This repository provides an automated framework for managing data flow at various stages using the Common Workflow Language (CWL). It allows seamless integration of processes and tasks in scientific workflows. To be able to use it, first one needs to install CWL on the system as well explained here. Each step contains:

• an input.json file containing required input information, e.g. input.json file determines the path of cloning the PIConGPU (picongpu_git_path), and where to save it in your local directory (picongpu_local_dir).
• a Python code doing the tasks.
• a tool.cwl file written in the CWL language that reads inputs from the .json file, runs the Python code and processes all steps defined there. Importantly, one should define the FULL path of the Python code used in each .cwl file, e.g. baseCommand in 1-git_clone_picongpu_tool.cwl.
  To run a step, one just needs to open a terminal where these files are present and type in the command line cwltool, relevant .cwl and .json files, e.g.:

cwltool 5-json_file_transfer_tool.cwl 5-json_file_transfer_input.json

which sends metadata to this website. We can also send files to the website using 6-file_transfer_tool.cwl.