selfhost.directory

The index of self-hostable projects — ranked, categorized and served with live updates.
Own your data and stop giving it to big tech.

Rigorous logo

Self-host Rigorous

AI multi-agent system for scientific manuscript peer review

AlphaOpen SourceMIT
Updatedlast year
LanguagePython
Setup difficultyModerate
Stars
253
Requires
OpenAI API (or local LLM)
Install with Python

What is Rigorous?

Rigorous is an open-source AI-powered tool for scientific manuscript analysis developed by the Agentic Systems Lab. It implements a multi-agent system with 24 specialized agents organized into three categories: section agents (S1-S10) that analyze each manuscript section, rigor agents (R1-R7) that evaluate originality, impact, ethics, data availability, statistical rigor, technical accuracy, and consistency, and writing agents (W1-W7) that assess language, narrative, clarity, terminology, inclusive language, citations, and audience alignment. Quality control agents validate outputs across categories, and an executive summary agent provides a synthesized three-paragraph overview. The tool runs locally via Python CLI, requires an OpenAI API key (configurable for local LLMs), processes PDF manuscripts, and generates detailed JSON results and PDF reports. Currently at v0.1 for self-hosted use, with v0.2 available as a hosted web service at rigorous.review.

Should you self-host Rigorous?

Rigorous scores 43/100 on selfhost.directory (how scoring works) and ranks #91 of 125 in AI. It has 253 GitHub stars.

There was no stable release in the past 12 months, so check the repository's recent activity before you rely on it.

Setup is moderate: expect some configuration beyond a single command. It is MIT-licensed and written mostly in Python.

Its closest alternative on selfhost.directory is ComfyUI (93/100, 143 668 stars). It scores higher overall, so compare the two before you pick. See Rigorous vs ComfyUI side by side.

Features

  • Configurable review focus
  • Executive summary generation
  • Local LLM Support
  • Multi-agent manuscript analysis
  • Outlet fit evaluation (in development)In active development, not yet ready for use
  • PDF report generation
  • Quality control validation
  • Scientific rigor assessment
  • Section-specific review agents
  • Structured JSON output
  • Writing quality analysis

Rigorous README

How to install Rigorous

Install Rigorous with Python

Environment variables
OPENAI_API_KEY, DEFAULT_MODEL
  1. Clone the repository
  2. Install dependencies:
pip install -r requirements.txt
  1. Configure your manuscript details in manuscript.json:
{
    "manuscript_src": "path/to/your/manuscript.pdf",
    "publicationOutlets" : "e.g. Nature Medicine",
    "reviewFocus" : "e.g. Statistics and Writing"
}
  1. Run the analysis:
python run_local_aipeer_review.py

Requires Python 3.7+ and OpenAI API key (or local LLM)

How to configure Rigorous

  • DEFAULT_MODELgpt-4.1-nanoDefault LLM model to use

    Default LLM model to use

  • OPENAI_API_KEYOpenAI API key for LLM access

    OpenAI API key for LLM access

More in Research

Rate this project

Click to rate — no account needed.

0 comments

Posting as guest
Loading comments…