ChaokunHong/MetaScreener

AI-powered tool for efficient abstract and PDF screening in systematic reviews.

74
/ 100
Verified

Employs a multi-LLM ensemble with calibrated confidence aggregation and per-element consensus scoring to route papers to auto-decision or human review based on confidence tiers. Integrates 15+ open-source models via OpenRouter API, features a hierarchical 4-layer screening pipeline combining parallel inference, rule-based filtering, and active learning, with a Vue 3 web UI supporting title/abstract and full-text screening, extraction, and quality assessment workflows.

1,304 stars and 48 monthly downloads. Actively maintained with 226 commits in the last 30 days. Available on PyPI.

Maintenance 22 / 25
Adoption 14 / 25
Maturity 24 / 25
Community 14 / 25

How are scores calculated?

Stars

1,304

Forks

47

Language

Python

License

Apache-2.0

Last pushed

Feb 27, 2026

Monthly downloads

48

Commits (30d)

226

Dependencies

22

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/ChaokunHong/MetaScreener"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.