research-review

Conduct multi-round critical research reviews via a secondary Codex agent.

Updated Apr 10, 2026
One-click install
npx skills add https://github.com/KwongFuk/codex-skills --skill research-review-kwongfuk
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research-review
Source: https://github.com/KwongFuk/codex-skills/tree/main/global/research-review
Command: npx skills add https://github.com/KwongFuk/codex-skills --skill research-review-kwongfuk

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill enables a deep, critical evaluation of research work by orchestrating a multi-round review via a secondary Codex agent, surfacing gaps and strengths.

Core Features & Use Cases

  • Iterative critique: conducts multi-round analysis to identify logical gaps, missing experiments, and narrative weaknesses across papers, proposals, or experiments.
  • Actionable deliverables: aggregates criticisms into concrete recommendations, design changes, and a final review document suitable for submission.
  • Use Cases: ideal for pre-submission manuscript reviews, grant proposals, or experimental plans awaiting rigorous evaluation.

Quick Start

Initiate a multi-round critical review of your research work using a secondary Codex agent.

Frequently Asked Questions about research-review

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

FAQPage Schema
How do I perform a multi-round critical review of a research paper?

To perform a multi-round critical review, this Skill delegates the evaluation to a secondary Codex agent. It conducts round-based dialogue to gather context, identify logical gaps, and deliver a final review document with actionable criticisms.

Can I use an automated agent to critique missing experiments in a grant proposal?

Yes, you can critique missing experiments in a grant proposal by initiating the iterative review workflow. The secondary agent performs multi-round analysis to surface narrative weaknesses and aggregate them into concrete recommendations for your submission.

What is the best way to get actionable feedback on an experimental plan before submission?

The best way to get actionable feedback on an experimental plan is using a delegated secondary agent for an ultra-deep critique. It systematically evaluates the research work across multiple rounds, compiling identified weaknesses into a final review document.

Does this critical review approach work for evaluating both papers and experimental designs?

Yes, this critical review approach works for evaluating papers, proposals, and experimental designs. The multi-round agent workflow assesses logical gaps and narrative weaknesses across various research outputs, aggregating the criticisms into actionable design changes.

How does the secondary agent gather context during the research review process?

The secondary agent gathers context during the research review process through round-based dialogue. This iterative mechanism allows the agent to progressively analyze the research work, collect necessary details, and formulate a comprehensive final review document.