Hydra actions provide the option to run a pre-defined task after each Scanner invocation using data from the Scanner output. As an example, you can use a Hydra action to create a Salesforce opportunity object using the data extracted from an RFP document. Action configuration is a part of the Scanner configuration. You have the option to configure one or many Actions for a given Scanner. If you have multiple Actions enabled for a Scanner, they are executed in the order shown on the UI.
Actions have three types of data available to them:
Label data: these are the data points from the Scanner output (i.e. predictions)
Action parameter data: each Action identifies a list of input parameters specific to them. As an example Send An Email Through SendGrid Action uses: From Name, From Email, To Name, To Email and Template as input parameters.
Context data: As Actions execute their tasks, they collect additional data points, which are made available to any downstream actions. As an example, Move Box Document to a Shared Folder adds box_shared_folder_link to the context data set. This data point is then available to any other actions that are run afterward. You can refer to each Action's documentation to identify the context data generated and made available by that Action.
Actions can use data from any of the above data types. The specific data points used depends on the Action.