self-review

Automate NeurIPS-format academic paper reviews using three personas and aggregated JSON reports.

Updated Apr 23, 2026
One-click install
npx skills add https://github.com/Embers-of-the-Fire/agent-research-skills-opencode --skill self-review-embers-of-the-fire
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: self-review
Source: https://github.com/Embers-of-the-Fire/agent-research-skills-opencode/tree/main/.opencode/skills/self-review
Command: npx skills add https://github.com/Embers-of-the-Fire/agent-research-skills-opencode --skill self-review-embers-of-the-fire

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pymupdf4llm, pymupdf, pypdf, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Automatically review academic papers using the NeurIPS review form with three reviewer personas, ensemble scoring, and reflection refinement. Extracts text from PDF, runs structured review, and outputs actionable feedback. Use when the user wants to review a paper before submission or get feedback on a draft.

Core Features & Use Cases

  • Multi-persona reviews: Harsh, critical, and open-minded reviewers produce independent assessments.
  • Text extraction and structuring: Extracts paper text from PDF/TeX, identifies sections, and formats outputs for review.
  • Aggregation and reporting: Combines reviews, computes consensus scores, and outputs an actionable report for authors.

Quick Start

Provide a PDF or TeX file to the skill and request a NeurIPS-formatted review; it will return a structured JSON-ready report.

Frequently Asked Questions about self-review

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

FAQPage Schema
How do I automate peer review for an academic paper using a NeurIPS form?

Automate peer review by providing a PDF or TeX file to extract text and run structured assessments. The system applies three reviewer personas, aggregates their scores, and outputs a final NeurIPS-formatted report.

Can I extract text from a LaTeX source or PDF for automated review?

Yes, text extraction from PDF and TeX sources is fully supported. The process uses pymupdf and pypdf dependencies to parse documents, identify sections, and format the extracted text for structured review.

How does multi-persona automated paper review work?

Multi-persona review works by deploying harsh, critical, and open-minded reviewer personas to produce independent assessments. These individual reviews are then aggregated to compute consensus scores and generate an actionable feedback report.

Do I need a specific format to get structured JSON output from a peer review?

No specific input format beyond PDF or TeX is required. The system processes the provided academic paper and automatically generates structured NeurIPS review JSON output, alongside an aggregated final report with optional references.

What is the best way to get actionable feedback on a draft before journal submission?

The best way to get actionable feedback is running an automated NeurIPS-formatted review on your draft. It combines ensemble scoring from multiple reviewer personas and reflection refinement to highlight weaknesses and suggest improvements.

What are the limitations of using automated reviewer personas for academic papers?

Automated reviewer personas are limited to the text extracted from the provided PDF or TeX files and follow a strict NeurIPS rubric. They cannot evaluate external dependencies, novel data not present in the text, or non-standard formatting outside the structured review scope.