The Agent Skills Directory

[SAFE]: The skill provides structured guidelines and a local script for evaluating trading backtest results. No malicious patterns, obfuscation, or unauthorized network operations were detected.
[COMMAND_EXECUTION]: The skill uses a local Python script evaluate_backtest.py to calculate scores based on user-provided metrics. The script utilizes standard libraries (argparse, json, pathlib) to perform calculations and save reports to a local reports/ directory.
[DYNAMIC_EXECUTION]: The test file conftest.py employs importlib.util to load the main script for unit testing purposes. This is a standard development practice for testing modules and does not present a risk in the skill's operational context.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied numeric metrics. The evaluation script implements input validation via the validate_inputs function, which enforces constraints on values like win rates (0-100) and trade counts, ensuring the data remains within expected logical boundaries.
Ingestion points: Command-line arguments in evaluate_backtest.py.
Boundary markers: Not applicable for numeric metric inputs.
Capability inventory: Local file writing (pathlib.Path.write_text) to the reports/ directory.
Sanitization: Validation logic in validate_inputs() ensures all inputs are positive numbers and win rates are within percentage limits.

backtest-expert