← Back to Skills Marketplace
unisound-llm

unisound-followup-record

by Unisound-LLM · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
66
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install unisound-followup-record
Description
将中文门诊复诊病历文本结构化为细粒度字段，输出 JSON（如现病史/既往史/诊断/处理意见等）。
README (SKILL.md)

门诊复诊病历结构化

概述

给定一份中文门诊 复诊病历文本（医生已书写），本技能抽取并规范化为细粒度字段 JSON，例如：

  • 现病史.病情概述
  • 现病史.药物
  • 现病史.其他治疗措施
  • 现病史.病情转归
  • 现病史.一般情况
  • 既往史.疾病
  • 既往史.其他信息
  • 既往史.手术史
  • 既往史.过敏史
  • 既往史.输血史
  • 婚育史
  • 月经史
  • 个人史
  • 家族史
  • 查体
  • 辅助检查
  • 诊断
  • 处理意见.药物
  • 处理意见.其他建议

数据安全、隐私与伦理声明

  • 最小必要原则：仅处理结构化抽取所必需的文本内容；不要求也不鼓励包含姓名、证件号、手机号、详细地址等身份信息。
  • 严格脱敏：在发送至任何模型/接口前，会对可识别个人身份的信息进行脱敏/去标识化处理（如姓名、证件号、手机号、详细地址、人脸/影像等）。仅传递脱敏后的必要信息用于本次 skill 调用。
  • 不做本地持久化：不将用户输入与中间结果写入本地持久化存储（包含磁盘文件、数据库、日志）。仅在内存中短暂处理；本次调用结束即销毁。
  • 第三方 API 风险提示：在功能需要时，可能会调用第三方模型/服务接口；此时仅会发送脱敏后的必要信息，并使用加密传输。除完成本次请求外，不用于任何其他用途（如训练、画像、营销）。
  • 医疗边界：本技能输出为文本抽取与结构化结果，不构成医疗诊断或治疗建议；如涉及临床判断请以执业医生意见为准。

输入格式

纯文本病历（UTF-8），可包含如下分段：

主诉：…… 现病史：…… 既往史：…… 婚育史：…… 月经史：…… 个人史：…… 家族史：…… 查体：…… 辅助检查：…… 诊断：…… 处理：……

也支持通过统一入口 scripts/run.py 直接输入 pdf/doc/docx/xls/xlsx/csv/txt/json。 预处理成功后，会先归一化为标准复诊病历文本，再调用本 skill 的原始结构化逻辑。

快速开始

# JSON 结构化输入
python doctor/emr-gen/followup-record/scripts/run.py \
  --input data/med-followup-record/gen_records.json \
  --appkey \x3Cyour-appkey>

# 普通病历文件（纯文本）
python doctor/emr-gen/followup-record/scripts/run.py \
  --input data/med-followup-record/record.txt \
  --appkey \x3Cyour-appkey>

参数说明

  • --input PATH：必填。输入 JSON 文件或病历文本文件路径。
  • --input-type auto|pdf|doc|docx|xls|xlsx|csv|txt|json：输入类型，默认 auto。
  • --sheet STRING：读取 Excel 时指定 sheet（可选）。
  • --encoding STRING：txt/csv 编码，默认 utf-8。
  • --base URL：内部大模型 base URL，默认 https://maas-api.hivoice.cn/v1。
  • --model STRING：模型名称，默认 u1-insuremed。
  • --timeout SECONDS：HTTP 超时秒数；0 表示一直等待，默认 0。
  • --appkey STRING：必填。内部医疗大模型鉴权 key，使用 Bearer 方式认证。
  • --output-json PATH：可选。保存输出 JSON。
  • --output PATH：可选：输出病历文本文件路径。
  • --save-prepared：可选：保存预处理后的文本，便于调试。

输出约定

  • 输出为 UTF-8 文本，每行格式：字段：值
  • 如果某个字段在原文中没有对应内容，返回 未提及
  • 标准化表达：如无过敏史、无手术史等统一简化为 无

依赖

前置 Skill

scripts/run.py 依赖 _shared/doc-preprocess 提供的公共文件预处理库（preprocess.py）。 请确保 _shared/doc-preprocess/ 位于 skills/ 根目录下。

运行环境

  • Python 3.7+

外部 API

  • 内部医疗大模型：https://maas-api.hivoice.cn/v1/chat/completions
    • 方法：POST，OpenAI 兼容格式
    • 需要传入 --appkey 参数进行 Bearer 认证

Python 第三方包（可选，run.py 使用非 txt/json 输入时需要）

包名 用途 必要条件
openpyxl 读取 .xlsx 文件 输入为 xlsx 时必须
pypdf 提取 PDF 文本 输入为 pdf 时必须

安装：pip install openpyxl pypdf

仅使用 TXT/JSON 输入时，无需安装任何额外包。

测试命令

从 skills 根目录执行：

# 离线自测（检查输入和构造请求）
python self_tests/med-followup-record-gen/self_test_followup_record_gen.py

# 在线自测（调用内部接口）
python self_tests/med-followup-record-gen/self_test_followup_record_gen.py --run-network
Usage Guidance
Use this skill only if you trust the publisher, the hivoice model endpoint, and the shared preprocessing dependency. Manually remove names, IDs, phone numbers, addresses, and other identifiers before use; protect the app key; avoid save-prepared/output files unless you can secure them; and verify the generated structured fields against the original medical record.
Capability Analysis
Type: OpenClaw Skill Name: unisound-followup-record Version: 1.0.0 The skill bundle is a legitimate tool for structuring Chinese medical records into JSON format using the Unisound (hivoice.cn) medical LLM API. The Python script `scripts/run.py` implements standard text processing, multi-stage LLM prompting, and HTTP POST requests to a documented endpoint (maas-api.hivoice.cn) without any evidence of data exfiltration, malicious execution, or unauthorized access to sensitive system files.
Capability Assessment
⚠ Purpose & Capability
The stated purpose—structuring Chinese follow-up medical records—is coherent, but the visible code sends record-derived prompts to a model API and does not show the promised de-identification step.
ℹ Instruction Scope
The extraction prompts are narrow and task-aligned, but raw document text is inserted into prompts, so adversarial content inside an input record could influence the structured output.
ℹ Install Mechanism
There is no install spec or required binary, but the script imports an external shared preprocessing helper outside this package and documents optional unpinned Python package installs.
⚠ Credentials
User-selected medical files and a medical LLM call are purpose-aligned, but the data is highly sensitive and the artifacts do not clearly demonstrate the claimed redaction before remote processing.
⚠ Persistence & Privilege
No background service or privilege escalation is shown, but the documentation claims no local persistence while also offering output and save-prepared options that can write processed medical text to disk.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install unisound-followup-record
  3. After installation, invoke the skill by name or use /unisound-followup-record
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release: structure Chinese outpatient follow-up medical records into detailed JSON fields. - Extracts key sections such as present history, past history, physical examination, diagnosis, and treatment advice. - Handles multiple input formats (txt, json, pdf, doc, docx, xls/xlsx, csv) via a command-line tool. - Ensures strict de-identification, data privacy, and no local data persistence. - Returns standardized field outputs; missing fields are marked as "未提及". - Requires Python 3.7+; optional dependencies for non-txt/json files. - Relies on an internal medical language model with API key authentication.
Metadata
Slug unisound-followup-record
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is unisound-followup-record?

将中文门诊复诊病历文本结构化为细粒度字段，输出 JSON（如现病史/既往史/诊断/处理意见等）。 It is an AI Agent Skill for Claude Code / OpenClaw, with 66 downloads so far.

How do I install unisound-followup-record?

Run "/install unisound-followup-record" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is unisound-followup-record free?

Yes, unisound-followup-record is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does unisound-followup-record support?

unisound-followup-record is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created unisound-followup-record?

It is built and maintained by Unisound-LLM (@unisound-llm); the current version is v1.0.0.

More Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463815 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259802 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200680 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191345 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190333 · by oswalpalash

💬 Comments