputs, and narrative documentation without inheriting the author's original reasoning.

What is independent fact-checking for high-stakes financial reports?

Independent fact-checking for high-stakes financial reports is an automated process that validates cited values, links, units, and formula correctness to prevent confirmation bias and ensure quantitative accuracy before shipping.

Can I verify data integrity and trace numbers in completed analysis artifacts?

Yes, you can verify data integrity and trace numbers by applying a comprehensive checklist to a structured analysis directory, validating source links, cited values, and reproducibility of the completed artifacts.

What do I need to provide for an automated analysis pipeline audit?

You need to provide a structured analysis directory containing the completed model files, outputs, and narrative documentation required for the subagent to perform the independent audit.

Does automated fact-checking work for high-stakes decision memos?

Yes, automated fact-checking works for high-stakes decision memos by ensuring all underlying data, calculations, and source verification are accurate and reproducible before final delivery.

Why does an independent subagent prevent confirmation bias in data analysis?

An independent subagent prevents confirmation bias by auditing analysis artifacts without inheriting the author's original reasoning, enforcing an objective verification of data integrity and formula correctness.