scientific-fact-classification

Classify empirical claims by evidence strength and methodological rigor.

3|2|Updated Apr 25, 2026
One-click install
npx skills add https://github.com/patricksavalle/investigate-journalism-skills --skill scientific-fact-classification
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: scientific-fact-classification
Source: https://github.com/patricksavalle/investigate-journalism-skills/tree/main/.agents/skills/scientific-fact-classification
Command: npx skills add https://github.com/patricksavalle/investigate-journalism-skills --skill scientific-fact-classification

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill addresses the common failure of treating unverified claims or consensus-based opinions as objective scientific facts, providing a rigorous framework to audit the epistemic status of any statement.

Core Features & Use Cases

  • Epistemic Auditing: Classifies claims along a spectrum from objective fact to unfalsifiable belief, ensuring that assumptions and opinions are clearly distinguished from evidence-backed findings.
  • Warrant Labeling: Applies standardized qualifiers like (traced), (deferred to consensus), or (memory — unverified) to every claim, forcing transparency regarding the source of knowledge.
  • Use Case: Use this when you need to determine if a scientific paper's conclusion is actually supported by its data, or when you need to distinguish between a proven causal mechanism and a mere correlation in a news article.

Quick Start

Use the scientific-fact-classification skill to audit the claims in this article and assign calibrated confidence levels to each.

Frequently Asked Questions about scientific-fact-classification

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

FAQPage Schema
How do I check if a scientific paper's conclusion is actually supported by its data?

To distinguish correlation from causal mechanism, audit the claim by evaluating evidence strength and warrant type. This applies Bayesian weighing to determine the epistemic status of empirical claims, preventing the conflation of consensus opinion with proven scientific fact.

What is epistemic auditing for empirical claims?

Epistemic auditing classifies empirical claims along a spectrum from objective fact to unfalsifiable belief. It ensures assumptions are distinguished from evidence-backed findings by applying standardized warrant labels like (traced) or (deferred to consensus) to force source transparency.

How do I calibrate evidence confidence for contested narratives?

Calibrate evidence confidence by applying Bayesian weighing to evaluate warrant types and methodological rigor against domain-specific standards. This systematic tracing of primary sources assigns calibrated confidence levels to contested narratives, preventing the treatment of unverified claims as objective facts.

Can I use this fact-checking framework to audit news articles?

Yes, you can audit news articles by evaluating evidence strength and methodological rigor. The framework applies standardized qualifiers to every statement, distinguishing between proven causal mechanisms and mere correlations in contested narratives.

Does fact-checking with Bayesian weighing work for all research domains?

Bayesian weighing works across research domains by evaluating evidence against domain-specific standards. It systematically traces primary sources to determine the epistemic status of load-bearing arguments, ensuring methodological rigor is maintained regardless of the specific research field.

What are the limitations of classifying claims as scientific facts?

A key limitation is the risk of conflating consensus-based opinions with objective scientific facts. This framework mitigates it by requiring systematic tracing of primary sources and applying Bayesian weighing to prevent treating unverified claims or assumptions as evidence-backed findings.