impact-analysis

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the SN-Execute-Background-Script tool to perform recursive dependency mapping and risk calculations. This involves the agent generating server-side GlideScript from templates and executing it within the ServiceNow instance. This high-privilege capability is used as intended for advanced automation.
  • [PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks because it processes metadata (CI names, class names, and descriptions) from an external ServiceNow CMDB.
  • Ingestion points: Data enters the agent context through tools like SN-Query-Table, SN-List-CmdbCis, and SN-Get-Record (referenced in SKILL.md).
  • Boundary markers: The procedure relies on standard markdown headers and code blocks but lacks specific delimiters or instructions to disregard potential commands embedded within CI data.
  • Capability inventory: The skill possesses significant capabilities, including server-side code execution (SN-Execute-Background-Script) and the ability to modify records (SN-Update-Record) in the connected environment.
  • Sanitization: There is no evidence of sanitization or validation of the retrieved CMDB content before it is used in subsequent logic or included in the final CAB reports.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 04:36 PM