# Bi-Weekly Product Release Notes - Oct 1, 2025 ## Features The following features have been added to the Aisera Gen AI Platform in this release. ### Agent Assist The following features have been added to **Agent Assist** in this release. #### Single Ticket Knowledge Ticket Generation During issue investigation, IT agents often encounter scenarios where no existing Aisera Answer exists, there is no Next Best Action, or the agent has already added detailed investigation notes. In such cases, waiting for a batch job to trigger knowledge generation delays the process. Because of this, IT agents can now manually trigger knowledge generation from Agent Assist. Agents can also publish this information directly to their **System of Record (SOR)**. This is currently supported only for the ServiceNow data source. **On-Demand Knowledge Generation** IT agents can now manually generate knowledge documents from individual tickets, irrespective of its status, directly from Agent Assist. Tickets which have already been used to generate articles, whether through batch processing or manual processing, will be directly visible in the Generate Knowledge widget. **To trigger On-Demand Knowledge Generation:** 1. Navigate to the **Agent Assist** widget within ServiceNow. 2. Open a ticket with investigation notes. 3. Click the **Generate Knowledge** button in the Agent Assist widget to Generate a knowledge document. A processing icon will appear while the document is being generated. The system will automatically refresh based on the number of attempts configured. If the document takes longer than the configured refresh attempts, you will be notified to check back later or navigate to the **Settings > Jobs** window to view the status. If 30 minutes pass and the document is still not available, the **Generate Knowledge** button will be re-enabled for another attempt. The first successfully generated document will always be retained. **Publishing to the System of Record** You can navigate to the Aisera Admin UI directly by clicking on a document title. within your Agent Assist widget. Before proceeding to the next steps, make sure you have permission to edit, publish, or add tags to your knowledge articles. You also need permission to access the Aisera Admin UI and view the **Knowledge Document Details** window. **To Publish a document to the SOR :** 1. Navigate to the **Agent Assist** widget within ServiceNow. 2. Select the document you just generated. 3. Click the **Publish** icon in the top-right corner. **To go to the Knowledge Document Details page in the Aisera Admin UI:** 1. Navigate to the **Agent Assist** widget in ServiceNow. 2. Locate and click on the **Document Title**. ### Knowledge Article Generation Previously, when configuring Confluence data sources with directory filtering, only the child pages under a specified directory were ingested—parent directory pages themselves were skipped. The ingestion process now includes the parent directory page along with all its child pages. This ensures that all relevant knowledge base articles, including top-level directory content, are available for search and retrieval. #### Resolution Summary for Document Chunks This release includes a **Resolution Summary** under the **Validated Chunk** section of the Knowledge Generation page to help you validate documents against the Ticket Comments. The **Resolution Summary** provides a brief overview of the resolution steps extracted from ticket comments. This gives you more visibility into the solutions mentioned in the comments and helps you compare the **Summary** with the actual resolution steps. **To Validate a Chunk from Ticket Comments:** 1. Navigate to **Content Generation > Knowledge Generation** in the Aisera Admin UI. 2. Choose a generated document. 3. In the side panel, open the **Validated Chunk** section. 4. You will now see a new section called **Resolution Summary**, which displays the extracted resolution details from the ticket comments, along with **Ticket Comments** (the raw comments extracted from the ticket). 5. This feature is available for all tickets included in the cluster.
### Hyperflows The following features have been added to **Hyperflows** in this release. #### Hyperflow Prompt Template Integration A Prompt Template is generated when a Hyperflow is created. This template combines the Hyperflow Description and Use Cases provided upon creation of the Hyperflow with the general best practices for creating a Hyperflow. After creation, you are free to edit, remove, replace, or refine the prompt to shape the Agent’s behavior. **To view and customize the Prompt Template :** 1. In the Aisera Admin UI navigate to **AI Automation > Hyperflows.** 2. Click on **+ New Hyperflow** in the top right corner. 3. Input the **Name**, **Description**, and optionally **Use Cases**, and click **OK** to create a new Hyperflow.
4. A template prompt is visible within the prompt window.
#### Separate Editing for Hyperflow Details and Prompt In previous releases, editing Hyperflow details and a Hyperflow prompt has been part of a single process. Now the editing has been separated into two distinct actions. You may now edit the Hyperflow **name**, **description**, and **use cases** directly from the details window by clicking the **Pencil** icon next to **Details**.
To edit the prompt, from the Hyperflow Details screen click **New Draft**.
Click the **Pencil** icon next to the word, **Prompt** to start editing the draft. A new editing window has been added to the application to provide a more convenient experience while editing Hyperflow prompts. When you complete editing the prompt, press the\*\* Publish\*\* button to save the changes. ## Enhancements The following enhancements have been added to the Aisera Gen AI Platform in this release. ### Hyperflows The following enhancements have been added to **Hyperflows** in this release. #### Hyperflows in Workflow Used By Column Hyperflows now appear in the Used By Column of Workflows.
#### Go Back Option in Hyperflows The **Go Back** option is now available for use in a Workflow when it is being executed through the Hyperflow. ### Connectors The following feature has been added to **Connectors** in this release. #### Generic Connector Now Supports Mutual Transport Layer Security (mTLS) With this release, the Aisera Gen AI platform supports the ability for both the client and the server to verify each other's identity using digital certificates, when your data source supports the mTLS protocol. In this release the mTLS protocol is available for use with the Generic Connector and the Workflow Action Node. **To configure mTLS with the Generic Connector:** 1. Navigate to **Settings > Integrations** in your Aisera Admin UI. 2. Choose the **Generic** Connector icon.
3. Click **Next**. 4. **Name** the Integration and write in the **Endpoint**. 5. Click **Next**. 6. Choose **OAuthClientCredentials** as the **Auth Type**. 7. Leave the **HTTP Operation** with the default value of **POST**.
8. Click **OK**. 9. The bottom portion of the New Integration Authentication parameters contains the mTLS options. The first two options are for the server authentication and the last four fields represent the client options. The server’s certificate is verified using a Certified Authority. If the servers certificate is self-signed or it cannot be verified through a CA, then you can use one of the two server options: * Provide the trusted Server SSL Certificate * Check the **Trust any SSL Server Certificate** checkbox
10. For Client verification, you need a bundle that contains a private key and a (public) certificate. This bundle can be also protected by a pass-phrase. The most popular text base format for these certificates are in PEM format [Privacy-Enhanced Mail](https://en.wikipedia.org/wiki/Privacy-Enhanced_Mail) in which the certificates are base64 encoded and prefixed by a header and a footer: ``` -----BEGIN PRIVATE KEY----- PRIVATE KEY IN BASE64 -----END PRIVATE KEY----- ``` Use the following field options for PEM format certificates: * Private key to use for client (Client Private Key in Pem Format) * Certificate to use for client (Client Certificate in Pem Format) * Optional passphrase (Client Private Key Passphrase)
You can also use a PEM certificate format known as .p12. This is also supported by a field on the New IntegrationAuthentication window: * Add a path to the file inside the pod (Path to Client Certificate in p12 Format)
All connector or REST call action requests made using this integration (even during authorization) will use information from any of the above fields. **Workflow Actions Node** After you have created a Generic Integration, you can use a similar set of options to point to a **.p12** certificate using the Custom Certificate field in the Edit Action Node window of the AI Workflow Studio.
You can add the client private key and certificate to be sent to the requests by checking the Use Client SSL Cert checkbox.The client’s key and certificate will be loaded when the actions-server starts. #### Freshworks Connector Sub-Folder Crawling The Freshworks connector knowledge Base ingestion logic has been enhanced to support crawling subfolders. Previously, only articles within top-level folders were ingested. The connector now recursively discovers and ingests articles from all subfolder levels, ensuring complete knowledge base coverage within hierarchical folder structures. #### Confluence Data Source Supports Parent Directory Filtering The Confluence connector now supports ingestion of both parent directory pages and their child pages when directory filtering is used. Previously, when configuring Confluence data sources with directory filtering, only the child pages under a specified directory were ingested—parent directory pages were not included. With this enhancement, the ingestion process now includes the parent directory page along with all its child pages. This ensures that all relevant knowledge base articles, including top-level directory content, are available for search and retrieval.
In the screenshot above, the **Spaces** parameter displays the parent directories. The **Space Directories** parameters allow you to add the sub-directory names. With this enhancement, data from the parent spaces as well as the sub-directories and sub-pages will be ingested. #### Export Audit Logs to Splunk The Aisera Gen AI platform now supports real-time export of audit logs directly to Splunk, enabling you to integrate Aisera audit data with your existing Security Information and Event Management (SIEM) infrastructure for enhanced security monitoring and compliance. **The Export Audit Logs Feature Includes:** * Real-time streaming of audit events to Splunk via HTTP Event Collector (HEC) * Configurable export options for event types and delivery schedules * Secure, PII/PHI-masked log format in JSON * Automated error detection and administrator alerting * Supports core audit event categories (authentication, configuration, account management, job execution, document management) You can now achieve immediate visibility into platform activities within your Splunk dashboards, improving threat detection, compliance, and operational efficiency. This feature reduces manual export overhead and supports scalable, enterprise-ready security operations. **To setup the audit log export for Splunk:** **Create the Splunk Integration:** 1. Navigate to the **Settings > Integrations** page in the Aisera Admin UI. 2. Click the **+ New Integration** button. 3. Search for and choose the **Splunk** icon. 4. **Name** the integration. 5. Provide the HTTP Event Collector (HEC) endpoint. 6. Click **Next**. 7. Set the authentication type and fill in the parameters. **Add the Splunk Data Source** 1. Navigate to **Settings > Data Source** in the Aisera Admin UI. 2. Choose **+ New Data Source**. 3. Search for and select the **Splunk** icon. 4. Choose **Event Trigger** as the function.
5. Choose **Audit History** and **Audit History Data Type** as the Data Types. 6. Click **Next**. **Set the Event Triggers** 1. Navigate to **AI Automation > Event Studio** in the Aisera Admin UI. 2. Click the **+ New Event Trigger** button. 3. Add a **Name** for the Trigger.
4. Select the new **Event Forwarder** option for the **Event Handler Type**. 5. Choose **Next**. 6. Set the **Event Forwarder** options.
7. Select **Splunk** as the **External System**. 8. Select your Splunk Data Source as the **Data Sources** option.
9. Choose **Audit History** as the **Data Type**. 10. Select **AuditEvent** as the **Event Type**. 11. Click **OK**. **Set the Delivery Schedule:** 1. Navigate to the **Audit Trail** page in the Aisera Admin UI. 2. Select your Splunk integration, choose event types, and set delivery schedules (real-time or batch). 3. Validation and Monitoring 4. Test the configuration to ensure connectivity. 5. Monitor export status and error alerts in the Admin UI. **Activation** Enable the export configuration to start streaming logs. **Known Limitation:** Only core audit events are supported in this release. #### OAuth 2.0 now supports Proof Key for Code Exchange (PKCE) You can now set OAuth 2.0 authentication with Proof Key for Code Exchange (PKCE) in the Integration configuration window of the Aisera Admin UI. Enabling this option sets the platform to handle the PKCE flow, requiring no additional configuration steps other than to enable the option. This enables a secure, standards-compliant authentication flow if you are using Okta or other providers requiring PKCE.
### User Permissions The following enhancements have been made to the **Use Permissions**in this release. #### Privacy Access added to Role Based Access Controls A new **Privacy Access** setting is now available in the Role Management UI, enabling administrators to control the visibility of Personally Identifiable Information (PII) at the role level. This permission can be set to **Read** or **None**. This new Role Based Access Control (RBAC) has defaulted to **None**.
When a role’s **Privacy Access** permission is set to **None**, user identities will be masked across the platform for that role. For example, emails will be masked if a user has **None** permissions. Information will also be masked or removed when downloading campaign reports. **To view or edit this new RBAC:** 1. From the Aisera Admin UI, open **Settings > User Accounts**from the left navigation menu 2. Select the **Roles** tab at the top of the **User Accounts** window 3. Select the role you want to modify 4. Click on the **pencil icon** in the top right to see the **Edit Role** window 5. Navigate through the **Edit Role** by clicking **Next** to locate the **Privacy Access** permission.
#### Data Type Permissions Update The **Data Type** permissions have been updated. Users with **Write** permissions will be able to view and edit data types. Users with **Read** permissions will be able to view Data Types, but will be unable to access the edit modal and make changes. The **Data Types**section within the Aisera Admin UI will be hidden for users with **None** permissions to Data Types. **To view or edit this RBAC:** 1. From the Aisera Admin UI, open **Settings > User Accounts** from the left navigation menu. 2. Select the **Roles** tab at the top of the **User Accounts** window. 3. Select the role you want to modify. 4. Click on the **pencil icon** in the top right to see the **Edit Role** window. 5. Navigate through the **Edit Role** by clicking **Next** to locate the **Data Types** permission.
## Known Issues The following known issue has been identified in this release. ### MS Teams Has Deprecated Support for Multi-Tenant Bots There have been reports of a streaming issue when using MS Teams Multi-Tenant Bots with the Aisera Gen AI platform. This is because Microsoft no longer supports Multi-Tenant Bots. If you’re using a Multi-Tenant Bot, remove it and create a Single Tenant Bot to replace it. **To Create a Single Tenant Bot:** 1. Select Single Tenant as the Type of App when you create the Azure Bot. 2. Choose the checkbox next to Is Single Tenant Bot? In the Aisera New Channel Configuration window. See [**Microsoft (MS) Teams (Azure Bot)**](https://docs.aisera.com/aisera-platform/channels/microsoft-ms-teams-azure-bot) for detailed instructions. ## Issues Resolved in the Release
Aisera IDProduct AreaSummary
10.1.1FlowsWebchat format issue with RAG responses via Hyperflow
10.1.2Approval/Expiry not working as expected in the Ask Approval node.
10.1.3AI Lens and Workflow test mode returning different results for the same test.
10.1.4Box workflow not displaying file owner properly
10.1.5KB GenExisting images from message node are not shown in bot response
9.10.27Validated chunks little info, whereas the solution has many steps
10.1.6ConversationIssue with nudging notifications on Webchat
10.1.7Webchat transcript sometimes missing or incomplete
10.1.8Webchat fails to load when domain whitelist is enforced
10.1.9HTML tags appearing in Copilot Notifications
10.1.10Agent AssistGraph Not Rendering for Agent Feedback When Closed Reason Filter Is Applied
9.10.37Aisera Admin UIWhen exporting to CSV, not all fields are exported
10.1.11Conversational AIEntering only the ticket number does not fetch ticket details on Conv AI 2.0
10.1.12Ingestion & ConnectorsData Source Ingestion Issue where KBs are missed during Ingestion
10.1.13Fresh Knowledge Base ingestion jobs could run indefinitely without importing articles, due to outdated rate limit handling
10.1.14Recycle Token and V2 check box functionality is not working while creating Integration