Parabola steps are stateless, which means that they act independent of previous runs. Consequently, there shouldn’t be a need to ‘clear’ data from a flow. If you input new data into the flow—whether in building or running the flow—steps will automatically update as new data moves through the flow. In fact, many steps will show errors if they don’t have any data flowing through, so it’s not possible to
If you want to share the logic of a flow, but don’t want a viewer to see sensitive data from previous runs, consider copying and pasting snippets, or duplicating the flow and inputting sample or dummy data.
Last modified on March 5, 2026