神經科病歷助手
PassAudited by ClawScan on May 1, 2026.
Overview
This instruction-only skill is coherent and shows no code, install, credential, or persistence behavior, but it processes sensitive medical notes and produces clinical suggestions that should be privacy-controlled and clinician-reviewed.
Before installing, ensure this is used only with patient data you are authorized to process, preferably de-identified and within an approved medical/privacy environment. Use the generated diagnosis and treatment suggestions as clinician-reviewed assistance, not as standalone medical advice.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Patient details could be exposed if the user pastes records into an unapproved model, workspace, or agent memory system.
The skill asks users to provide raw outpatient medical-record text, which may contain sensitive patient information. This is expected for the stated purpose, and no storage or sharing behavior is shown.
1. 輸入門診病歷原始文字
Use only in an approved clinical/privacy environment, de-identify records where possible, and avoid saving outputs to persistent memory unless permitted.
A user could treat generated clinical suggestions as more authoritative than they are, especially outside clinician supervision.
The skill outputs possible diagnoses, further testing suggestions, and treatment plan suggestions, and its listed audience includes broad medical/AI users. This is disclosed and purpose-aligned, but users should not over-trust the output.
- 可能的診斷 - 進一步檢查建議 - 治療方案建議 ## 適用對象 - 神經內科醫生 - 門診護士 - 醫學生成AI愛好者
Treat outputs as drafting or structuring assistance only; diagnoses and treatment plans should be reviewed by a qualified clinician.