Reviewing your agentic AI assessment

  • Release version: Yokohama
  • Updated September 18, 2025
  • 2 minutes to read
  • This automated assessment process evaluates potential implementation impacts and provides actionable insights for agentic AI in Now Assist for IT Service Management (ITSM) and Now Assist for Customer Service Management (CSM).

    Navigate to Workspaces > Now Assist Readiness Evaluation and select the agentic_AI assessment icon to review your agentic AI assessment.

    Note:
    Before viewing your agentic AI assessments, you must first have run the scheduled job to get the assessment results. If you haven't run your scheduled jobs, you are prompted to do so on an individual job level. For more information on running scheduled jobs, see Run the GenAI/AgenticAI Assessment scheduled job.

    The summary tab provides a consolidated summary of overall findings along with summarization and resolution notes for ITSM and CSM. The results shown are estimates. You should evaluate results provided by Now Assist Readiness Evaluation for accuracy and appropriateness for your use case.

    To better understand the estimated remediation effort, select the Legends button. The legend shows the estimated remediation effort using sizing verbiage and colors, along with an estimated timeline of remediation in development days.

    Figure 1. Example of Legends estimated remediation effort sizing
    Legends button displays estimated remediation efforts such as Small (1-5 days), Medium (6-10 days), Large (11-30 days), XL (31-90 days).

    After reviewing the summary, select the individual assessment tabs to review those specific assessments. For the Agentic AI - ITSM tab, you must first choose a use case from the Select Use Case drop-down menu: Categorize Incidents or Generate change request plans. The CSM page doesn’t have any specific use cases to select. The questions for each assessment differ. For example, the questions and answers on the Agentic AI - ITSM > Categorize Incidents assessment are different than the questions and answers on the Agentic AI - CSM assessment.

    Note:
    The assessments gather information from tables. To view any of the tables' available questions, answers, assessments, or assessment run successes or failures, search for the applicable table in the navigation filter:
    • Questions [sn_assess_question] table
    • Answers [sn_assess_answer] table
    • Assessment [sn_assess_assessment] table
    • Assessment Run [sn_assess_assessment_run] table

    For more information about how to search for tables in the navigation filter, see Navigate directly to a table.

    The assessments include visual indicators to guide you on non-blocker and blocker results. The following list details the descriptions of the assessment's visual indicators:

    • : Non- blocker area and no further action needed.
    • : Blocker area and further action is needed to resolve issue before implementation.
    • : Non-blocker area warning, but the issue should be evaluated before implementation.

    You can select Download Report to export the entire agentic assessment report in PDF format from the Summary tab. The exported report still includes hyperlinks to the specific individual assessments. Selecting Download Report on an individual assessment downloads only that report in PDF format. After selecting a hyperlink in the PDF, that record or table opens in a separate browser tab.

    Note:
    You don’t have to select Download Report to work directly with the problematic records or tables found in the assessment. Each assessment includes hyperlinks to the records or tables that need attention. Selecting a hyperlink opens the record or table in a separate browser tab.
    Figure 2. Example of the Agentic AI - ITSM assessment
    The Agentic AI - ITSM Categorize Incidents use case assessment is shown, along with the user case's relevant questions and records.

    After reviewing the assessment and making the suggested assessment changes, rerun your scheduled jobs to see how your results improve. For more information on rerunning a specific job or all jobs, see Run the GenAI/AgenticAI Assessment scheduled job.