EAGER: Towards an International Open-source Repository Browser (iORB)

Open-source software is now part of the infrastructure of scientific research, but the research-security community lacks reliable data and methods for studying it. This gap is especially important in fast-moving fields such as drones, embodied artificial intelligence, and autonomous systems, where open-source code supports both civilian research and capabilities relevant to national defense. The same software may be developed across countries, maintained by small groups, reused through long dependency chains, and shaped by funding or institutional relationships that are difficult to see. This project establishes the foundations for research on research security related to open-source software. It identifies which projects matter most to U.S. research, who maintains and funds them, how influence is distributed, and how risks propagate across software supply chains. The resulting evidence supports more precise security policies while preserving the openness and international collaboration that benefit science. The project also creates a basis for studying emerging challenges, including open-source software increasingly written, modified, or maintained by artificial intelligence systems. The project creates a canonical, reproducible, and longitudinal dataset of high-impact research software. It links repositories to dependencies, downstream users, contributors, organizational affiliations, funding sources, scholarly use, development activity, and documented security events. These data support formal models of influence, concentrated control, and risk propagation across research-software ecosystems. Statistical and network methods produce validated indicators of scientific importance, contributor concentration, systemic criticality, and potential single points of failure. Comparative studies test whether these indicators explain historical disruptions and whether governance, due diligence, and other policy interventions reduce risk. Drones and embodied artificial intelligence provide important test cases because they combine rapid open-source development with clear civilian, scientific, and national-defense relevance. An international open-source repository browser makes the data and methods available for replication and extension. Collaboration among computer scientists, social and political scientists, cybersecurity researchers, and nonproliferation experts provides a scalable model for future Research on Research Security studies and for evaluating AI-generated open-source software. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.