Action types

Form actions

Form Actions ask for input through forms and they are compatible with the form.io framework only. Orchestrator handles form visualization, action assignment, and completion.

According to your use cases, you can further customize your form actions. Visit Working with Form Actions to learn what features you can use inside form actions.

App actions

App actions are custom user actions, similar to Form actions, for which Apps provides the visual interface for rendering the task in Action Center. Apps also offers pre-made templates which allow you to customize out-of-the-box customizations for your requirements.

From Action Center, you can view your Action App, corresponding to your task, by navigating to Admin Settings, and then Tenant. Under Action Apps, you can view an Action App history, and navigate to App Studio to edit it.

Visit Working with App Actions to learn how to use the feature.

External actions

External Actions are handled in external systems, such as JIRA or Salesforce, contributing to the suspend-resume cycle in an Orchestration Process, as external automation can create the Action in Orchestrator and suspend the workflow until the Action is completed. A form schema does not restrict them, and there is no visualization provided for such Actions. Upon completion, data input is mapped back to the workflow variables.

Action assignment is handled externally. Therefore, External Actions cannot be assigned to users.

In case the external automation cannot complete the action, you can do it instead based on an action performed within the corresponding external system. To complete an external action, you need Edit permissions on Actions and access to the associated folder.

Data Labeling actions

Data Labeling Actions handle user input to identify data coming from attached files, and add specific labels to it. The labels indicate what the data is about. These labels are later used to train ML models in AI CenterTM. Data labeling Actions, based on classification, allow users to select one or more classes. Data labeling Actions based on (NER) Named Entity Recognition, allow users to recognize specific entities in data.

When you select Submit inside the Data Labeling action, you save the labels that will later appear in AI Center.

You can create a Data Labeling action either through the Create Labeling Task activity, or by selecting Send to labelers. You can find the Send to labelers button in AI Center, in the Data Labeling page, inside the Label tab.

Document validation actions

Document validation Actions handle user input in order to review and correct automatic data extraction results.

Document validation is performed in Orchestrator using the validation station. At action creation, document data is uploaded into a storage bucket, enabling the user to act upon it. Upon completion of the action, the validated extracted data is downloaded, and the workflow resumes. To delete the downloaded data, automatically, check the Remove Data from Storage box in the Wait For Document Validation Action And Resume activity.

Document classification actions

Document classification Actions handle user input in order to review and correct document classification.

Document classification is performed in Orchestrator using the classification station.

At action creation, document data is uploaded into a storage bucket, enabling the user to act upon it. Upon completion of the action, the validated classified data is downloaded, and the workflow resumes. To delete the downloaded data, automatically, check the Remove Data from Storage box in the Wait For Document Classification Action And Resume activity.

Note:

Starting with Action Center 2021.10, document types used in Document Validation and Document Classification Actions can be created either as single files or as part of a group in their corresponding taxonomy.

Read more in our Document Understanding guide.

Important:

Web browser access to Amazon and Azure storage buckets can sometimes be restricted due to Cross Origin Resource Sharing (CORS) and Content Security Policy (CSP) settings.

Allow identified entities in the web application.

For details on how to configure CORS and CSP settings, click here.