The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a bundled Python script scripts/analyze_log.py to process workout data. The script is self-contained, uses standard Python libraries, and lacks dangerous functions like eval() or exec(). It specifically processes the local log.jsonl file to generate training trends.
[EXTERNAL_DOWNLOADS]: The skill references several external resources for health guidelines and fitness formulas, including the Israeli Ministry of Health, NASM, and Wikipedia. These are well-known, authoritative sources and are provided for informational purposes only. The skill does not perform any automated background downloads or remote code execution.
[DATA_EXPOSURE]: User training history and profiles are stored in a dedicated workout-coach/ folder within the agent's working directory. This is standard practice for maintaining persistence in AI agent skills. No access to sensitive system files or credentials was observed.
[SAFE]: The skill demonstrates high safety standards by implementing mandatory health screenings, medical emergency routing (MCD 101), and specific guardrails for heat-related illness and disordered eating. It clearly defines its scope as a training assistant and not a medical or dietetic substitute.

israeli-workout-coach