backtest
Pass
Audited by Gen Agent Trust Hub on Jul 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local shell script
scripts/upbit-strategy-toolkit.shto run backtests. This is the primary function of the skill and is controlled via defined arguments. - [EXTERNAL_DOWNLOADS]: The shell script
scripts/upbit-strategy-toolkit.shusesuvxto download and run the toolkit from the official Upbit GitHub repository (github.com/upbit-official/upbit-strategy-toolkit.git) if the local environment is not already set up. This is a trusted vendor source. - [PROMPT_INJECTION]: The skill implements robust defensive instructions (Guardrails) to prevent the agent from providing investment recommendations, recalculating metrics, or attempting live trading. It includes specific Korean and English response templates to enforce these boundaries.
- [DATA_EXPOSURE]: The skill manages data within a defined
{data_root}(either the repo root or a hidden home directory folder). It does not attempt to access sensitive system paths or exfiltrate private credentials.
Audit Metadata