peer-review_skill

Overview

This skill is a systematic peer review toolkit for evaluating manuscripts and grant proposals across disciplines. It focuses on methodology, statistics, experimental design, reproducibility, ethics, figure integrity, and reporting standards. The goal is to produce clear, actionable reviews that prioritize scientific validity and transparent reporting.

How this skill works

The skill guides reviewers through staged inspections: initial assessment, section-by-section critique, methodological and statistical rigor checks, reproducibility and transparency evaluation, figure and data integrity checks, ethical review, and writing quality. For each stage it suggests concrete checks, flags common pitfalls, and structures feedback into summary, major comments, minor comments, and questions for authors. It also recommends visual schematics to clarify workflows and complex critiques.

When to use it

• Preparing peer reviews for journal manuscripts
• Evaluating grant proposals and research applications
• Assessing experimental designs and statistical analyses
• Checking reproducibility, data sharing, and code availability
• Verifying compliance with reporting guidelines (CONSORT, PRISMA, STROBE)

Best practices

• Start with a concise summary and overall recommendation before detailed comments
• Prioritize major issues that affect validity, then list actionable minor edits
• Request specific data, code, or methods needed to reproduce results
• Distinguish speculation from data-supported conclusions and ask for limitations
• Use a constructive, respectful tone and balance criticisms with strengths

Example use cases

• Produce a structured review for a randomized clinical trial, checking CONSORT adherence
• Assess a computational biology manuscript for code availability, software versions, and validation
• Evaluate a preclinical animal study for IACUC compliance, replication, and appropriate controls
• Review a meta-analysis for PRISMA checklist compliance, search strategy, and heterogeneity handling
• Provide targeted statistical review focusing on sample size justification and multiple testing correction