transformer-architecture
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Prompt Injection] (SAFE): No malicious instruction overrides or safety bypass attempts were detected.
- [Data Exposure] (SAFE): No hardcoded credentials or sensitive file access patterns were found.
- [Remote Code Execution] (SAFE): No evidence of external script downloads or package installations.
- [Indirect Prompt Injection] (SAFE): The skill refers to internal markdown files. 1. Ingestion points: references/patterns.md, references/sharp_edges.md, references/validations.md. 2. Boundary markers: Absent. 3. Capability inventory: No code execution or network operations. 4. Sanitization: N/A. No high-privilege capabilities are exposed to ingested data.
Audit Metadata