How does this skill help with research novelty?

It includes a 'Find the Gap' stage that evaluates potential research directions based on novelty, difficulty, impact, and scope to ensure the work isn't incremental or trivial.

What does the final output look like?

The skill generates a Research Question Brief containing a territory map, candidate gaps, refined questions with hypotheses, and immediate next steps.

Who is this skill for?

It is designed for PhD students, academic researchers, R&D engineers, and anyone looking to move from a vague idea to a concrete technical investigation.

Can this skill help with project feasibility?

Yes, it guides users through assessing compute resources, data availability, time requirements, and technical risks before starting a project.

What is the Research Question Refiner?

It is a Claude Code skill that provides a structured framework to transform broad research interests into specific, tractable, and measurable research questions.

Research Question Refiner

Name: Research Question Refiner
Author: GhostScientist

byGhostScientist

•

Data Science & ML

Transforms vague research interests into concrete, measurable, and tractable research questions through systematic refinement and feasibility analysis.

The Research Question Refiner is a specialized workflow designed to help researchers, developers, and academics bridge the gap between broad curiosity and actionable investigation. It employs a multi-stage process that includes excavating core interests, mapping the current state of the art, and identifying high-impact gaps in existing knowledge. By providing structured formulas for hypothesis generation and rigorous feasibility checks, this skill ensures that research projects are properly scoped, novel enough to matter, and realistic enough to complete within available resource constraints.

Key Features

01Six-stage systematic refinement process from interest to action

02Gap-finding framework to balance novelty, difficulty, and impact

03Comprehensive territory mapping for knowns, unknowns, and controversies

04Template-based formula for generating concrete, measurable research questions

051 GitHub stars

06Rigorous feasibility and risk assessment for resource management

Use Cases

01Scoping a PhD thesis chapter or academic research paper

02Narrowing a broad technical interest into a viable 6-month research roadmap

03Refining an R&D project idea for machine learning or software engineering

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add ghostscientist/skills research-question-refiner

For use in Claude.ai and ChatGPT

Download Skill