Does this skill require an existing data file?

Yes, this skill is designed to work with an existing 'papers/extraction_table.csv' file generated during the earlier stages of a systematic review.

What scale does the Bias Assessor use for evaluation?

The skill uses a standardized three-level scale: low, unclear, and high risk of bias, ensuring consistency across the entire dataset.

How is the overall risk of bias (rob_overall) determined?

It uses a conservative logic: the score is 'high' if any domain is high, and 'unclear' if there are no high risks but at least one domain is marked unclear.

How does it handle missing methodological details in a paper?

The skill is programmed to prefer an 'unclear' rating with a specific note regarding what information was missing, rather than guessing at the risk level.

Which domains of bias are evaluated by this skill?

It evaluates selection bias, measurement bias, confounding bias, and reporting bias, then generates an overall summary score.

Systematic Review Bias Assessor

Name: Systematic Review Bias Assessor
Author: WILLOSCAR

byWILLOSCAR

•

Ciencia de Datos y ML

Standardizes and populates risk-of-bias assessment fields in evidence extraction tables to ensure research quality and transparency.

The Bias Assessor skill is designed for researchers and data scientists performing systematic reviews who need to evaluate the quality of their evidence base. It automatically appends standardized risk-of-bias (RoB) domains—including selection, measurement, confounding, and reporting—to existing extraction tables. By enforcing a strict 'low/unclear/high' scale and requiring justifying notes for every entry, the skill ensures that evidence synthesis is grounded in an 'evidence-first' methodology. This prevents hollow reporting and provides an auditable trail for why specific studies were rated as they were, significantly speeding up the critical appraisal phase of research pipelines.

Características Principales

01Mandatory short-form justifying notes for every assessment to ensure auditability

02Automated injection of standardized RoB columns (Selection, Measurement, Confounding, Reporting)

03Conservative overall risk calculation logic based on domain-specific ratings

0489 GitHub stars

05Strict enforcement of a 3-level assessment scale (low, unclear, high)

06Data normalization for inconsistent table headers and research artifacts

Casos de Uso

01Audit-proofing research findings by linking quality ratings to specific methodological limitations

02Conducting systematic literature reviews where evidence quality must be quantified and standardized

03Updating existing CSV extraction tables with rigorous, auditable bias metrics before data synthesis

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add willoscar/research-units-pipeline-skills bias-assessor

For use in Claude.ai and ChatGPT

Download Skill