Abstract. The important role of digital resources relies on whether metadata is available and has been correctly catalogued and indexed so that the user can discover and use geospatial datasets. However, the cost and the error-proneness in the manual metadata creation, the lack of information provided by the producers of geospatial datasets and the lack of experience in cataloguing have motivated us to propose a new workflow for the automated metadata generation for geospatial datasets. This paper describes this workflow based on tasks synchronization that gives support for four metadata functions: discovery, use, evaluate and retrieval of digital geodata. The workflow was implemented using a multi-tier architecture system where the Data, Application and User Tiers can run a single use application as well as web services. The prototype evaluation is discussed in terms of the type of metadata being generated and the type of metadata function being supported by the workflow.