persona-assistant

Pass

Audited by Gen Agent Trust Hub on Aug 3, 2026

Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill accesses sensitive Personally Identifiable Information (PII) including employment type, hire date, job title, and benefits eligibility from the sn_hr_core_profile and sys_user tables. This data is used solely to provide persona-based routing and guidance as described in the skill's primary function.
  • [PROMPT_INJECTION]: The skill processes untrusted data from knowledge base articles (kb_knowledge) and case notes (sn_hr_core_case), presenting a surface for indirect prompt injection.
  • Ingestion points: Data enters the agent context via ServiceNow table queries in Step 4 and Step 6 of the procedure.
  • Boundary markers: The skill does not define explicit delimiters or instructions to ignore potential commands within the retrieved data.
  • Capability inventory: The agent has the ability to create records in ServiceNow and execute shell commands via the Bash tool.
  • Sanitization: No sanitization or filtering logic is specified for the data retrieved from external tables before it is used to generate responses.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 3, 2026, 02:24 PM
Security Audit — agent-trust-hub — persona-assistant