GRC case summarization skill for compliance cases

  • Versão de lançamento: Australia
  • Atualizado 12 de mar. de 2026
  • 3 min. de leitura

    • The GRC case summarization skill uses a large language model (LLM) to generate a structured, AI-powered summary of a compliance case record. The summary is generated on demand from case data and can be saved to the record for future reference.

    Overview of the GRC case summarization skill

    Compliance cases can span multiple action tasks, stakeholders, and regulatory requirements. The GRC case summarization skill generates a concise AI summary of key case details, so case analysts and managers can quickly understand the context and take informed action.

    The skill collects data from predefined fields and related lists across the case record. This data is assembled into a prompt and sent to the configured LLM service provider, which then returns a structured summary.

    To summarize compliance case records, the skill must be activated from the Now Assist Admin console. Once it's activated, case analysts with the appropriate user role can trigger the skill on a compliance case.

    User roles

    The user roles required to activate, configure, and use the GRC case summarization skill are:
    • sn_nowassist_admin.nsa_admin: Grants an admin access to activate or edit a Now Assist skill.
    • sn_comp_case.compliance_case_analyst: Grants access to compliance case records.
    • sn_grc_sharegenai.grc_case_ai_user: Grants users access to use the GRC case summarization skill.
      Nota:
      The sn_grc_sharegenai.grc_case_ai_user role must be explicitly assigned and isn’t inherited through the case analyst role.

    LLM service providers

    An administrator must set a default LLM provider before the skill can be used. The following providers are supported:

    • Azure OpenAI
    • AWS Claude
    • Now LLM Service
    • Now LLM Service LTS
    • Google Gemini

    To set a default provider for the GRC case summarization skill, see Manage model providers.

    Components of a compliance case summary

    The summary reflects case data at the time of generation. As the case progresses, you can regenerate the summary to capture the latest information. Once generated, you can review and edit the summary before saving it to the case record.

    Tabela 1. Compliance case summary structure
    Section What it captures
    Case Overview Core case details, such as name, description, start date, priority, and assigned analyst, captured from the Details tab.
    Events Timeline Date of occurrence, date of discovery, investigation start and end dates, and remediation start and end dates, captured from the Schedule section on the Details tab.
    Scope of Impact Impact breakdown- Summary of the areas and regulatory frameworks that are impacted by a compliance case, captured from the Impacted Areas, Related Areas, and Regulations tabs.
    • Impacted areas- Number of areas, such as the entities, controls, companies, and users, impacted by the reported compliance case.
    • Related areas- Number of areas, such as policies, citations, control objectives, control, and risk events, related to the compliance case.
    • Regulations/Standards- Number of regulations that are or can be impacted by the reported compliance case.
    Causes & Consequences Causes and consequences of the compliance case, captured from the Causes and Consequences tab.
    • Cause (Confirmed)- Confirmed cause of the incident.
    • Contributing Cause (Suspected)- Suspected cause of the incident.
    • Consequences- Resulting impact or penalties arising from the incident.
    Actions & Outcomes Summary of actions taken during the investigation, captured from the Action Tasks tab.
    • Investigation- Initial investigation tasks to examine the reported incident.
    • Assessment- Impact, loss, breach scope, and breach assessments to evaluate the incident.
    • Additional Actions Taken- Supplementary actions to support a resolution.
    Evidence & Worknotes Work notes and comments recorded during the investigation, captured from the Activity section on the Details tab.
    Lessons Learned AI-powered analysis of the case timeline and overall investigation effort.
    • Velocity analysis- Time to detect, investigate, and remediate.
    • Investigation effort level- Effort assessment ranked as low, medium, or high, based on case complexity, activity log, and timelines.