stock-signal-analyzer
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from a CSV file in scripts/compute_indicators.py. Although no explicit boundary markers are used, the risk is mitigated by strong sanitization: the script enforces numeric types on all input columns using pd.to_numeric with errors='coerce', which removes text-based instructions. The skill also lacks network and dynamic execution capabilities.
- [SAFE]: The tool performs local mathematical calculations using standard libraries and contains no malicious code or suspicious network activity.
Audit Metadata