testing-patterns
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- Prompt Injection (SAFE): No patterns of instruction override, safety bypass, or role-play injection were detected.
- Data Exposure & Exfiltration (SAFE): The skill does not reference sensitive system paths, hardcode credentials, or perform network operations.
- Unverifiable Dependencies & Remote Code Execution (SAFE): No external package managers or remote code execution patterns (e.g., curl to bash) are utilized.
- Indirect Prompt Injection (SAFE): The skill serves as a static reference and does not ingest untrusted external data for processing.
- Dynamic Execution (SAFE): No evidence of runtime compilation, unsafe deserialization, or dynamic script generation was found.
Audit Metadata