spoke-generation

Warn

Audited by Gen Agent Trust Hub on Aug 3, 2026

Risk Level: MEDIUMCREDENTIALS_UNSAFEREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill instructs the agent to generate JavaScript code for ServiceNow script steps. This code is stored in the sys_hub_script_step table and executed within the ServiceNow environment.
  • [CREDENTIALS_UNSAFE]: The procedure involves creating records in the sys_alias_credential and sys_connection_alias tables, which manage sensitive authentication secrets.
  • [PROMPT_INJECTION]: The skill processes untrusted integration requirements to generate code and configuration, creating an indirect prompt injection surface.
  • Ingestion points: Integration requirements analysis (Step 1).
  • Boundary markers: Absent. No instructions are provided to the agent to delimit or ignore instructions within the requirement data.
  • Capability inventory: Record creation and modification via SN-Create-Record and SN-Update-Record tools.
  • Sanitization: Absent. There is no mention of escaping or validating requirement data before it is used in record creation or script generation.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 3, 2026, 02:24 PM
Security Audit — agent-trust-hub — spoke-generation