skills/serac-labs/serac/virtual-agent/Gen Agent Trust Hub

virtual-agent

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill generates ServiceNow scripts that ingest untrusted user input (e.g., inputs.user_email, inputs.user_question) and interpolate them into database queries and record creation logic.
  • Ingestion points: Script blocks in SKILL.md process inputs from the inputs object.
  • Boundary markers: None identified in the script templates.
  • Capability inventory: The skill utilizes GlideRecord.query(), GlideRecord.insert(), and several snow_ tools for artifact management.
  • Sanitization: The examples rely on platform-native GlideRecord escaping but do not demonstrate explicit validation or sanitization of user-provided strings.
  • [DYNAMIC_EXECUTION]: The skill facilitates the creation of 'Script Blocks' containing server-side JavaScript (ES5) that is assembled at runtime.
  • Script generation: String concatenation and template-like structures are used to generate functional code blocks for ServiceNow execution (e.g., verify_user and createIncidentScript in SKILL.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 07:16 PM
Security Audit — agent-trust-hub — virtual-agent