The Agent Skills Directory

[SAFE]: No security issues detected across the analyzed files. The skill follows best practices by separating logic into a specialized script and providing clear ethical constraints.- [COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python utility (scripts/bazi_calc.py) to perform date conversions and chart calculations. The script uses standard libraries and performs basic input validation by casting arguments to integers.- [PROMPT_INJECTION]: The skill contains no instructions attempting to bypass safety filters or override agent behavior. On the contrary, it includes an explicit ETHICS.md and ethics sections in SKILL.md that provide guidance on handling sensitive topics like mental health or crisis situations.- [DATA_EXPOSURE]: No hardcoded credentials, sensitive file paths, or unauthorized data transmission patterns were found. All calculations are performed locally without network dependencies.

bazi-mingli