component-development
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- [Prompt Injection] (SAFE): No override or bypass instructions detected.
- [Data Exposure & Exfiltration] (SAFE): No sensitive file access or unauthorized network operations found.
- [Remote Code Execution] (SAFE): No external script downloads or dynamic execution patterns present.
- [Indirect Prompt Injection] (LOW): While the skill uses Read and Edit tools which create an attack surface, the instructions themselves are benign. Ingestion points: Read, Grep. Boundary markers: Absent. Capability inventory: Edit, Write. Sanitization: Absent.
Audit Metadata