itsm-suggested-steps

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection because it ingests and processes untrusted data from multiple external sources without sufficient sanitization or boundary markers.
  • Ingestion points: Untrusted data enters the agent context from the incident table (short_description, description), the kb_knowledge table (text), and historical resolution data (close_notes) found in similar incidents.
  • Boundary markers: The instructions do not define clear delimiters or use "ignore instructions" warnings when interpolating retrieved incident data into the troubleshooting logic.
  • Capability inventory: The skill possesses the capability to perform network operations via Bash (curl/REST API) and write content back to the ITSM system using the SN-Add-Work-Notes tool and PATCH requests.
  • Sanitization: No sanitization, escaping, or validation of the ingested external content is mentioned before it is processed to generate the structured resolution plan.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 01:34 PM
Security Audit — agent-trust-hub — itsm-suggested-steps