algorithmic-trading-dma-handbook

Pass

Audited by Gen Agent Trust Hub on Jun 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a local Python script, scripts/algorithmic_trading_dma_handbook_diagnostics.py, to perform quantitative analysis on trading datasets. This execution is scoped to the skill's purpose of generating performance diagnostics.\n- [PROMPT_INJECTION]: The skill features a data ingestion surface through input.csv, which is processed to generate diagnostics.json. This represents an indirect prompt injection surface where untrusted data is brought into the agent's context.\n
  • Ingestion points: input.csv (referenced in SKILL.md and processed by scripts/algorithmic_trading_dma_handbook_diagnostics.py)\n
  • Boundary markers: Absent\n
  • Capability inventory: Local script execution, file system read (CSV), and file system write (JSON)\n
  • Sanitization: The script performs basic data type validation (numeric and datetime conversion), but does not sanitize content for embedded instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 27, 2026, 07:05 PM
Security Audit — agent-trust-hub — algorithmic-trading-dma-handbook