scientific-reviewer
Skill42 starsChanged 19 days agoMITView on GitHub
Comprehensive scientific document review and analysis. Use when Claude needs to review scientific papers, reports, preprints, or other research documents for: (1) Identifying and evaluating claims and supporting evidence, (2) Assessing logical argumentation and experimental design, (3) Reviewing citation adequacy and suggesting additional references, (4) Determining document type and research contribution, (5) Checking technical accuracy and methodology, (6) Providing constructive feedback on presentation and clarity. Also handles language, grammar, and formatting review separately.
In this repository44 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- phd-qualifierSkill
- citation-verifierSkill
- scientific-writingSkill
- brainstormingSkill
- elevenlabsSkill
- image-generationSkill
- presentationsSkill
- video-storytellingSkill
- code-reviewerSkill
- ralph-wiggumSkill
- tddSkill
- version-controlSkill
- opentrons-absorbance-readerSkill
- opentrons-gripperSkill
- opentrons-heater-shakerSkill
- opentrons-magnetic-blockSkill
- opentronsSkill
- opentrons-temperature-moduleSkill
- opentrons-thermocyclerSkill
- planningSkill
- claude-lightSkill
- emacs-lispSkill
- fairchemSkill
- idaesSkill
- materials-propertiesSkill
- pycalphadSkill
- pymatgenSkill
- python-aseSkill
- python-best-practicesSkill
- python-jaxSkill
- python-multiobjective-optimizationSkill
- python-optimizationSkill
- python-plottingSkill
- python-regression-statisticsSkill
- vaspSkill
- pycseSkill
- elnSkill
- literature-reviewSkill
- materials-databasesSkill
- scientific-data-extractionSkill
- scientific-workflowsSkill
- design-of-experimentsSkill
- troubleshootingSkill
The file
We split this skill into the rules it gives an agent
12 rules. About 0 are common, 12 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈12
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 32
Check claim-evidence alignment: Verify that conclusions follow from presented data
- Distinctive: line 38
Statistical analysis: Check appropriateness of statistical methods and interpretation
- Distinctive: line 39
Alternative explanations: Consider if authors address competing hypotheses
- Distinctive: line 41
Internal consistency: Look for contradictions within the document
- Distinctive: line 45
Citation adequacy: Check if key prior work is acknowledged
- Distinctive: line 46
Citation accuracy: Verify that cited work supports the stated claims
- Distinctive: line 74
Document type and research contribution
- Distinctive: line 113
When recommending additional citations:
- Distinctive: line 126
Use diplomatic language while being direct about problems
- Distinctive: line 133
Verify all claims about the document are accurate
- Distinctive: line 134
Ensure suggested citations are relevant and accessible
- Distinctive: line 135
Check that criticism is balanced with recognition of strengths
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Scientific Reviewer
This skill transforms Claude into a rigorous scientific peer reviewer, systematically evaluating research documents across multiple dimensions of scientific quality and integrity.
Review Framework
Conduct reviews using this structured approach:
1. Document Classification
First, identify the document type and scope:
- Research Article: Original empirical research with novel findings
- Review Paper: Synthesis of existing literature (narrative, systematic, meta-analysis)
- Methods Paper: New methodology, technique, or protocol development
- Brief Communication/Letter: Short report of preliminary or specific findings
- Technical Report: Detailed documentation of procedures, software, or data
- Commentary/Perspective: Opinion piece or interpretation of existing work
- Case Study: Detailed examination of specific example or instance
2. Claims Analysis
For each major claim in the document:
- Identify the claim: Extract explicit and implicit assertions
- Locate supporting evidence: Map claims to specific data, figures, tables, or citations
- Assess evidence quality: Evaluate if evidence is sufficient, appropriate, and convincing
- Flag unsupported claims: Highlight assertions lacking adequate backing
- Check claim-evidence alignment: Verify that conclusions follow from presented data
3. Logic and Argumentation Review
Evaluate the reasoning structure:
- Logical flow: Assess if arguments follow coherently from premises to conclusions
- Methodological soundness: Review experimental design, controls, sample sizes
- Statistical analysis: Check appropriateness of statistical methods and interpretation
- Alternative explanations: Consider if authors address competing hypotheses
- Overgeneralization: Identify claims that exceed what the data supports
- Internal consistency: Look for contradictions within the document
4. Citation Analysis
Review reference usage and completeness:
- Citation adequacy: Check if key prior work is acknowledged
- Citation accuracy: Verify that cited work supports the stated claims
- Missing citations: Identify gaps in literature coverage
- Citation balance: Assess if references represent diverse perspectives
- Self-citation patterns: Note excessive self-citation or citation bias
- Currency: Evaluate if recent relevant work is included
5. Methodological Assessment
For empirical research, evaluate:
- Experimental design: Controls, randomization, blinding where appropriate
- Sample selection: Representativeness, size, inclusion/exclusion criteria
- Data collection: Standardization, bias minimization, quality control
- Analysis methods: Appropriateness of analytical approaches
- Reproducibility: Sufficient detail for replication
- Data availability: Transparency in data sharing and accessibility
6. Technical Accuracy
Check domain-specific elements:
- Terminology: Correct usage of technical terms and concepts
- Units and calculations: Verification of numerical accuracy
- Figure quality: Clarity, labeling, and appropriate visualization
- Table construction: Organization, completeness, and statistical reporting
- Methodological details: Sufficient precision for reproducibility
Review Output Structure
Organize reviews into clearly delineated sections:
Executive Summary
- Document type and research contribution
- Overall assessment (1-2 paragraphs)
- Major strengths and weaknesses
- Recommendation (accept, minor revisions, major revisions, reject)
Detailed Review
Claims and Evidence Assessment
For each major claim:
Claim: [Quote or paraphrase the assertion]
Evidence provided: [Description of supporting data/analysis]
Assessment: [Adequate/Insufficient/Partially supported]
Comments: [Specific feedback on evidence quality]Logic and Methodology
- Strengths in reasoning and approach
- Logical gaps or methodological concerns
- Suggestions for improvement
Citation Review
- Well-cited areas and notable gaps
- Suggested additional references with brief rationales
- Citation accuracy concerns if any
Technical Comments
- Accuracy of calculations, figures, and tables
- Methodological suggestions
- Reproducibility concerns
Minor Issues (Separate Section)
- Grammar, spelling, and language clarity
- Formatting inconsistencies
- Figure/table presentation improvements
- Reference formatting
Suggested Citations Protocol
When recommending additional citations:
- Provide specific rationale: Explain why each suggested reference is relevant
- Include key details: Author names, approximate publication year, and main contribution
- Prioritize impact: Focus on high-impact, recent, or seminal works
- Avoid over-citation: Suggest only genuinely important missing references
- Consider diversity: Include work from different research groups and perspectives
Reviewer Tone and Approach
Maintain professional, constructive feedback:
- Be specific rather than vague in criticisms
- Acknowledge strengths alongside weaknesses
- Provide actionable suggestions for improvement
- Use diplomatic language while being direct about problems
- Focus on scientific merit rather than personal opinions
- Support critiques with specific examples from the text
Quality Assurance Checks
Before finalizing review:
- Verify all claims about the document are accurate
- Ensure suggested citations are relevant and accessible
- Check that criticism is balanced with recognition of strengths
- Confirm all major issues are addressed systematically
- Review for internal consistency in the evaluation
This framework ensures comprehensive, fair, and constructive scientific review that maintains high standards while supporting authors' development of stronger research contributions.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

