OOPR Tool v1.0-alpha

Open Ontological Peer Review — Reproducible AI-Based Peer Review for Ontological Framework

What is OOPR?

OOPR enables rigorous, reproducible peer review of ontological and foundational frameworks using AI reviewers. It formalizes review as an open, transparent, and repeatable procedure.

The Problem: Traditional peer review fails systematically with pre-physical models, cross-disciplinary frameworks, and paradigm-questioning work.

The Solution: OOPR separates review methodology from disciplinary authority through explicit criteria, transparent prompts, and full reproducibility.

Quick Start

Use in Claude Artifacts (Recommended)

1. Go to claude.ai
2. Ask Claude: "Create the OOPR Tool v1.0-alpha as an artifact"
3. Paste the React code from src/oopr-tool.jsx
4. Start reviewing immediately

Local Development

git clone https://github.com/Christianfwb/oopr-tool.git
cd oopr-tool
npm install
npm start

7 Review Axes

OOPR evaluates frameworks along seven rigorous dimensions:

• A. Axiomatic Clarity - Are foundational assumptions explicit and non-circular?
• B. Internal Consistency - Are there contradictions or undefined behaviors?
• C. Logical Derivation - Do conclusions follow necessarily from premises?
• D. Conceptual Precision - Are key terms unambiguous and operationalizable?
• E. Model Scope & Boundaries - What does it explain? Where does it stop?
• F. Falsifiability & Kill-Test - What experiment would prove it wrong?
• G. Terminological Economy - Are new terms necessary or just renamings?

Key Features

• ✅ Queue-Based Reviews - Run all 7 axes automatically in sequence
• ✅ Stop Control - Abort reviews mid-process if needed
• ✅ Dual Export - JSON snapshots (scientific) + Markdown reports (human-readable)
• ✅ Full Reproducibility - Every review includes prompts, versions, and metadata

How to Use

1. Enter Your Framework - Paste your ontological model with axioms, principles, and definitions
2. Choose Review Mode - "Run Missing" (skip completed) or "Re-run All" (fresh start)
3. Start Review - AI analyzes each axis with brutal honesty
4. Export Results - Download JSON for reproducibility or copy Markdown for documentation

Philosophy

"Do not ask whether a model is correct. Ask whether it is coherent, explicit, and inspectable."

OOPR is not about declaring truth—it's about making review transparent, reproducible, and honest.

∞ − 1 = you: AI reviewers are finite agents, not omniscient judges. They serve understanding and dialogue, not verdicts.

Current Limitations (Alpha)

⚠️ Use in trusted environments only:

• API calls made directly from frontend (no backend proxy)
• No rate limit handling or automatic retry
• No persistence between sessions
• Best for: Claude artifacts, local testing, research use

Roadmap

v1.0-alpha (Current)

• ✅ 7 review axes, queue system, exports

v2.0 (Next)

• 🔨 Serverless backend with secure API handling
• 🔨 Rate limit detection and retry logic
• 🔨 Public deployment ready

v3.0 (Future)

• 💭 Meta-review synthesis across all axes
• 💭 Multi-AI comparison (Claude vs GPT vs Gemini)
• 💭 Review evolution tracking

Citation

@software{berrang2025oopr,
  author = {Berrang, Christian},
  title = {OOPR: Open Ontological Peer Review Protocol},
  year = {2025},
  version = {1.0-alpha},
  url = {https://github.com/Christianfwb/oopr-tool}
}

Contributing

OOPR is an open protocol. Contributions welcome:

• Code: Backend implementation, UI/UX improvements, new review axes
• Testing: Bug reports, diverse framework testing
• Documentation: Examples, tutorials, translations
• Research: Compare with human review, develop meta-review algorithms

See ROADMAP.md for detailed plans.

Contact

• 📧 Email: christianfwb@gmail.com
• 🔗 GitHub: github.com/Christianfwb

License

MIT License - See LICENSE file for details.