Abstract: [Purpose/Significance] The aim of this paper is to identify open science capabilities and skills for researchers, build a capability framework of open science, and explore its development path, in order to improve the open science capabilities of team leaders and decision makers, researchers and scientific research support personnel in the open science ecosystem, and to provide reference for promoting open science and building an open science innovation ecosystem in our country. [Method/Process] In this paper, the relevant research on open science definition, open science capabilities, and competence model at home and abroad were analyzed and summarized. Based on the competency onion model, the open science capability framework was designed as three layers from the inside out: embracing open science values and principles and developing an open science mindset as the core layer, acquiring open science knowledge composed of open science elements as the second layer outside, and building open science skills embedded in the scientific research life cycle and data life cycle as the outermost layer. Then, the possible development methods were discussed. At last, libraries' role in the open science training was emphasized. [Results/Conclusions] The open science capability framework includes three dimensions. Firstly, embracing open science values, including quality and integrity, collective benefit, equity and fairness, and diversity and inclusiveness, and open science guiding principles covering transparency, scrutiny, critique and reproducibility, equality of opportunities, responsibility, respect and accountability, collaboration, participation and inclusion, flexibility and sustainability. Secondly, acquiring open science knowledge based on open science elements, including open science definitions, initiatives, best practices, implications, open access, open data (scientific research data, government data, and industry data), open scientific and technological reports, open educational resources, open scientific infrastructure (virtual and physical), open source code, open reproducible research, open peer review, open science evaluation, citizen science, and open science policy. Thirdly, building open science core competencies embedded in scientific research lifecycle and data lifecycle, including (1) open scholarly communication skills, which mean open publication literacy skills, like making the correct choices about where and how to publish their results, how and what to self-archive and how to communicate their research for scholarly and societal impact; (2) data science skills, which mean data management plan, research data management, data analysis, data sharing, and data storage skills; (3) scientific research literacy skills, which mean research management skills, research integrity and ethics skills, and legal skills; (4) citizen science skills, which mean researchers learn how to engage with citizens, and citizens learn how to conduct research. The open science capability development path includes formulating relevant policies, laws and strategic plans for skills building, strengthening open science promotion and training, and providing comprehensive support throughout the entire scientific research process. At the same time, the library is a key player in the construction of open science capacity, and should lead the design of a scientific and rational open science training system, and play an active role in promoting the paradigm shift of open science. In addition, this research has shortcomings due to time limitation, and we will continue to conduct in-depth research on open science capabilities and refine its indicators, in order to provide theoretical basis and reference for libraries to design open science training system and promote open science training in China.

Key words: open science, open science capability, competency model, competency framework, development path