The Agent Skills Directory

[SAFE]: The skill provides expert guidance and a local evaluation script for financial backtesting analysis. All code logic is contained within the skill's own scripts, and no external network dependencies or remote code downloads were detected.
[COMMAND_EXECUTION]: The skill includes a Python script (scripts/evaluate_backtest.py) that performs numerical calculations on user-provided trading metrics. The script uses standard libraries, performs input validation (range and type checks), and writes evaluation reports to a local directory. This is expected functionality for the skill's purpose.
[DATA_EXFILTRATION]: No network access or data exfiltration patterns were identified. The script operates entirely on local data provided as command-line arguments and outputs results to the local filesystem.
[PROMPT_INJECTION]: The instructions follow a structured workflow for strategy development and validation. No attempts to override agent safety guidelines or extract system prompts were found.
[DYNAMIC_EXECUTION]: The test suite uses importlib to load the evaluation script for unit testing. This is a standard Python testing pattern and does not represent a security risk as it targets a specific local file.

backtest-expert