Shift Handover Summarizer
Generate structured shift handover summaries from EHR updates, highlighting critical events that occurred during the shift.
Clinical Disclaimer: This tool generates summaries for handover support only. All clinical decisions must be verified by qualified medical staff. Patient data must comply with applicable data protection regulations (e.g., HIPAA).
Quick Check
CODEBLOCK0
When to Use
- - Use this skill when generating a structured handover summary from EHR records at the end of a clinical shift.
- Use this skill when prioritizing patients by event severity for incoming staff.
- Do not use this skill as a substitute for direct clinical handover, real-time patient assessment, or emergency triage.
Workflow
- 1. Confirm the patient records file, shift start/end times, and optional department filter.
- Validate that the input records are within the declared shift time range.
- Timezone validation: If
--shift-start or --shift-end lacks a timezone offset (e.g., 2026-02-06T00:00:00 without Z or +HH:MM), emit a warning: "Shift times appear to lack a timezone offset. Assuming UTC. Specify timezone explicitly (e.g., 2026-02-06T00:00:00+08:00) to avoid incorrect event filtering." - Run the summarizer script or apply the manual extraction path.
- Return a structured summary with patients ranked by priority, key events, and pending tasks.
- If inputs are incomplete, state exactly which fields are missing and request only the minimum additional information.
Usage
CODEBLOCK1
Parameters
| Parameter | Type | Required | Description |
|---|
| INLINECODE6 | file path | Yes | JSON file of EHR records for the shift |
| INLINECODE7 |
ISO 8601 | Yes | Shift start time |
|
--shift-end | ISO 8601 | Yes | Shift end time |
|
--department | string | No | Department filter |
|
--output | file path | No | Output file path (default: stdout) |
|
--no-vitals | flag | No | Exclude vital signs summary |
Event Priority Levels
| Priority | Event Type |
|---|
| High | Resuscitation, deterioration, serious complications, abnormal vitals |
| Medium |
New symptoms, abnormal findings, medication adjustments, special procedures |
| Low | Routine treatment, condition improvement, daily care |
Output
- - Shift summary with total patients and critical patient count
- Per-patient priority ranking, key events, vitals summary, medication summary, and pending tasks
- Plain-text handover narrative
Scope Boundaries
- - This skill processes structured EHR records; it does not access live hospital systems.
- Event classification is based on preset thresholds and keywords; adjust thresholds for department-specific needs.
- This skill does not replace direct verbal handover or physician sign-off.
Stress-Case Rules
For complex multi-constraint requests, always include these explicit blocks:
- 1. Assumptions
- Shift Period and Inputs Used
- Summary Output
- Critical Flags
- Risks and Manual Checks
Error Handling
- - If required inputs are missing, state exactly which fields are missing and request only the minimum additional information.
- If the task goes outside the documented scope, stop instead of guessing or silently widening the assignment.
- If
scripts/main.py fails, report the failure point, summarize what still can be completed safely, and provide a manual fallback. - Do not fabricate patient data, clinical events, or execution outcomes.
Input Validation
This skill accepts: a structured EHR records file with shift start and end times for handover summary generation.
If the request does not involve shift handover summary generation from EHR records — for example, asking for real-time patient monitoring, clinical diagnosis, or direct treatment recommendations — do not proceed with the workflow. Instead respond:
"shift-handover-summarizer is designed to generate structured handover summaries from EHR records. Your request appears to be outside this scope. Please provide a records file and shift times, or use a more appropriate clinical tool."
Response Template
Use the following fixed structure for non-trivial requests:
- 1. Objective
- Inputs Received
- Assumptions
- Workflow
- Deliverable
- Risks and Limits
- Next Checks
If the request is simple, you may compress the structure, but still keep assumptions and limits explicit when they affect correctness.
班次交接摘要生成器
从电子健康记录更新中生成结构化的班次交接摘要,突出显示班次期间发生的关键事件。
临床免责声明: 本工具仅用于生成交接支持摘要。所有临床决策必须由合格医务人员核实。患者数据必须遵守适用的数据保护法规(如HIPAA)。
快速检查
bash
python -m py_compile scripts/main.py
python scripts/main.py --help
使用时机
- - 在临床班次结束时,需要从电子健康记录生成结构化交接摘要时使用此技能。
- 需要按事件严重程度对患者进行优先级排序以便交接给接班人员时使用此技能。
- 请勿将此技能用作直接临床交接、实时患者评估或紧急分诊的替代方案。
工作流程
- 1. 确认患者记录文件、班次开始/结束时间以及可选的科室筛选条件。
- 验证输入记录是否在声明的班次时间范围内。
- 时区验证: 如果--shift-start或--shift-end缺少时区偏移量(例如,2026-02-06T00:00:00没有Z或+HH:MM),则发出警告:班次时间似乎缺少时区偏移量。默认使用UTC。请明确指定时区(例如2026-02-06T00:00:00+08:00)以避免事件筛选错误。
- 运行摘要生成脚本或应用手动提取路径。
- 返回结构化摘要,按优先级对患者排序,包含关键事件和待办任务。
- 如果输入不完整,明确指出缺少哪些字段,并仅请求最少量的补充信息。
使用方法
text
python scripts/main.py \
--records data/shift_records.json \
--shift-start 2026-02-06T00:00:00Z \
--shift-end 2026-02-06T08:00:00Z \
--department Cardiology \
--output summary.json
参数
| 参数 | 类型 | 必填 | 描述 |
|---|
| --records | 文件路径 | 是 | 班次电子健康记录JSON文件 |
| --shift-start |
ISO 8601 | 是 | 班次开始时间 |
| --shift-end | ISO 8601 | 是 | 班次结束时间 |
| --department | 字符串 | 否 | 科室筛选条件 |
| --output | 文件路径 | 否 | 输出文件路径(默认:标准输出) |
| --no-vitals | 标志 | 否 | 排除生命体征摘要 |
事件优先级级别
| 优先级 | 事件类型 |
|---|
| 高 | 复苏、病情恶化、严重并发症、异常生命体征 |
| 中 |
新症状、异常发现、药物调整、特殊操作 |
| 低 | 常规治疗、病情改善、日常护理 |
输出
- - 班次摘要,包含患者总数和危重患者数量
- 每位患者的优先级排序、关键事件、生命体征摘要、用药摘要和待办任务
- 纯文本交接叙述
范围边界
- - 此技能处理结构化电子健康记录;不访问实时医院系统。
- 事件分类基于预设阈值和关键词;可根据科室特定需求调整阈值。
- 此技能不能替代直接口头交接或医生签字确认。
压力案例规则
对于复杂的多约束请求,始终包含以下明确模块:
- 1. 假设条件
- 班次时段和使用的输入
- 摘要输出
- 关键标记
- 风险与手动检查
错误处理
- - 如果缺少必要输入,明确指出缺少哪些字段,并仅请求最少量的补充信息。
- 如果任务超出文档范围,则停止执行,而不是猜测或默默扩大任务范围。
- 如果scripts/main.py执行失败,报告失败点,总结仍可安全完成的内容,并提供手动回退方案。
- 不得捏造患者数据、临床事件或执行结果。
输入验证
此技能接受:包含班次开始和结束时间的结构化电子健康记录文件,用于生成交接摘要。
如果请求不涉及从电子健康记录生成班次交接摘要——例如,要求实时患者监测、临床诊断或直接治疗建议——则不执行工作流程。而是回复:
shift-handover-summarizer旨在从电子健康记录生成结构化交接摘要。您的请求似乎超出此范围。请提供记录文件和班次时间,或使用更合适的临床工具。
响应模板
对于非简单请求,使用以下固定结构:
- 1. 目标
- 收到的输入
- 假设条件
- 工作流程
- 交付物
- 风险与限制
- 后续检查
如果请求简单,可以压缩结构,但当假设条件和限制影响正确性时,仍需明确说明。