skills.rest
lightrain-a
lightrain-aCOMMUNITY
Explore all skills by this author →
GitHub

research-review

Critically review ML/AI research artifacts and validate results-to-claims.

Updated May 20, 2026
Search Tags:#experiment design#ml#research-review#paper-writing#critical feedback#claims matrix#codex mcp
One-click install
npx skills add https://github.com/lightrain-a/medtrace-aris --skill research-review-lightrain-a
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: research-review
Source: https://github.com/lightrain-a/medtrace-aris/tree/main/.vendor/aris/skills/research-review
Command: npx skills add https://github.com/lightrain-a/medtrace-aris --skill research-review-lightrain-a

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires claude mcp codex server, mcp__codex__codex, mcp__codex__codex-reply.

What problem does it solve?

This Skill helps you get a deep critical review of your research so you can find logical gaps, missing experiments, and narrative weaknesses before you waste time writing or running the wrong work.

Core Features & Use Cases

  • Senior ML-style critique with high reasoning depth: Uses Codex MCP with xhigh reasoning effort to produce reviewer-grade feedback.
  • Iterative multi-round improvement loop: Runs a conversation-style process that asks targeted follow-ups and requests concrete deliverables (e.g., experiment designs, claims matrices).
  • Actionable documentation outputs: Guides you to save a self-contained, round-by-round review and a prioritized experiment/TODO plan in the project root.
  • Use case: When you have a draft paper, research idea, or experimental results, you can prompt for an external reviewer response that pressures the work to meet top-venue standards.

Quick Start

Use research-review to critically evaluate your research and propose the minimal experiment plan needed to address the most important reviewer concerns.

Frequently Asked Questions about research-review

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

FAQPage Schema
How do I get a brutally honest critical review of my ML research paper draft?▼

A critical review of your ML research paper draft is produced by running Codex MCP with xhigh reasoning effort, simulating reviewer-grade feedback to identify logical gaps, missing experiments, and narrative weaknesses before publication.

What is a claims matrix and how does it validate experimental results?▼

A claims matrix validates experimental results by mapping research claims to specific evidence, generated through iterative multi-round Codex MCP review to pinpoint exactly where your experimental methodology lacks sufficient support.

How do I critique my research methodology and plan follow-up experiments?▼

To critique research methodology and plan follow-up experiments, this skill runs an iterative dialogue that produces a prioritized experiment and TODO plan in your project root addressing the most important reviewer concerns.

Does the research review skill work with Claude MCP Codex server?▼

Yes, the research review skill requires the Claude MCP Codex server, specifically utilizing the codex and codex-reply tools to enable multi-round iterative dialogue and high-depth critical reasoning.

What is the best way to find logical gaps in AI experimental work?▼

The best way to find logical gaps in AI experimental work is applying a results-to-claims validation process, which pressures your work to meet top-venue standards by exposing unsupported claims and narrative weaknesses.

What are the limitations of using automated critical feedback for paper writing?▼

Limitations of automated critical feedback for paper writing include its dependency on Codex MCP availability and the need for comprehensive context briefing in the initial request to avoid misdirected methodology critiques.

Related Skills

Explore More Education & Research Skills→
GRIND-Lab-CoreGRIND-Lab-Core
100

research-review

Review research papers and experimental results through iterative dialogue.

Official
Advanced
LambenthanLambenthan
77

/rebuttal

Parse reviewer comments into Rvx-Cy concerns and generate evidence-backed rebuttals.

Community
Advanced
nyxoraAInyxoraAI
5

research-paper-writing

Automate ML research paper writing from literature review to submission preparation.

Community
Advanced
LingrongyeLingrongye
1

research-review

Perform multi-round critical research reviews via Codex MCP with xhigh reasoning.

Community
Advanced