Referee — Data & AI agent for Claude Code
Simulated blind peer reviewer.
How to install Referee
Installs to ~/.claude/agents/matthewmair42-claude-config-referee.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/MatthewMair42/claude-config/HEAD/agents/referee.md -o ~/.claude/agents/matthewmair42-claude-config-referee.md Restart Claude Code, or start a new session, for it to be picked up.
What Referee does
name: referee description: Simulated blind peer reviewer. Evaluates papers across 5 dimensions (contribution, identification, data, writing, journal fit) with weighted scoring. Two referees are assigned independently by the Editor — neither sees the other's report. Use for paper review or pre-submission quality check. tools: Read, Grep, Glob model: inherit source: clo-author author: Hugo Sant'Anna adapted_by: Matthew Mair adaptation: light upstream_url: https://github.com/hugosantanna/clo-au
Alternatives in Data & AI
- Peer Reviewer Computational — A simulated peer reviewer specializing in computational social science methods — NLP/text-as-data, machine lea 134 ★
- Book Evaluator — Independent evaluator for the book pipeline 114 ★
- Gsd Eval Planner — Designs a structured evaluation strategy for an AI phase 57 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Peer Reviewer Ethics
A simulated peer reviewer specializing in research ethics, IRB compliance, informed consent, vulnerable popula
Health Reviewer
Healthcare / clinical-data specialist. Use PROACTIVELY when code touches PHI / ePHI, HIPAA-regulated workflows
Outreach Drafter
Intent Outreach Phase-3 worker. Scores one lead's fit against the ICP and drafts grounded outreach (email or L
Repo Finder
Discovers high-value open-source repositories worth contributing to. Scores repos by maintainer responsiveness
Zp Verifier
ZeroProtocol adversarial verifier. Use to try to REFUTE a candidate finding before it is written up - re-repro
Holdings Financial Scorer
Grades the current financial condition of one ETF's holdings: each holding's financial grade weighted by its s
Related Skills
Agentic Eval
Niche-agnostic agentic evaluator using CLEAR v2.0 framework — 6-domain assessment, 8 analysis dimensions, 6-ti
Contribution Dashboard
Read-only dashboard for the opensource-contributor state. Shows active run, score history, plateaued dimension
Bug Race
Use when a bug has already survived at least one debugging attempt and you have 2-3 genuinely distinct root-ca