ai-ml-data-science
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill demonstrates a strong security posture by utilizing standard Python libraries for its utility scripts (ml_toolkit.py and leakage_scan.py), avoiding the risks associated with untrusted third-party dependencies or remote code execution.
- [INDIRECT_PROMPT_INJECTION]: The ml_toolkit.py script processes user-supplied JSON specifications to generate Markdown reports. This creates an attack surface for indirect prompt injection where malicious Markdown or text could be embedded in report fields like 'model_name' or 'description'. While the current risk is low as the output is static documentation, it is recommended to sanitize external inputs or use explicit boundary markers when interpolating untrusted data into reports.
Audit Metadata