Research Integrity Screening automates the complex and time-consuming process of academic fraud detection. By integrating with AI agents like Claude and Cursor, it connects to a comprehensive network of 7 live data sources, including OpenAlex, ORCID, PubMed, and NIH Grants. This allows for swift identification and flagging of potential issues such as paper mill output, citation manipulation, and grant-publication discrepancies. The tool provides consistent, reproducible integrity scores and clear verdicts (CLEAR, MINOR_CONCERNS, INVESTIGATION_NEEDED, HIGH_RISK), transforming manual, inconsistent reviews into a streamlined, automated workflow for grant reviewers, journal editors, and research integrity officers.
Key Features
01Four independent scoring models (Researcher Integrity, Paper Mill, Journal Quality, Funding Risk) for comprehensive assessment.
02Benford's law citation analysis to detect statistical anomalies in citation distributions.
030 GitHub stars
04Retraction and correction detection across OpenAlex, PubMed, and Semantic Scholar.
05Weighted composite integrity score (0-100) with five-tier verdicts and deterministic required actions.
06Paper mill template detection by identifying repeating title prefixes across publication sets.
Use Cases
01Journal submission integrity review to detect paper mill output and suspicious author patterns before peer review.
02Faculty hiring due diligence to verify ORCID identity, assess publication rates, and check for retraction history.
03Pre-award grant screening for federal agencies and private foundations to vet principal investigators.
