Overview
Data Management Plans (DMPs) are commonly required by funders with every grant proposal, but the formats and requirements imposed by each funder keep evolving, and no two funders' PDFs are structured the same way. DMP Bridge is an open-source (MIT License), Python-based pipeline that converts any DMP PDF, regardless of funder format, into the DMP Tool JSON format, unlocking the ability to import, edit, and continue expanding existing DMPs rather than starting from scratch. This project builds on our earlier DMP Chef pipeline and is part of a broader extension of the DMP Tool platform.
What is DMP Bridge?
DMP Bridge is an open-source, Python-based pipeline that converts DMP PDFs from any funder format into DMP Tool JSON, combining a narrative portion that mirrors DMP Tool's internal structure with the RDA DMP Common Standard JSON for machine-actionable output.
How does it work?
DMP Bridge works in three steps. First, it extracts text from PDF using a tool like pdfplumber. Then, it feeds it to an LLM that produces a structured narrative portion following the DMP Tool schema. Finally, it uses that narrative portion with another LLM extraction process, constrained to the RDA schema, to generate a machine-actionable DMP.
What challenges does it address?
Every funder has its own DMP policies, formats, and compliance criteria, and existing DMPs are usually locked away as static PDFs. DMP Bridge addresses this by turning any DMP PDF into an interoperable, machine-actionable format, laying the foundation for a funder-agnostic tooling ecosystem that's robust to changing DMP guidelines.
Why is this important?
Turning DMPs into machine-actionable objects enables automated compliance checking, AI-based DMP support, and continued editing throughout the data lifecycle, transforming DMPs from static, one-time funding documents into living objects that evolve as data is collected, managed, and shared.
Development Approach
DMP Bridge is developed using an open-source approach. This allows for transparency and collaboration with the open source community. The GitHub repository will be linked here as we test and develop the tool.
Team
Members
Researchers, engineers, and collaborators behind this project.
Research partners
We are collaborating on this project with the developers and maintainers of the DMP Tool at the California Digital Library (University of California).
Funding
This project is a National Science Foundation (NSF) and Chan Zuckerberg Initiative (CZI)-funded collaboration between the FAIR Data Innovations Hub and California Digital Library (University of California Office of the President) as part of a broader extension of the DMP Tool platform.
Project milestones
Key milestones in the development of DMP Bridge.
Deployment and Integration
October 2026