Security audit
Security checks across malware telemetry and agentic risk
The skill broadly matches its stated video-analysis purpose, but it needs Review because it sends sensitive shelter footage to cloud services while silently creating, reusing, and storing identity/token data.
Install only in an authorized emergency-response setting after reviewing the configured endpoints, especially the dev/private HTTP URLs. Treat use of this skill as remote processing of sensitive shelter footage and crisis reports, and confirm consent/legal authority, tenant scoping, retention limits, face blurring, and secure credential storage before using real footage.
if filters:
for key, value in filters.items():
query = query.filter(getattr(self.__model__, key) == value)
if offset:
query = query.offset(offset)if filters:
for key, value in filters.items():
query = query.filter(getattr(self.__model__, key) == value)
return query.scalar()
finally:"source": ConstantEnum.DEFAULT__SKILL_HUB_NAME
}
try:
_response = requests.post(_url, json=_data)
if _response.status_code == 200:
_response_json = _response.json()
if _response_json and _response_json.get("success"):| 🔎 使用提醒 | 应急救援场景采用"**宁可多触发不可漏报**"原则，但高危预警必须**人工复核**后再升级到救援调度，避免误判造成现场骚动 | | 🔎 使用提醒 | 儿童 / 老人 / 孕妇 / 残障人士等脆弱群体阈值降一档，系统更敏感 | | 🔎 使用提醒 | 应注意区分正常疲倦休息（静坐）与木僵：木僵需 ≥ 5 分钟且对外界刺激**无响应** | | 🧑⚖️ 结果性质 | 红线约束：**禁止**输出 ASD / PTSD 等临床诊断；**禁止**给予药物建议；**禁止**长期存储原始视频（≤ 7 天清理，仅留聚合事件日志）；**禁止**将受灾人群视频用于媒体传播 / 社交媒体 / 商业研究 | | 🔎 使用提醒 | 公共指挥屏展示必须做**面部模糊化**处理（保护受灾者尊严） | | 🔎 使用提醒 | 合规要点：必须经由**应急指挥中心 / 卫健委授权部署**，配合**现场持证心理救援人员**（中国心理学会临床心理学注册委员会注册人员、红十字心理救援队等）使用；遵守《突发事件应对法》《精神卫生法》 | | 📁 格式支持 | 任何预警都附 **PFA 6 步要点**（建立连接 → 安全保障 → 平静化 → 联系亲友 → 实际支持 → 转介资源）+ 转介资源（当地精神卫生中心 / 12320 / 400-161-9995） |
VirusTotal findings are pending for this skill version.
Detected: suspicious.install_untrusted_source