> For the complete documentation index, see [llms.txt](https://docs.nexthink.com/platform/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.nexthink.com/platform/user-guide/spark/setting-up-and-managing-spark/managing-spark-data-inputs.md). # Managing Spark data inputs Configure Spark to incorporate organization-specific data inputs into its response generation. This ensures it provides responses aligned with your internal processes and knowledge. As a result, Spark increases automated resolution rates, reduces escalations and rework, and shortens resolution time. This is achieved through the following data inputs: * Knowledge base articles * Service request catalog * Past ticket resolution data * Web search ## Knowledge base articles Knowledge base articles support Spark in providing guidance aligned with company policies and recommended practices. This includes both employee-facing and IT support-related knowledge. Spark can classify articles by intended audience to determine how to use the content during search and resolution. Depending on your ITSM, you have the option to: * Import multiple knowledge bases from your ITSM using CSV files. * Refer to the [Managing knowledge sources](/platform/user-guide/spark/setting-up-and-managing-spark/managing-spark-data-inputs/managing-knowledge-sources.md) documentation for more information. * Ingest the knowledge base from ServiceNow by configuring the relevant inbound connector. * Refer to [ServiceNow Knowledge Base connector](/platform/configuring_nexthink/bringing-data-into-your-nexthink-instance/integrating-nexthink-with-third-party-tools/inbound-connectors/connector-for-servicenow-knowledge-base.md) for more information. ## Service request catalog Service request catalog access enables Spark to guide employees to the appropriate service request and submission process for their request. Configure the ServiceNow request catalog connector to ingest the ServiceNow catalog structure, forms, and request metadata, which Spark can use to identify and recommend service requests and provide associated guidelines. Refer to the [ServiceNow Request Catalog connector](/platform/configuring_nexthink/bringing-data-into-your-nexthink-instance/integrating-nexthink-with-third-party-tools/inbound-connectors/connector-for-servicenow-request-catalog.md) documentation for more information. ## Past ticket resolution data Past ticket resolution data enables Spark to continuously learn from past resolutions. This allows Spark to assist employees in complex scenarios by leveraging previously applied solutions and expert knowledge. Spark can use resolution notes from incidents and service requests previously resolved by support agents to recommend remediations. This capability helps reduce manual intervention and escalations over time. For configuration instructions, refer to the [ServiceNow Tickets connector](/platform/configuring_nexthink/bringing-data-into-your-nexthink-instance/integrating-nexthink-with-third-party-tools/inbound-connectors/servicenow-tickets-connector.md) documentation. ## Web search If Spark cannot find sufficient information in customer-specific sources, such as knowledge base articles or past ticket resolution data, it can search external web sources for additional guidance. These include documentation sites of relevant device manufacturers, application providers, and related parties. Spark always prioritizes customer-provided content and does not override high-confidence internal guidance with external information. Spark also records the accessed URLs in its reasoning logs to support monitoring and troubleshooting. ### Spark trusted websites To ensure control and trust, Spark only searches the Nexthink-defined allowlist of trusted domains. The following allowlist criteria are designed to ensure that approved domains provide trustworthy, relevant, and actionable information that supports Spark’s troubleshooting and support capabilities.
Security * Domain reputation is verified and not associated with malware, phishing, or known CVEs. * HTTPS is enforced with a valid, non-expired certificate. * No known history of domain hijacking or typosquatting activity.
Legitimacy & Ownership * The domain is owned by a known and trusted vendor or official organization, such as Microsoft, Cisco, or NIST. * Domain ownership can be verified through WHOIS records. * The domain is not recently registered or anonymized.
Necessity * The domain serves a clear and documented purpose tied to a specific troubleshooting use case. * The domain is included according to the principle of least privilege and only when required for Spark to perform a specific task.
Data Exposure Risk * The domain does not unintentionally receive sensitive telemetry or credentials. * The domain does not aggregate or resell access logs.
Stability * The domain is stable, long-lived, and maintained by an established provider. * The domain is not ephemeral or anonymously proxied through unmanaged infrastructure. * The domain hosts official documentation or API references rather than unofficial mirrors or reposted content.
Auditability * Each allowed domain has a documented justification and assigned owner. * Allowed domains are reviewed on a defined schedule, such as quarterly
Usefulness * The domain is relevant to technologies and services used by the company. * The domain contains actionable troubleshooting content, such as how-to guides, knowledge base articles, error code references, CLI references, configuration steps, or API documentation. * The domain provides information relevant to the IT support scope, such as endpoints, SaaS applications, identity providers, networking, security tools, or hardware. * The domain includes operational or incident-related information, such as status pages, incident reports, known issues, or Common Vulnerabilities and Exposures (CVE) advisories. * The content is well-structured, versioned, stable, and suitable for automated processing. * Access to the domain helps reduce manual escalations and improve autonomous issue resolution.
{% hint style="info" %} Only domains that meet all allowlist criteria can be used by Spark to provide external troubleshooting guidance in conversations with employees. {% endhint %} {% hint style="info" %} For the complete and up-to-date list of approved domains, log in to Nexthink Community and open [Trusted web domains for Spark web search](https://edocs.nexthink.com/nexthink-infinity/infinity-specifications/trusted-web-domains-for-spark-web-search). {% endhint %} ### Enabling web search in Spark To enable web search in Spark: 1. Open the **Spark** module from the main menu and select **Manage settings** from the navigation panel. 2. Navigate to the **Knowledge and Tools** tab. 3. Select the **Web search** checkbox. 4. Save the changes. Once enabled, Spark can access external websites and search them when internal sources do not provide sufficient guidance.
--- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.nexthink.com/platform/user-guide/spark/setting-up-and-managing-spark/managing-spark-data-inputs.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.