Release Notes
Digital Enterprise Suite - Release Notes - November 13 2026
Version: 13.2.0
New and improved features:
- Analytic data can now be accessed and queried with the assistance of an AI model.
- Insight Analyzer now offers a generative AI integration to access the Digital Enterprise Graph using natural language.
- Generation of models using AI was greatly improved.
- A new AI custom FEEL library is available to integrate both generative AI and decisioning AI.
- Testing a single element of a workflow model can now be achieved using the "Test Element" button in the execution ribbon.
- A specific trigger (start event) can now be selected when testing a workflow process.
- AI Performer reasoning is now shown when testing a workflow model.
- Active and completed case instances can now be viewed graphically in the instance viewer.
- When testing models that invoke other models, those models are now automatically opened. This can be disabled by clicking on "Block Models" in the Test ribbon.
- The automatic layout of workflow models has been improved.
- Conditions can now be viewed when testing a workflow model.
- Information icon now appears over the data inputs when testing a decision model.
- A new knowledge source relation type has been added to the Digital Enterprise Graph and they are now displayed in Insight Analyzer.
- The container base image for the Digital Enterprise Suite and the Digital Distributed Container has been changed from alpine to Red Hat Universal Base Image (UBI).
Bug fixes:
- Workflow tokens can't be created, deleted or moved on a call activity in the workflow instance viewer.
- When executing a case using the form interface, the activation and completion of tasks could not be done as two separate operations.
- When testing cases, discretionary tasks need to be added to the case to be available for completion.
- When testing a case, an active process task does not display an active token.
- When testing a workflow process and a token was deleted from a task or moved to another task, the original task remains highlighted as completed.
- A workflow process without proper start conditions could be deployed or tested.
- When creating a new identity for an AI Agent, the test button may fail.
- Testing an AI Agent connectivity would fail on some more recent models.
- Some knowledge sources could have the wrong name in the Digital Enterprise Graph.
- Publishing an older version of model to automation could result in no deployment actually occurring.
- Using the trim FEEL function in a filter did not produce the proper result.
- The service library does not automatically select the new group when a model is published in a new group.
- When mapping data for workflow tasks, the previous entries are now properly available for autocompletion.
- Fixed multiple small inconsistencies with the default data mapping when no expression is provided.
- The PUT operation on the data endpoint of a case should not delete values if a null is passed, only the DELETE operation should do that.
- A BPMN XML import of a model without sourceRef or targetRef on sequence flow failed without a precise error message.
- When consuming messages from a queue (Kafka, RabbitMQ, SQS, ...), messages that fail to generate a process instance should be marked in error.
- The terms auto-linking does not always bring the codes associated with the terms.
- Link labels could move to their default position when a very small movement of an anchor occurred.
- AI model generation now asks for a confirmation before replacing the current diagram.
- In a rare case, the Decision Table analysis would provide just null as an explanation instead of the actual root cause.
- REST service calls returning a plain JSON element outside of an object would not always parse properly.
- When the email server is not configured, it is not possible anymore to send an email invite for places.
- Null values could be incorrectly rejected by the validation when they were allowed by an enumeration.
- Some datastore persistence function could silently fail to persist the data under rare conditions.
- Difference could sometime be reported in error when comparing imported models.
- Some Business Knowledge Packages (BKP) containing files edited prior to version 13 could fail to generate.
- Under some rare conditions, the button to complete a user task in the email notification would have a relative URL that did not work to complete the task.
Learn how it works
Request DemoConfirm your budget
Request PricingDiscuss your project
Request Meeting
We would like to use third party cookies to improve the functionality of this website.