Can this skill access scientific literature?

Yes, it includes specific strategies for searching PubMed for biomedical topics and general web sources for other domains to ground hypotheses in existing evidence.

Why does it generate competing hypotheses?

Generating 3-5 distinct explanations helps avoid confirmation bias and allows for critical comparisons to identify the most likely mechanism behind a phenomenon.

Does it provide specific experimental steps?

It proposes experimental design elements including what to measure, necessary controls, appropriate methods, and potential sample sizes to test specific predictions.

How does this skill ensure scientific rigor?

It follows a structured workflow that evaluates every hypothesis against quality criteria such as falsifiability, parsimony, and explanatory power, ensuring they are more than just descriptive statements.

Which scientific fields can use this skill?

The skill is designed for cross-domain inquiry, making it suitable for biomedicine, chemistry, physics, social sciences, and computational research.

Scientific Hypothesis Generator

Name: Scientific Hypothesis Generator
Author: Zehong-Wang

byZehong-Wang

0•

Data Science & ML

Generates testable, evidence-based scientific hypotheses and structured experimental designs across research domains.

Scientific Hypothesis Generator provides a systematic framework for developing robust, mechanistic explanations for scientific phenomena. It guides Claude through a rigorous process of literature synthesis via PubMed and web search, the formulation of competing hypotheses, and quality assessment against criteria like falsifiability and parsimony. By moving from raw observations to detailed experimental designs and quantitative predictions, this skill ensures that scientific inquiry remains evidence-based and structurally sound, making it an essential tool for researchers, academics, and data scientists.

Key Features

01Quality evaluation framework based on testability, scope, and parsimony

02Integrated literature search strategies for PubMed and general scientific domains

03Systematic multi-step workflow for generating and refining scientific hypotheses

04Mechanistic explanation modeling to identify how and why phenomena occur

05Detailed experimental design patterns including controls and measurement strategies

060 GitHub stars

Use Cases

01Designing controlled experiments to test specific biological, physical, or social phenomena

02Developing research questions and hypotheses for academic papers or grant proposals

03Exploring alternative mechanistic explanations for unexpected experimental results or data patterns

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add zehong-wang/kosmos hypothesis-generation

For use in Claude.ai and ChatGPT

Download Skill