中医舌像分析

专业中医舌象分析系统，基于深度学习自动识别舌色、舌形、舌苔特征，提供中医辨证建议。适用于中医临床诊断辅助、中医教学、健康监测等场景。使用当用户需要分析舌象照片、获取中医辨证建议、或进行舌诊学习时触发。

Security Scan

The skill and included code are coherent with a local, offline tongue-image analysis tool; no unexpected credentials, network exfiltration, or privileged persistence were found, though there are minor documentation/packaging inconsistencies to be aware of.

Purpose & Capability

The name, description, SKILL.md and the Python scripts align: this is a tongue-image analysis and TCM-diagnosis helper. Small inconsistencies: SKILL.md shows CLI commands as `tcm-tongue` but the repo provides tongue_analyzer.py (no packaging/console entrypoint provided), the manifest/package scripts mention a price (6 RMB) while SKILL.md/README claim MIT-0; SKILL.md lists heavy ML deps (TensorFlow/PyTorch) but manifest's packages list is minimal (opencv/pillow/numpy). These are documentation/packaging mismatches, not evidence of malicious behavior.

Instruction Scope

Runtime instructions tell the agent to accept image paths/folders, install typical imaging/ML Python packages, and run local CLI scripts. The included code only accesses files the user specifies, performs local image-file checks and simulated analysis, and generates reports. There are no instructions to read unrelated system files or to send data to external endpoints.

Install Mechanism

There is no network-based install spec and no arbitrary download/extract. The package includes local Python scripts and a packaging helper. The packaging script operates on local files and references npx clawhub commands for publishing, but nothing in the repo fetches or executes remote code at runtime.

Credentials

The skill declares no required environment variables, no credentials, and no config paths. The code does not attempt to access secrets or unrelated environment variables. Requested dependencies (OpenCV, Pillow, numpy, optional TF/PyTorch) are appropriate for image analysis.

Persistence & Privilege

The skill is not force-included (always: false) and does not request persistent elevated privileges or modify other skills' configs. It runs as local scripts and prints/writes reports only when invoked.

Assessment

This package appears to be a local/offline tongue-image analysis tool and is internally consistent. Before installing: 1) Be aware the repository provides Python scripts but not a packaged console entrypoint — running the SKILL.md 'tcm-tongue' commands may require manual installation or invoking python tongue_analyzer.py. 2) Heavy ML libs (TensorFlow/PyTorch) may be optional but large — confirm which are actually needed and install in a virtual environment. 3) The manifest mentions a 6 RMB price while README claims MIT-0; verify licensing and payment expectations with the publisher before paying. 4) The tool processes image files you provide — treat sensitive health photos carefully and test in a sandbox if you are concerned about accidental uploads. 5) No network endpoints or credential access were found in the code, but verify any build/publish steps you run (e.g., npx clawhub publish) before executing them.

中医舌像分析技能

产品概述

中医舌像分析是一款专业的中医舌诊辅助工具，通过人工智能技术自动分析舌象照片，提供准确的中医辨证建议。

核心价值

• 专业准确：基于经典中医理论和大量临床数据
• 操作简便：一键上传，自动分析
• 快速高效：30秒内完成全面分析
• 学习辅助：详细解释分析结果

功能特性

1. 舌色自动分类

• 淡白舌：阳虚、气血不足
• 淡红舌：正常或轻度气虚
• 红舌：热证、阴虚火旺
• 绛舌：营血热盛
• 紫舌：血瘀、寒凝

2. 舌形识别

• 胖大舌：脾肾阳虚、水湿内停
• 瘦薄舌：阴血亏虚
• 齿痕舌：脾虚湿盛
• 裂纹舌：阴液亏耗

3. 舌苔分析

• 苔色：白、黄、灰、黑
• 苔质：薄、厚、润、燥、腻、腐
• 分布：均匀、剥落、地图舌

4. 特殊特征识别

• 瘀点瘀斑：血瘀证
• 舌下络脉：青紫、迂曲
• 舌乳头：充血、萎缩

使用方法

基本命令

# 分析单张舌象照片
tcm-tongue analyze --image "舌象照片路径"

# 批量分析
tcm-tongue batch --folder "舌象照片文件夹"

# 生成详细报告
tcm-tongue report --image "照片路径" --output "报告文件"

参数选项

• --image：舌象照片路径（支持JPG、PNG格式）
• --format：输出格式（text/json/html）
• --detail：详细程度（basic/standard/detailed）
• --compare：对比分析多张照片

输出示例

基本分析结果

舌象分析报告
============

患者信息：匿名
分析时间：2026-03-29 12:30:00

【舌色】淡红舌（轻度气虚）
【舌形】轻度齿痕（脾虚湿盛）
【舌苔】薄白苔（正常或表证）
【瘀点】无

中医辨证：
1. 脾气虚弱证
2. 轻度湿浊内停

治疗建议：
1. 主方：运中补土
2. 辅方：祛湿降浊
3. 穴位：足三里、阴陵泉、中脘、脾俞

JSON格式输出

{
  "analysis": {
    "tongue_color": "淡红",
    "tongue_shape": "轻度齿痕",
    "coating": "薄白",
    "special_features": []
  },
  "diagnosis": [
    "脾气虚弱证",
    "轻度湿浊内停"
  ],
  "recommendations": {
    "formulas": ["运中补土", "祛湿降浊"],
    "acupoints": ["足三里", "阴陵泉", "中脘", "脾俞"]
  }
}

技术原理

图像处理流程

1. 预处理：标准化光线、角度、大小
2. 舌体分割：分离舌体与背景
3. 特征提取：颜色、纹理、形状特征
4. 分类识别：基于深度学习模型分类
5. 辨证推理：结合中医知识库推理

准确率指标

• 舌色分类准确率：92%
• 舌形识别准确率：88%
• 舌苔分析准确率：85%
• 整体辨证准确率：80%

应用场景

临床诊断辅助

• 中医门诊快速舌诊
• 远程医疗舌象分析
• 治疗前后对比评估

教学培训

• 中医学生舌诊练习
• 典型舌象案例库
• 辨证思维训练

健康管理

• 个人健康监测
• 体质辨识
• 养生调理指导

安装要求

系统要求

• Python 3.8+
• OpenCV 4.5+
• TensorFlow 2.8+ 或 PyTorch 1.10+

依赖安装

pip install opencv-python pillow numpy
pip install tensorflow scikit-learn

更新日志

v1.0.0 (2026-03-29)

• 初始版本发布
• 基础舌色分类功能
• 简单辨证建议

v1.1.0 (计划中)

• 增加舌形识别
• 优化辨证算法
• 添加批量处理功能

许可证与使用

许可证

• 许可证类型：MIT-0（完全免费）
• 使用权限：可自由使用、修改、分发
• 商业用途：允许商业用途，无需授权

免费声明

本技能完全免费，旨在促进中医数字化和AI辅助诊断的发展。欢迎中医师、学生、研究者使用和改进。

技术支持

如有问题或建议，请联系：

• 邮箱：[e-mail]
• 网站：https://tcm-tongue.clawhub.ai
• 社区讨论：欢迎在ClawHub社区分享使用经验

---

免责声明：本工具为中医诊断辅助工具，不能替代专业医师诊断。临床决策请咨询执业中医师。