portfolio-algorithmic-trading

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external CSV data, which constitutes an indirect prompt injection surface. The impact is limited by the lack of dangerous capabilities in the skill's environment.\n
  • Ingestion points: scripts/portfolio_algorithmic_trading_diagnostics.py reads data from a user-provided input.csv file.\n
  • Boundary markers: No explicit delimiters or instructions are used to distinguish untrusted data from valid diagnostic output.\n
  • Capability inventory: The skill is restricted to local file operations and statistical calculations; it contains no networking, subprocess execution, or dynamic code evaluation (eval/exec) capabilities.\n
  • Sanitization: The script enforces numeric types for data columns, but CSV header names are reflected directly into the JSON summary without validation.\n- [SAFE]: The Python diagnostic script uses established libraries and follows deterministic logic. All operations are consistent with the skill's stated purpose for financial quantitative research.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 08:46 AM
Security Audit — agent-trust-hub — portfolio-algorithmic-trading