predictive-intelligence

Pass

Audited by Gen Agent Trust Hub on Feb 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides legitimate tools and code examples for ServiceNow machine learning development. All identified tools (snow_query_table, snow_execute_script_with_output) and scripts serve the primary purpose of managing Predictive Intelligence solutions.
  • [PROMPT_INJECTION]: Although the skill processes data from ServiceNow records (such as incidents) which are technically untrusted inputs, this is an inherent part of machine learning classification and prediction workflows. The ServiceNow Predictive Intelligence APIs (sn_ml) used here process text as data for analysis, not as executable commands, mitigating the risk of indirect prompt injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 23, 2026, 08:21 PM