# Bi-Weekly Product Release Notes - Nov 13, 2025
## Features
The following features have been added to the Aisera Gen AI Platform in this release.
### Agent Assist
The following feature has been added to **Agent Assist** in this release.
#### Separate Generate Knowledge and Publish Button Permissions in Salesforce
The Agent Assist channel has been updated to support separate permissions for the **Generate Knowledge** and **Publish** buttons when used in Salesforce. Users with the Salesforce Generate Knowledge Access permission will be able to use the Generate Knowledge button. Users with the Salesforce Publish Knowledge Access Permission will be able to use the Publish button.
These permissions are set within Salesforce using custom permissions. To use these permissions, create or edit a custom user profile and set the following permissions based on the desired access level for that profile.
* AISERA\_TIQ.Generate Knowledge Access
* AISERA\_TIQ.Publish Knowledge Access
A user attempting to access either of these features when they do not possess the required permissions will be notified with a message such as, “`You don’t have permission to generate documents. Contact your admin to get access.`”
### Channels
The following feature has been added to **Channels** in this release.
#### Support for WhatsApp
As described in the [previous Release Notes](https://docs.aisera.com/product-release-notes#new-whats-app-channel), are excited to announce that Aisera has added support for WhatsApp as a new conversational channel.
This integration allows organizations to meet their users and customers right where they are, offering seamless, AI-powered support and automation directly through one of the world's most popular messaging apps. By extending our omnichannel capabilities, Aisera makes it more convenient than ever for users to get instant help, 24/7.
#### New Agent Skills Field for Domain-Specific Agents
You can now add specific skills, keywords, or capability descriptions to the Agent Skills field when creating and configuring domain-specific agents.
When handling a user request, the Universal Agent will now consider this new field along with the existing auto-generated capabilities summary of each agent’s domain. This provides you with an additional layer of governance, and significantly improves the Universal Agent's accuracy when encountering various unanticipated edge cases.
### Live Agent Handler Improvements
This release includes various minor fixes and improvements to our Live Agent handler.
### Workflows and Hyperflows
The following feature has been added to **Workflows and Hyperflows** in this release.
#### New Action Node Introduced
A new **Create Service Request (via data source configuration)** action node has been introduced in the AI Workflow Studio. This node accepts a Data Source through a Data Source ID input parameter and creates a Service Catalog Request.**To use this Action Node :**
1. In the Aisera Admin UI, navigate to **AI Automation > AI Workflow Studio.**
2. Create or edit an existing Workflow.
3. Drag and drop an **Action**node\*\*\*\*from the panel of nodes on the left.
4. Search for \*\*Create Service Request (via data source configuration),\*\*select it from the list, and click **Next.**
5. Type in the remaining configurations and select **OK** to create this node.\
\
This creates the dataSourceId variable node inside your workflow.
6. Add the **Create Service Request** Action Node to your workflow. This Action Node uses the **dataSourceId** as a variable.
7. [Test your Workflow within the AI Workflow Studio](https://docs.aisera.com/aisera-platform/ai-automation-optimize-flow/ai-workflow-studio/testing-workflows) to make sure it is creating Service Requests.
#### Hyperflows now Support Import and Export
Each Application (or bot) on the Aisera platform is designed to handle a specific type of issue or domain operation. Within an Application you can create multiple Hyperflows, combining Prompts, Actions, APIs, and Workflows to perform well defined tasks.
To make your work more efficient you can now export Hyperflows from one Application and import them into another. This allows you to reuse, share, or transfer automation logic across different applications without rebuilding it from scratch.
The Export and Import are handled through System Jobs. Files exported from the platform will be in the .tar.gz file format. Files imported to the platform must also be in the . [tar.gz](http://tar.gz/) file format.
**To Export a Hyperflow:**
1. In the Aisera Admin UI, navigate to **Settings > Aisera GPT.**
2. Select the Application you would like to export from.
3. Navigate to **AI Automation > Hyperflows.**
4. Select the desired Hyperflows to export by checking the box next to the Hyperflow in the list.
5. In the top right click **Actions** **> Export.**
6. Navigate to **Settings > Jobs.**
7. Refresh the page to check the status of the job.
8. When the job is complete, click **Link to download** to download the exported Hyperflow.\
**To Import a Hyperflow:**
1. In the Aisera Admin UI, navigate to **Settings > Aisera GPT.**
2. Select the Application you would like to export from.
3. Navigate to **AI Automation > Hyperflows.**
4. Click **Actions > Import.**
5. Browse to the Hyperflow file on your machine.
6. Navigate to **Settings > Jobs**to monitor the status of the job.
7. When the status is complete navigate back to **AI Automation > Hyperflows.**
8. Click the **Configure Imported Resources** link at the top of the page.
9. Review and adjust any **integration maps**and then click **OK** to begin the Final Import job.
10. Navigate to **Settings > Jobs**to monitor the status of the job.
11. Once the import job has completed successfully, the imported Hyperflows will appear on the **Hyperflows**page.
### Ontology
The following feature has been added to **Ontology** in this release.
#### Ability to Change Entity Parent Class
The Ontology module of the Aisera platform is designed to help the Virtual Agent understand, classify, and respond to user queries more effectively by explicitly adding semantic structure to Domain knowledge.
The Ontology associated with your Agent is a representation of a set of Entities (concepts) within a Domain and the relationships between those Entities.
With this release, you can generate Entities that are structured in terms of a Class Hierarchy rooted under the **Generated** Class.
Entities within a Class are semantically related on the basis of the **is-a** (is of type) **relationship** (such as ‘an admin user **is-a** user role’ or ‘ticket status **is-a**ticket attribute’).
If you are editing an **Entity** in **Review Mode**, where the **Review Status** is Needs Review, you can change the **Entity Class** and select a more appropriate one from the **Entity Classes hierarchy** drop-down menu.
When you revise and **Publish** an **Entity**, the newly-associated **Class** is also **Published**.
## Enhancements
The following enhancements have been added to the Aisera Gen AI Platform in this release.
### Connectors
The following enhancements have been made to **Connectors** in this release.
#### ServiceNow Connector Enhanced for AI Ops
This release introduces a major enhancement to the ServiceNow Connector, enabling a configurable, relation-first ingestion flow for Configuration Management Database (CMDB) data sources. You can now control whether or not the ingestion process starts with CMDB Relationships. If CMDB relations are ingested, then Configuration Items (CIs) and tickets will be filtered based on these relationships, significantly improving performance and filtering for large-scale environments.
The backend logic has been refactored to take this new flow into account. The data ingestion flow is now dynamically controlled depending on the configuration of the Data Source. Because of this, the connector is able to maintain high data integrity while reducing the unnecessary API calls when interacting with large CMDBs.
**To ingest CMDB Relations:**
1. In the Aisera Admin UI navigate to **Settings > Data Source.**
2. Click **+ New Data Source.**
3. Select **ServiceNow**and click **Next.**
4. Input the general parameters for the New Connector. Select **CMDB Learning** under \*\*Functions,\*\*and then click **Next.**
5. Select **CMDB Relations Discovery Path.**
6. Input the remaining parameters and click **Next** here and on the remaining windows of the Data Source configuration.
#### Event Forward Data Sources Field Maps
Event Forwarding Data Sources are data sources using the **Event Forwarding**function. Previously, event forwarding data sources did not show field maps for the data types selected during configuration. You can now view detailed Fields and Field Maps for Event Forwarder data sources, similar to other data source types. The maps are dynamically displayed based on the Data Type you selected during data source creation, to provide greater transparency and easier validation of integration configurations.
This change reduces ambiguity and provides the ability to manually verify the configurations of an event forwarding Data Source by surfacing all relevant field maps in the **Data Source Details Page**.
Navigate to **Settings > Data Sources** and select an event forwarding Data Source to view the maps. The external fields will be displayed at the bottom of the **Data Source Details**page.
### AI Automation
The following feature has been added to **AI Automation** in this release.
#### Event Trigger Condition Tooltip
Event triggers allow you to trigger workflows based when certain criteria are met or events occur. Triggering conditions determine what triggers the event. If no conditions are set all events will be processed by the event trigger. A new tooltip or label has been added to the Triggering Condition field in the Event Trigger UI when configuring **EventForwarder**event triggers. If no specific condition is selected, the UI now clearly indicates that all events will be processed by default. The tooltip message reads: All Events.
This will help prevent confusion during the configuration of filters without triggers.
### Knowledge Generation
The following enhancement has been added to **Knowledge Generation** in this release.
#### Refined Similarity Algorithm
The similarity detection algorithm has been refined to deliver more accurate, context aware matches between AI-generated and existing knowledge articles.
The new approach enhances semantic understanding to identify related content even when the documents vary significantly in length, structure, or phrasing. This improvement enables consistent and precise similarity detection across diverse knowledge formats, resulting in higher accuracy and reliability in knowledge matching.
## Issues Resolved in the Release
| **Aisera ID** | **Product Area** | **Summary** |
| ------------- | ---------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------- |
| 11.12.2025.1 | Workflows | Callback Nodes in subflows timeout is not working properly |
| 11.12.2025.2 | Issue with users concurrently editing workflows affect other users | |
| 11.12.2025.3 | Hyperflows | Hyperflow Details Page - Valid workflow with draft version shows under Pending workflow section |
| 11.12.2025.4 | Hyperflow - UI rendering errors when creating a New Draft for Hyperflow | |
| 11.12.2025.5 | Conversation | Form Node Data Appeared with Duplicate values in case of multiple formNodes Configured in Sequence |
| 11.12.2025.6 | Fix contract source config variable type | |
| 11.12.2025.7 | Ingestion/Connection | Confluence Connector: V1 Rest API for Spaces has been deprecated |
| 11.12.2025.8 | Admin UI: Not sending tenant id correctly when authenticating with OAuth | |
| 11.12.2025.9 | Agent Assist | ServiceNow - \[Agent Assist] New Aisera logo is not updated in Agent Assist widget |
| 11.12.2025.10 | Knowledge Generation | Document generation from single ticket hitting error 'Error generating knowledge. Please contact Administrator' |
| 11.12.2025.11 | Ticket upload task is failing on azure dev. | |
| 11.12.2025.12 | Document publish is failing with "Multiple active KCS template events found" | |
| 11.12.2025.13 | System | Ticket Learning Job Error - RuntimeError: module compiled against API version 0x10 but this version of numpy is 0xf |
| 11.12.2025.14 | Connector Server - IllegalArgumentException: Unknown src: POD NOT FOUND | |