epistemics

Enforce evidence hierarchies and citation requirements for biomedical research claims.

Updated Nov 9, 2025
One-click install
npx skills add https://github.com/markusstrasser/skills --skill epistemics
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: epistemics
Source: https://github.com/markusstrasser/skills/tree/main/epistemics
Command: npx skills add https://github.com/markusstrasser/skills --skill epistemics

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Bio/medical research often accumulates speculative claims without solid, citable evidence. Epistemics enforces a structured evidence framework to prevent hallucinations and to ensure traceable sources.

Core Features & Use Cases

  • Enforces an evidence hierarchy and mandatory citations for non-trivial claims.
  • Guides researchers through domain-specific failure modes and proper interpretation of genetic and clinical data.
  • Use Case: During a literature review of a biomarker, Epistemics ensures every claim is paired with a DOI, PMID, or official URL and clearly separates mechanistic vs clinical evidence.

Quick Start

Provide a research question and attach any available sources; Epistemics will outline citation requirements and the evidence hierarchy to govern the workflow.

Frequently Asked Questions about epistemics

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

FAQPage Schema
How do I prevent hallucinations in biomedical literature reviews?

To prevent hallucinations in biomedical literature reviews, use an evidence hierarchy framework that mandates citations for non-trivial claims. This ensures traceable sources and separates mechanistic from clinical evidence.

How do I enforce citation requirements for medical research claims?

You can enforce citation requirements for medical research claims by applying a structured evidence framework. This approach pairs every non-trivial assertion with a DOI, PMID, or official URL to maintain source traceability.

What is the best way to separate mechanistic vs clinical evidence in a meta-analysis?

The best way to separate mechanistic vs clinical evidence in a meta-analysis is to implement formal evidence hierarchies with explicit INFERENCE labeling. This clearly distinguishes speculative claims from validated clinical data.

Can I use an anti-hallucination guardrail for pharmacogenomics interpretation?

Yes, anti-hallucination guardrails can be used for pharmacogenomics interpretation. They guide researchers through domain-specific failure modes and ensure proper interpretation of genetic and clinical data.

Does biomedical evidence synthesis work without structured guardrails?

Biomedical evidence synthesis without structured guardrails often accumulates speculative claims without solid, citable evidence. Applying a formal evidence framework prevents hallucinations and ensures traceable sources during research.