The Agent Skills Directory

[SAFE]: The skill is focused on professional quality management procedures for medical devices. All analyzed files contain legitimate instructional content and templates.
[COMMAND_EXECUTION]: The provided scripts (scripts/capa_tracker.py and scripts/root_cause_analyzer.py) use standard Python libraries (argparse, json, dataclasses, datetime). They perform local data processing and do not use subprocesses, shell execution, or any other dangerous system-level commands.
[DATA_EXFILTRATION]: No network operations or unauthorized file access patterns were detected. The scripts only read data from user-specified JSON files and output results to the console.
[PROMPT_INJECTION]: The instructions in SKILL.md and the reference documents follow professional standards and do not contain any patterns intended to bypass agent safety filters or override system instructions.
[INDIRECT_PROMPT_INJECTION]: While the scripts ingest external data from JSON files, they do not perform any execution or prompt interpolation that would allow for malicious command injection or agent manipulation based on the content of those files.

capa-officer