unified-reafference
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWNO_CODE
Full Analysis
- [Prompt Injection] (SAFE): No instructions found that attempt to bypass safety filters or override agent behavior.
- [Data Exposure & Exfiltration] (SAFE): No file access or network communication patterns identified.
- [Unverifiable Dependencies & Remote Code Execution] (SAFE): No external packages or remote code execution vectors are present.
- [Indirect Prompt Injection] (SAFE): While describing a database concept, the skill provides no functional code or data ingestion surfaces that could be exploited.
Audit Metadata