医疗设备问题报告生成器

This skill is internally coherent and the bundled Python script is small and readable, but take these precautions before installing or enabling it in production: - Manually inspect SKILL.md for hidden unicode control characters (the static scanner flagged them). Remove or clarify any suspicious invisible characters. - Review the included script (scripts/generate_issue_report.py) locally — it is self-contained and only formats user input into Markdown, which appears safe. Verify it does not call external networks or read sensitive files (it does not in the provided version). - If the platform will execute the example shell command, ensure the agent runtime escapes user-provided text (to avoid shell command injection) or prefer invoking the Python script with safe argument passing rather than via an interpolated shell line. - Avoid pasting real patient-identifiable data into the skill unless you have confirmed your environment and policies cover protected health information (PHI). Prefer redacted examples for testing. - Because of the prompt-injection flag, consider testing the skill in an isolated environment first and request the author to explain/remove the control characters. If you want, I can show the exact lines in SKILL.md that contain control characters and suggest sanitized replacements.