score-and-rank

Score and rank research questions by blending data feasibility with literature signals.

Updated Mar 8, 2026
One-click install
npx skills add https://github.com/DamarisDeng/paper-writing-system --skill score-and-rank
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: score-and-rank
Source: https://github.com/DamarisDeng/paper-writing-system/tree/main/workflow/skills/score-and-rank
Command: npx skills add https://github.com/DamarisDeng/paper-writing-system --skill score-and-rank

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Score-and-rank automates the critical early-stage decision in research pipelines by combining data feasibility with literature-informed signals to identify and select the top candidate research question for analysis.

Core Features & Use Cases

  • Reads Stage 2 outputs (research_questions.json) and computes a composite score by blending data feasibility, novelty, and literature support, then ranks candidates.
  • Supports a fast-track mode that reuses prior scoring details to skip web searches during rapid iteration.
  • Outputs backward-compatible ranked_questions.json and detailed scoring results for downstream stages.
  • Provides a structured selection_metadata block and progress tracking to support auditability and reproducibility.
  • Useful for teams automating topic selection in health research pipelines, clinical data studies, or any workflow requiring reproducible candidate evaluation.

Quick Start

Invoke the score-and-rank process with your pipeline output folder to compute scores and select the top candidate.

Frequently Asked Questions about score-and-rank

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

FAQPage Schema
How do I automate ranking candidate research questions in a health data pipeline?

Automating research question ranking involves computing a composite score from data feasibility and literature signals. This Skill reads pipeline outputs to blend novelty and support metrics, selecting the top candidate for downstream analysis.

Can I score research questions without performing web searches for novelty every time?

Yes, you can score research questions without web searches using fast-track mode. It reuses prior scoring details to skip online novelty checks, enabling rapid iteration while maintaining deterministic ranking outputs.

What is the best way to combine data feasibility with literature support for topic selection?

Combining data feasibility with literature support requires a composite scoring workflow. This process integrates data readiness with web-backed novelty signals to evaluate and rank candidate research topics reproducibly.

Does this research question scoring process output backward-compatible formats for downstream stages?

Yes, the research question scoring process outputs backward-compatible ranked_questions.json files. It also provides detailed scoring results and a structured selection_metadata block to support auditability across pipeline stages.

How do I evaluate research topic candidates when I only have a research_questions.json file?

Evaluating research topic candidates from a research_questions.json file requires an end-to-end scoring workflow. The process reads Stage 2 outputs, computes blended feasibility and literature scores, and tracks progress for reproducible topic selection.