Scholarpeer Econ — AI skill for Claude Code

ScholarPeer-Econ

**Multi-Agent Peer Review Simulation for Finance and Economics Manuscripts**

Overview

ScholarPeer-Econ simulates a journal peer review panel by dispatching multiple specialized AI reviewer agents in parallel. Each reviewer evaluates your manuscript from their expertise perspective, then a discussion phase synthesizes findings into a consensus report with actionable recommendations.

Installation

1. Clone this repository into your Claude Code skills directory:

   cd ~/.claude/skills/
   git clone https://github.com/yourusername/scholarpeer-econ.git

2. The skill will be automatically available in Claude Code.

Usage

In Claude Code, invoke the skill:

/scholarpeer-econ path/to/manuscript.docx

With journal targeting:

/scholarpeer-econ paper/main.tex --journal JF

How It Works

Reviewer Personas

Persona	Focus Area
Econometrician	Identification strategy, causality, specification, standard errors
Domain Expert	Literature positioning, contribution, institutional context
Methodologist	Data quality, measurement validity, replication potential

Process

1. Phase 1: Setup — Read manuscript, select reviewer personas, load journal profile
2. Phase 2: Independent Review — Dispatch 2-3 reviewers in parallel (blind to each other)
3. Phase 3: Discussion — Identify disagreements, facilitate resolution
4. Phase 4: Consensus Report — Synthesize into unified assessment with scores

Scoring

Overall score aligned with quality gate system:

• ≥95: Ready for submission
• 90-94: Minor revisions needed
• 80-89: Major revisions needed
• <80: Fundamental issues require rethinking

Output

Reports saved to `quality_reports/reviews/`:

• Individual reviewer reports
• Consensus report with actionable recommendations
• Required revisions checklist

Supported Formats

• .docx (Word documents)
• .tex (LaTeX manuscrip