Snowih is an open-source Python tool for automating forward and backward snowballing of research papers. It identifies references from uploaded studies and recursively explores their cited papers, presenting the entire snowballing pathway through an interactive, AI-screened graph that enables reviewers to expand literature searches clearly and reproducibly.

• AI-Driven Screening: Users specify inclusion criteria (keywords, title relevance, study type, publication year, etc.), and Snowih’s AI automatically evaluates each reference

• Recursive Snowballing: Snowih snowballs the uploaded papers and the references of those references—extending the search efficiently across multiple layers

• Graph-Based Visualisation: Displays all identified papers as an interactive network graph

  • Green nodes → meet inclusion criteria
  • Red nodes → excluded based on screening

• Transparent Decisions: Each node can be expanded to view AI-generated reasoning for inclusion/exclusion (migth add in future, not done for now)

• Streamlined Workflow: Greatly reduces time spent on manual screening in systematic reviews, scoping reviews, and meta-analyses

• Export Options: Save screened datasets, graphs, and logs for documentation or PRISMA reporting

Snowih — Created and Developed by Vihaan Sahu

Maintained under the Snowih Organization

NOTE: Snowih is under development currently and is planned to be in production in 2026.

Copyright (c) 2025 Vihaan Sahu

Licensed under the Apache License 2.0.