rct-bias-assessment-rob2

Assess risk of bias in randomized controlled trials using the ROB2 framework.

Updated Sep 15, 2026
One-click install
npx skills add https://github.com/mrsonord2240/openscience-specialists --skill rct-bias-assessment-rob2-mrsonord2240
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rct-bias-assessment-rob2
Source: https://github.com/mrsonord2240/openscience-specialists/tree/main/specialists/critical-appraisal-specialist/versions/1.0.0/package/skills/rct-bias-assessment-rob2
Command: npx skills add https://github.com/mrsonord2240/openscience-specialists --skill rct-bias-assessment-rob2-mrsonord2240

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires PyPDF2, and includes scripts (resource) and references (resource) components.

What problem does it solve? Evaluating the methodological quality of randomized controlled trial papers is slow and inconsistent when done manually. This Skill automates the Cochrane Risk of Bias 2 (ROB2) assessment by analyzing trial text against the five ROB2 domains and producing a structured, reproducible judgement. ## Core Features & Use Cases - Five-Domain ROB2 Assessment: Evaluates randomization, deviations from intended interventions, missing outcome data, outcome measurement, and selective reporting, then synthesizes an overall Low / Some concerns / High judgement. - PDF Text Extraction: Includes a PyPDF2-based script that extracts full text from trial PDFs so assessments can run directly on source papers. - Structured JSON Output: Produces a standardized result object with per-domain and overall judgements, plus a detailed report. - Use Case: A systematic reviewer screening dozens of RCTs can feed each paper's text or PDF into the Skill and receive consistent, domain-level bias judgements ready for inclusion in a review table. ## Quick Start Provide the full text or PDF path of a randomized controlled trial paper and ask for a ROB2 risk of bias assessment with per-domain judgements.

Frequently Asked Questions about rct-bias-assessment-rob2

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

FAQPage Schema
How do I assess risk of bias in a randomized controlled trial?▼

Provide the full text or PDF of the RCT paper, and the Skill evaluates the five ROB2 domains: randomization, deviations from intended interventions, missing outcome data, outcome measurement, and selective reporting. It then synthesizes an overall Low, Some concerns, or High judgement.

What is the ROB2 tool used for?▼

ROB2 is the Cochrane Risk of Bias 2 tool for evaluating bias in randomized trials. It uses signalling questions across five domains to produce domain-level and overall risk-of-bias judgements for use in systematic reviews and critical appraisal.

Can I assess risk of bias directly from a PDF file?▼

Yes. The bundled extract_pdf.py script uses PyPDF2 to extract text from a PDF into a text file, which is then used as input for the ROB2 assessment. Scanned PDFs without a text layer may not extract correctly.

What output format does the ROB2 assessment produce?▼

The assessment returns a JSON object with the study reference, judgements for domains D1 through D5, and an overall rating of Low, Some concerns, or High, along with a detailed narrative report.

When should I not use automated ROB2 assessment?▼

Do not use it when the paper's full text or key methodological sections are unavailable, or when the study is not a randomized trial, since ROB2 applies only to RCTs. Non-randomized studies require tools like ROBINS-I instead.