醫療病歷結構化助手

自動將門診或住院病歷內容轉換為結構化摘要，包括主訴、病史、檢查發現、診斷建議及治療方案。

This skill appears coherent and low-risk from a technical-permission perspective, but consider safety and privacy before using on real patient data: 1) The source/homepage is unknown—vet the author if possible. 2) Clinical outputs should be reviewed by a qualified clinician; models can hallucinate or give unsafe treatment suggestions. 3) Ensure use complies with applicable privacy/regulatory rules (e.g., de-identify PHI or follow HIPAA policies). 4) Limit logging and storage of any clinical text, and restrict who can invoke the skill. 5) If you plan to deploy at scale, add prompt-level safety checks, audit trails, and clinical validation steps.

Type: OpenClaw Skill Name: medical-note-assistant Version: 1.0.0 The skill bundle is a straightforward medical note structuring assistant designed to format clinical information into standard categories like Chief Complaint and Assessment. It contains only metadata and descriptive instructions in Traditional Chinese; there is no executable code, network activity, or evidence of malicious prompt injection. Files analyzed: SKILL.md, skill.yaml, and _meta.json.