legal-request-triage
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it processes and acts upon untrusted data from ServiceNow legal requests.\n
- Ingestion points: The agent is instructed to analyze the
short_descriptionanddescriptionfields from thesn_legal_requesttable in SKILL.md.\n - Boundary markers: Absent. There are no delimiters or instructions to ignore or isolate instructions that might be embedded within the legal request text.\n
- Capability inventory: The agent has the authority to update legal requests (
SN-Update-Record), create new legal cases, and append work notes (SN-Add-Work-Notes).\n - Sanitization: Absent. The skill does not define any steps to validate or sanitize the incoming text before it is used for classification and prioritization decisions.
Audit Metadata