code-review

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
  • [PROMPT_INJECTION]: Indirect prompt injection surface detected where the skill ingests untrusted script content and possesses write capabilities. * Ingestion points: Script content is retrieved from ServiceNow tables using SN-Query-Table and SN-Get-Record tools in multiple steps and examples. * Boundary markers: No delimiters or instructions to ignore commands within the analyzed scripts are present. * Capability inventory: The skill uses the SN-Update-Record tool and curl -X PATCH commands to modify records in the target instance. * Sanitization: No sanitization of the retrieved script content is performed before processing.
  • [COMMAND_EXECUTION]: The skill utilizes the Bash native tool and curl utility to execute network operations against ServiceNow REST APIs using environment variables.
  • [DATA_EXFILTRATION]: Technical script data including business rules and script includes are retrieved from the ServiceNow instance and processed within the agent context.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 04:36 PM