SuperAPI AI Directory

risk-identification

Office & productivity

View repository

Updated 10-15 22:12

About this skill

Risk Identification

Overview

This skill supports proactive risk identification in ServiceNow® before full assessment and treatment. It helps teams:

  • Gather risk signals from business applications, issues, and operational events
  • Classify and prioritize risk candidates
  • Route high-priority candidates for formal assessment
  • Maintain governance-ready evidence for risk intake decisions

Use this when risk managers need structured early detection, not only post-event summarization.

Prerequisites

  • Roles: sn_grc.manager, sn_risk.manager, or admin
  • Plugins: GRC / Risk Management enabled
  • Access: Read/write access to risk identification, profile, risk, and issue tables
  • Policy Inputs: Risk taxonomy, scoring thresholds, and escalation criteria

Procedure

Step 1: Confirm Risk Identification Schema

Validate table and fields used in your instance for risk identification workflows.

Using MCP:

Tool: SN-Discover-Table-Schema
Parameters:
  table_name: sn_risk_identification

If your instance uses alternate table names, capture mappings before continuing.

Step 2: Ingest Risk Candidate Signals

Pull new or updated risk candidates and supporting context.

Using MCP:

Tool: SN-Query-Table
Parameters:
  table_name: sn_risk_identification
  query: stateINnew,under_review^sys_updated_on>=javascript:gs.daysAgoStart(7)
  fields: number,short_description,state,category,profile,owner,likelihood,impact,sys_updated_on
  limit: 100

Step 3: Enrich with Profile and Historical Context

Gather profile posture and related historical risk records.

Using MCP:

Tool: SN-Query-Table
Parameters:
  table_name: sn_grc_risk
  query: profile=[profile_sys_id]^active=true
  fields: number,short_description,state,residual_risk,inherent_risk,risk_score
  limit: 50

Step 4: Score and Prioritize Candidates

Apply an intake score based on impact, likelihood, velocity, and control coverage.

Suggested decision bands:

  • High: escalate immediately to formal assessment
  • Medium: assign analyst review with due date
  • Low: monitor and capture rationale

Step 5: Create Follow-Up Actions

Route high/medium items into accountable next steps.

Using MCP:

Tool: SN-Create-Record
Parameters:
  table_name: task
  fields:
    short_description: "Risk identification follow-up"
    description: "Perform formal assessment for high-priority risk candidate."
    priority: 2

Step 6: Record Governance Evidence

Write final intake rationale with scoring factors and escalation decision.

Tool Usage

Tool Purpose
SN-Discover-Table-Schema Validate table names and field availability
SN-Query-Table Retrieve candidate, profile, and historical risk data
SN-Get-Record Inspect individual candidate details
SN-Create-Record Create follow-up tasks/issues for escalated candidates
SN-Update-Record Update candidate state and priority
SN-Add-Work-Notes Capture audit-ready intake rationale

Best Practices

  • Separate risk identification from risk treatment decisions
  • Keep scoring model transparent and versioned
  • Use consistent category mapping to avoid duplicate risk records
  • Time-box analyst review for medium-priority candidates
  • Link every escalation to an owner and target completion date

Troubleshooting

Risk Identification Table Not Found

Symptom: Query fails for sn_risk_identification.
Cause: Instance uses alternate table naming or scoped table variants.
Solution: Use schema discovery and update table mappings for your environment.

Too Many Low-Quality Candidates

Symptom: Intake queue grows without actionable prioritization.
Cause: Risk signal sources are noisy or scoring thresholds are too permissive.
Solution: Tighten intake criteria and require minimum evidence before escalation.

Related Skills

  • grc/risk-assessment-summarization - Summarize formal risk assessments
  • grc/risk-event-summarization - Summarize materialized risk events
  • grc/issue-validator - Validate quality of downstream GRC issues

References