general-peer-review

Evaluate research plans and manuscripts for scientific rigor with structured reviewer output.

144|21|Updated Jan 8, 2026

Search Tags:#reproducibility #research validation #peer review #scientific rigor #methodology critique #literature grounding

Ready to run

npx skills add https://github.com/learningmatter-mit/AtomisticSkills --skill general-peer-review

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: general-peer-review
Source: https://github.com/learningmatter-mit/AtomisticSkills/tree/main/.agents/skills/general-peer-review
Command: npx skills add https://github.com/learningmatter-mit/AtomisticSkills --skill general-peer-review

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents weak or under-validated research plans and manuscripts from moving forward by forcing a critical, adversarial evaluation of assumptions, baselines, sampling quality, and reproducibility.

Core Features & Use Cases

Critical scientific review: Identifies missing baselines, flawed statistics, inadequate sampling/ensemble choices, and unjustified assumptions.
Reproducibility and validation checks: Verifies whether protocols, hyperparameters, and reporting are sufficient for independent replication.
Structured reviewer output: Produces a clear recommendation plus Major/Minor Concerns and Questions for authors.
Use case: Review an AI-driven materials simulation workflow before running expensive computations by checking MD duration, supercell size, theory level, and validation against known references.

Quick Start

Ask your AI to perform peer review on the provided research plan by loading the target document and returning a structured set of Major/Minor Concerns and author questions focused on rigor and validity.

Frequently Asked Questions about general-peer-review

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

FAQPage Schema

How do I check my research plan for scientific rigor before running expensive simulations?▼

To check scientific rigor, evaluate your research plan for missing baselines, flawed statistics, inadequate sampling, and unjustified assumptions before execution. This peer review process identifies methodological weaknesses and verifies if protocols, hyperparameters, and reporting are sufficient for independent replication.

What does a peer review of a manuscript cover for materials and chemistry workflows?▼

A peer review of materials and chemistry manuscripts covers critical scientific validation, focusing on MD duration, supercell size, theory level, and validation against known references. It ensures reproducibility by verifying protocols and reporting are sufficient for independent replication.

How do I assess reproducibility and validation in an AI-driven simulation task?▼

Assess reproducibility and validation in AI-driven simulations by verifying whether protocols, hyperparameters, and reporting are sufficient for independent replication. The review checks baseline assessments, sampling quality, and validation against known references to ensure scientific credibility.

Can I get actionable improvements for a research proposal with flawed statistics?▼

Yes, you can get actionable improvements for research proposals with flawed statistics through a structured reviewer output. This includes a clear recommendation plus Major and Minor Concerns, along with specific questions for authors focused on validity and methodological critique.

Does literature grounding affect methodology critique in machine learning research?▼

Literature grounding directly affects methodology critique in machine learning research by providing the context needed to identify unjustified assumptions and missing baselines. It ensures the research plan is evaluated against existing knowledge before execution or publication.

When do I need a structured peer review output with major and minor concerns?▼

You need a structured peer review output with major and minor concerns when stress-testing research plans or manuscripts before publication or expensive computation. It prevents under-validated work from moving forward by forcing an adversarial evaluation of assumptions and reproducibility.