advocate-sourcing
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- NO_CODE (SAFE): The skill consists entirely of descriptive text and templates. There are no scripts, binaries, or command executions provided.\n- DATA_EXFILTRATION (SAFE): No sensitive file paths, credentials, or unauthorized network communication patterns were identified.\n- PROMPT_INJECTION (SAFE): No malicious instructions or attempts to bypass agent safety filters were found in the skill text or metadata.\n- INDIRECT_PROMPT_INJECTION (SAFE): The workflow describes collecting customer signals (NPS, support interactions), which represents an ingestion surface, but since no automation or code is provided, there is no exploit path. Evidence: 1. Ingestion points: Signal Collection (SKILL.md); 2. Boundary markers: Absent; 3. Capability inventory: None; 4. Sanitization: Absent.
Audit Metadata