Built in the open. Here's how to take part.
LitFlow is an open research project. You can contribute as a coder, a researcher, a methods reviewer, or an evaluation participant — here is what is currently open.
What's open right now.
The fastest way to take part. Each call is a concrete opportunity with its own scope.
Testers for screening
Researchers running a systematic review, willing to join a methods-evaluation study.
Take part →Calls above are drafted from the project plan — confirm scope and status before launch.
Code is one way of several.
Non-code contributions count as much as code. Translations follow once the UI is internationalised.
Where things stand.
From ideas we're weighing, to what's in active development, to what's already shipped.
- Topic Modeling
- Dual AI Use
- OSF / Zenodo Integration
- Team Screening
- Journal Ranking
- Living Literature Reviews
- Closed alphalive at the University of Münster
Have an idea, or want to help build it?
Propose a feature, contribute, or share review data — tell us what you're working on and we'll take it from there.
What we're trying to figure out.
Methodological questions LitFlow is actively researching. Each is an invitation, not a settled position.
Anchoring effects from AI recommendations
Do AI suggestions bias a reviewer's decision, even when shown with a justification and a confidence score? We're designing screening modes to test it.
Per-criterion reliability across reviewers
How consistently do independent reviewers agree at the level of individual criteria — with and without AI assistance?
Cross-discipline schema transfer
Can an extraction schema built in one field be reused in another without losing fidelity?
The people behind it.
The core maintainer team sits at the University of Münster's Department of Information Systems.
Designs the platform, writes the prompts, runs the evaluations, and merges the pull requests.
Co-leads the AI augmentation architecture and the LLM provider abstraction.