For the complete documentation index, see llms.txt. This page is also available as Markdown.

Setting up and managing Spark

This page describes how to configure Nexthink Spark, including the required technical prerequisites and step-by-step setup instructions.

Before you begin

Nexthink Administrators must complete the following technical prerequisites before setting up Nexthink Spark.

  • To ensure Spark can identify the user:

    • Ensure all employee devices have Collector configured to gather the User Principal Name (UPN) in clear text (set to Enabled in the device configuration profile). Refer to Configuring Collector-level anonymization for more information.

    • Configure the Microsoft Entra ID inbound connector for your Microsoft tenant. Refer to Microsoft Entra ID (Azure AD) connector for more information.

      • Confirm that the Microsoft Entra ID connector's field mapping uses the userPrincipalName attribute for employee name identification (this can differ from the onPremisesUserPrincipalName or mail attribute).

  • To use Microsoft Teams or Microsoft 365 Copilot as a communication channel:

  • To use a third-party enterprise chatbot as a communication channel:

    • Configure Spark Agent2Agent integration to allow employees to stay within the same chatbot experience.

    • Configure Spark Handoff API to transfer conversations from MS Teams to Spark while preserving context.

Enabling Spark to escalate conversations to a human agent requires additional prerequisites, including administrator rights and ITSM connector credentials. Refer to Configuring ticket escalation with Spark > Prerequisites for more information.


Setting up Nexthink Spark

Ensure that you meet the prerequisites before setting up Nexthink Spark. Follow these steps to ensure Spark is set up correctly:

1

Set up a communication channel for Spark interactions

Set up a Communication channel in Nexthink to enable Spark interaction with MS Teams and M365 Copilot. A single communication channel supports both integrations.

  • Use the welcome message to inform employees about the Spark scope and remind them to exercise judgment when reading AI-generated replies.

  • After setting up the communication channel, install the Spark version-specific application package (.zip).

2

Grant permissions

To configure and monitor Spark, update your roles to include the required permissions:

  • Set the Agent conversations NQL table visibility to Visible for the relevant roles.

  • Enable Spark permissions for the relevant roles.

For detailed information about available permissions, View domain options, Data model visibility and Data privacy granularity settings, see:

3

Import knowledge base articles

Ensure Spark can effectively support employees by providing access to your IT knowledge base, just as you would for human support agents. This includes both employee-facing and IT support-related knowledge.

Depending on your ITSM, you have the option to:

4

Import service request catalog

Configure the ServiceNow Request Catalog connector to enable Spark to map the employee’s intent to the appropriate service request. Spark then assists the employee with submission by sharing links to the request and its associated guidelines.

5

Enable diagnosis and remediation actions for Spark

Validate and enable actions for Spark in Nexthink.

Enable built-in Agent actions

From the main navigation menu:

  1. Go to Spark > Manage actions and review the Agent actions designed to work with Spark.

  2. Enable the desired Agent actions for Spark use.

Refer to the Managing Spark actions documentation for more information.

Activate custom remote actions of workflows

Configure custom or Nexthink Library remote actions or workflows and enable them in Spark by selecting Spark trigger method and activating them in Spark cockpit.

Refer to the Remote actions and workflows for Spark documentation for more information

6

Enable Spark to continuously learn from resolved tickets

To enable Spark to continuously learn from resolved tickets in ServiceNow, configure the ServiceNow Tickets Connector to ingest past incident and service request data.

Refer to the ServiceNow Tickets connector documentation for more information.

7

Configure ticket escalation with Spark

Configure Spark to automatically create ITSM tickets when an issue cannot be resolved. Integrate Nexthink with your ITSM platform and define how Spark generates escalation tickets. This integration ensures that unresolved issues are escalated with the relevant context.

Refer to the Configuring ticket escalation with Spark documentation for more information.

8

Enable web search

Enable web search to allow Spark access to a curated list of trusted web domains when internal knowledge sources do not provide sufficient guidance.

Refer to Managing Spark settings and data inputs documentation for more information.

9

Deploy Spark to employees and communicate the rollout

Select the employee group for Spark deployment, deploy Spark to them, and prepare communications.

  • Use the controls in the MS Teams admin console to select the employees with Spark access.

  • Inform employees about the scope of the Spark agent and remind them to exercise judgment when reading AI-generated replies.

Last updated

Was this helpful?