This project provides AI-powered tools to enhance the transparency, affordability, and efficiency of scientific research creation, evaluation, and dissemination. Targeting researchers and scientists, it offers automated manuscript analysis and feedback, aiming to streamline the peer-review process and improve knowledge sharing.

How It Works

The core is Agent1_Peer_Review, a multi-agent system designed for comprehensive manuscript analysis. It provides detailed feedback on specific sections, scientific rigor, and writing quality, incorporating quality control loops. The system outputs actionable recommendations in JSON format and can generate PDF reports. This approach leverages AI to offer a scalable and consistent review process, aiming for greater objectivity and speed than traditional methods.

Quick Start & Requirements

• Primary install: Python 3.7+ is required. Dependencies are detailed in each tool's requirements.txt.
• Prerequisites: An OpenAI API key is necessary, though the system can be adapted for alternative LLMs, including locally hosted ones.
• Input: PDF manuscripts for analysis.
• Online Interface: AI Reviewer v0.2 is accessible at https://www.rigorous.review/ with progress tracking.
• Documentation: Detailed documentation is mentioned but no direct link is provided.

Highlighted Details

• AI Reviewer v0.2 is live online, offering structured feedback via an interactive interface.
• Module prompts are planned for open-sourcing post-initial testing to foster transparency and community contributions.
• Agent2_Outlet_Fit, a tool for evaluating manuscript fit with target journals/conferences, is currently in development.
• Future ideas include embedding-based similarity analysis for literature comparison and journal suggestions.

Maintenance & Community

Development is active, with Agent1_Peer_Review (v0.1) ready for use and Agent2_Outlet_Fit under active development. Contributions via Pull Requests are welcomed. The project is authored by Robert Jakob and Kevin O'Sullivan. A feedback form is available to help improve the system.

Licensing & Compatibility

License: Not specified in the README. Compatibility for commercial use or linking within closed-source projects is unclear due to the unspecified license.

Limitations & Caveats

The Agent2_Outlet_Fit module is still in development. The system's core functionality relies on an external OpenAI API key, posing a dependency unless adapted to local LLMs. The open-sourcing of prompts for Agent1_Peer_Review is planned but not yet implemented.