スキル一覧に戻る
braselog

peer-review

by braselog

Research Assistant - A reproducible scientific workflow and manuscript builder

2🍴 0📅 2026年1月19日
GitHubで見るManusで実行

SKILL.md


name: peer-review description: Systematic peer review and self-evaluation toolkit. Evaluate methodology, statistics, experimental design, reproducibility, ethics, and reporting standards. Use during the REVIEW phase to assess manuscript quality before submission, or when reviewing others' work.

Scientific Peer Review

Rigorously evaluate scientific work for quality, validity, and reproducibility.

When to Use

  • Self-reviewing manuscript before submission (REVIEW phase)
  • Evaluating methodology and experimental design
  • Checking statistical analyses and reporting
  • Assessing reproducibility and data availability
  • Reviewing others' manuscripts for journals
  • Evaluating grant proposals
  • Quality checking your own work during ANALYSIS phase

Review Workflow

1. INITIAL SCAN      → Overall impression, scope, significance
2. SECTION REVIEW    → Detailed evaluation of each section
3. METHODOLOGY       → Rigor, assumptions, controls
4. STATISTICS        → Appropriate tests, effect sizes, reporting
5. REPRODUCIBILITY   → Data, code, materials availability
6. FIGURES/TABLES    → Clarity, integrity, accessibility
7. ETHICS            → Approvals, consent, conflicts
8. WRITING           → Clarity, organization, accuracy
9. SYNTHESIZE        → Major/minor issues, recommendation

Stage 1: Initial Assessment

Quick Questions (5 minutes)

  1. What is the central research question?
  2. What are the main findings?
  3. Is the work scientifically sound?
  4. Are there any immediate major flaws?
  5. Is it appropriate for the intended venue?

Initial Summary Template

## Initial Assessment

**Research Question**: [One sentence summary]

**Main Findings**: [2-3 key results]

**Initial Impression**: [Sound/Concerning/Major issues]

**Significance**: [Novel contribution to field?]

Stage 2: Section-by-Section Review

Abstract & Title

CheckQuestionStatus
AccuracyDoes abstract reflect the actual study?
ClarityIs the title specific and informative?
CompletenessAre key findings summarized?
AccessibilityUnderstandable to broad audience?

Introduction

CheckQuestionStatus
ContextIs background adequate and current?
RationaleIs the research question justified?
NoveltyIs originality clearly stated?
LiteratureAre relevant papers cited?
ObjectivesAre aims/hypotheses clear?

Methods

CheckQuestionStatus
ReproducibilityCan another researcher replicate this?
RigorAre methods appropriate for the question?
DetailProtocols, reagents, parameters described?
EthicsApprovals and consent documented?
StatisticsMethods described and justified?
ControlsAppropriate controls included?

Critical Details to Verify:

  • Sample sizes and power calculations
  • Randomization and blinding
  • Inclusion/exclusion criteria
  • Software versions
  • Statistical tests and corrections

Results

CheckQuestionStatus
PresentationLogical and clear?
FiguresAppropriate, clear, labeled?
StatisticsEffect sizes, CIs, p-values?
ObjectivityResults without interpretation?
CompletenessNegative results included?

Common Issues:

  • Selective reporting
  • Inappropriate statistical tests
  • Missing error bars
  • Over-fitting
  • Batch effects or confounders
  • Missing controls

Discussion

CheckQuestionStatus
InterpretationConclusions supported by data?
LimitationsAcknowledged and discussed?
ContextPlaced appropriately in literature?
SpeculationDistinguished from data-supported claims?
SignificanceImplications clearly stated?

Red Flags:

  • Overstated conclusions
  • Ignoring contradictory evidence
  • Causal claims from correlational data
  • Mechanistic claims without evidence

Stage 3: Methodological Rigor

Statistical Assessment

CheckQuestionStatus
AssumptionsAre statistical assumptions met?
Effect sizesReported alongside p-values?
Multiple testingCorrection applied?
Confidence intervalsProvided?
Sample sizeJustified with power analysis?
Missing dataHandled appropriately?
Exploratory vs confirmatoryClearly distinguished?

Experimental Design

CheckQuestionStatus
ControlsAppropriate and adequate?
ReplicationBiological and technical?
ConfoundersIdentified and controlled?
RandomizationProperly implemented?
BlindingAdequate for the study?

Stage 4: Reproducibility Assessment

Data Availability

CheckQuestionStatus
Raw dataDeposited in repository?
Accession numbersProvided for databases?
RestrictionsJustified (e.g., privacy)?
FormatsStandard and accessible?

Code and Materials

CheckQuestionStatus
Analysis codeAvailable (GitHub, Zenodo)?
ProtocolsDetailed enough to reproduce?
MaterialsAvailable or recreatable?

Reporting Standards

Check adherence to discipline-specific guidelines:

Study TypeGuidelineStatus
Randomized trialCONSORT
ObservationalSTROBE
Systematic reviewPRISMA
Diagnostic studySTARD
Animal researchARRIVE
Case reportCARE

Stage 5: Figure and Table Review

Quality Checks

CheckQuestionStatus
ResolutionHigh quality?
LabelsAll axes/columns labeled with units?
Error barsDefined (SD, SEM, CI)?
StatisticsSignificance markers explained?
ColorColorblind-friendly?
Scale barsIncluded for images?

Integrity Checks

CheckQuestionStatus
ManipulationAny signs of image manipulation?
SplicingGels/blots appropriately presented?
RepresentativeImages truly representative?
CompleteAll conditions shown?

Stage 6: Writing Quality

Structure and Organization

CheckQuestionStatus
LogicManuscript logically organized?
FlowSections flow coherently?
TransitionsClear between ideas?
NarrativeCompelling and clear?

Writing Quality

CheckQuestionStatus
ClarityLanguage clear and precise?
JargonMinimized and defined?
GrammarCorrect throughout?
ConciseNo unnecessary complexity?

Structuring the Review Report

Summary Statement (1-2 paragraphs)

## Summary

[Brief synopsis of the research]

**Recommendation**: [Accept / Minor revisions / Major revisions / Reject]

**Key Strengths**:
1. [Strength 1]
2. [Strength 2]
3. [Strength 3]

**Key Weaknesses**:
1. [Weakness 1]
2. [Weakness 2]

**Bottom Line**: [Overall assessment of significance and soundness]

Major Comments

Issues that significantly impact validity or interpretability:

## Major Comments

1. **[Issue Title]**
   - *Problem*: [Clear statement of the issue]
   - *Why it matters*: [Impact on conclusions]
   - *Suggestion*: [How to address it]

2. **[Issue Title]**
   ...

Major issues typically include:

  • Fundamental methodological flaws
  • Inappropriate statistical analyses
  • Unsupported conclusions
  • Missing critical controls
  • Reproducibility concerns

Minor Comments

Less critical issues that would improve the manuscript:

## Minor Comments

1. [Page/Figure X]: [Issue and suggestion]
2. [Methods section]: [Missing detail]
3. [Figure 2]: [Clarity improvement]

Review Tone Guidelines

Do ✓

  • Be constructive and specific
  • Acknowledge strengths
  • Provide actionable suggestions
  • Focus on the science
  • Be thorough but proportionate

Don't ✗

  • Use dismissive language
  • Make personal attacks
  • Be vague or sarcastic
  • Request unnecessary experiments
  • Impose personal preferences as requirements

Self-Review Checklist (Before Submission)

Use this during your REVIEW phase:

Methodology

  • Methods are reproducible
  • Controls are appropriate and documented
  • Statistical methods are justified
  • Sample sizes are adequate

Results

  • All results support conclusions
  • Effect sizes are reported
  • Negative results are included
  • Figures are clear and accessible

Reproducibility

  • Data will be available
  • Code is documented and available
  • Protocols are detailed
  • Reporting guidelines followed

Writing

  • Abstract accurately summarizes the work
  • Conclusions are supported by data
  • Limitations are acknowledged
  • References are current and complete

Integration with RA Workflow

REVIEW Phase Activities

  1. Run self-review using this checklist
  2. Document issues in tasks.md
  3. Address each issue systematically
  4. Re-review until checklist passes
  5. Update .research/logs/activity.md

Pre-Submission Verification

Before calling a manuscript complete:

  • Self-review completed
  • All major issues addressed
  • Figures meet journal requirements
  • Data/code deposited
  • Reporting checklist complete
  • Cover letter prepared

スコア

総合スコア

50/100

リポジトリの品質指標に基づく評価

SKILL.md

SKILL.mdファイルが含まれている

+20
LICENSE

ライセンスが設定されている

0/10
説明文

100文字以上の説明がある

0/10
人気

GitHub Stars 100以上

0/15
最近の活動

3ヶ月以内に更新

+5
フォーク

10回以上フォークされている

0/5
Issue管理

オープンIssueが50未満

+5
言語

プログラミング言語が設定されている

+5
タグ

1つ以上のタグが設定されている

0/5

レビュー

💬

レビュー機能は近日公開予定です