optimize

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses Write and Bash tools to create and execute Python scripts for backtesting. This behavior is consistent with the skill's primary purpose of automating strategy optimization.
  • [EXTERNAL_DOWNLOADS]: The generated scripts download historical market data from the OpenAlgo API, which is a legitimate and expected network operation for financial analysis tools.
  • [PROMPT_INJECTION]: This finding analyzes the attack surface for Indirect Prompt Injection. Ingestion points: The skill ingests user-provided arguments (strategy, symbol) and external data from the OpenAlgo API. Boundary markers: No explicit delimiters are used to wrap external data, though the data is primarily processed as numerical arrays. Capability inventory: The agent possesses Read, Write, and Bash capabilities, which are used to generate and save local reports. Sanitization: Input parameters are used to define file names and strategy types without explicit validation, though they are processed in a structured script environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 07:21 PM