ReproDB is an open-source effort to unify fragmented artifact evaluation data across security and systems conferences, enabling cross-venue analysis and artifact discovery at scale.
Website: reprodb.github.io
ReproDB combines data collection, schema governance, publication, and research artifacts into one open reproducibility platform:
- reprodb-pipeline: Collects and normalizes artifact, committee, author, and repository metadata
- data-schemas: Defines and versions the machine-readable data model
- reprodb-pipeline-results: Stores generated datasets and snapshots
- reprodb.github.io: Publishes cross-venue analysis and search on the public website
- Reproducibility by default
- Open, inspectable data transformations
- Transparent provenance and documentation
- Practical tooling for chairs, evaluators, and researchers
- Open issues with bug reports or feature requests
- Contribute schema updates, pipeline improvements, or documentation
- Share reproducibility practices from your venue or community
