Samtree

AI-driven multi-agent medical education platform for academic discussion, knowledge library management, knowledge tree generation, and learning history tracking.

项目背景与来源 / Project Background

• 本项目由 浙江大学基础医学院实验教学中心人工智能教学分中心 支持开发。
• 本项目为 NVIDIA DGX Spark 全栈 AI 开发黑客松大赛 参赛作品。

版权声明 / Copyright Notice

本仓库（mtc0205/Sam-tree）及其包含的所有源代码、文档、资源文件均由 Sam tree团队 开发并拥有完整版权。

本项目并非开源项目。 除非另有书面许可，任何个人或组织：

1. 禁止 将本项目代码用于任何商业用途。
2. 禁止 对本项目代码进行修改、二次开发或作为其他项目的组成部分。
3. 禁止 在未经作者许可的情况下进行二次分发或公开传播。

允许的行为：

• 仅限于出于学习、评价或学术交流目的查看、下载（Clone/Download）及在本地环境中运行。
• 如果你在学术论文中引用了本项目的思路或架构，请务必注明出处并联系作者。

---

All rights reserved. The source code, documentation, and assets in this repository (mtc0205/Sam-tree) are the intellectual property of Sam tree.

This is NOT an open-source project. Unless otherwise authorized in writing:

1. Commercial use of any part of this project is STRICTLY PROHIBITED.
2. Modification, derivation, or integration into other projects is PROHIBITED.
3. Redistribution or public dissemination without prior permission is PROHIBITED.

Permitted Actions:

• Viewing, cloning, and running the code locally for educational or academic review purposes only.
• If you reference the concepts or architecture of this project in academic publications, please provide proper attribution and contact the author.

---

联系方式 / Contact: e-mail：wcc180220@outlook.com

Evaluation Quick Start

1) Install dependencies

pip install -r requirements.txt
cd interface && npm install && cd ..

2) Start services (single entry)

python start_all_servers.py

3) Health check and stop (single entry)

python scripts/ops/check_services.py
python scripts/ops/stop_services.py

Default web entry: http://localhost:3000

Core Service Ports

• 3000: Next.js frontend
• 8080: Academic chat API
• 8081: Knowledge library API
• 8082: Auth API
• 8083: Knowledge tree API
• 8084: Assistant API

Repository Layout

Samtree/
├── README.md
├── NVIDIA_DGX_Spark_Hackathon_项目报告书.md
├── 02_部署指南.md
├── agents/
├── core/
├── interface/
├── deploy/
├── examples/
├── openspec/
├── socratic-platform/        # historical prototype (not primary workflow)
├── fs ui/                    # legacy UI exploration (not primary workflow)
├── interface_backup/         # backup snapshot (not primary workflow)
├── scripts/
│   ├── ops/
│   ├── smoke/
│   ├── verify/
│   └── archive/
├── docs/
│   ├── README.md
│   ├── guides/
│   └── archive/
└── tests/

Test Commands

• Unit tests:
  • pytest tests -v
• Smoke tests (services must be running):
  • python scripts/smoke/test_api.py
  • python scripts/smoke/test_login.py

Documentation

• Project report: NVIDIA_DGX_Spark_Hackathon_项目报告书.md
• Deployment guide: 02_部署指南.md
• Full docs index: docs/README.md

For Judges

• Product walkthrough: NVIDIA_DGX_Spark_Hackathon_项目报告书.md
• One-command startup: python start_all_servers.py
• Quick verification: python scripts/ops/check_services.py
• Smoke checks: python scripts/smoke/test_api.py and python scripts/smoke/test_login.py
• Deployment reference: 02_部署指南.md

Frontend Source Of Truth

• Active frontend for evaluation: interface/
• fs ui/ and interface_backup/ are retained for historical traceability only and are not required to run the current system.

Notes for Reviewers

• The primary operational flow uses only three commands:
  • start: python start_all_servers.py
  • health: python scripts/ops/check_services.py
  • stop: python scripts/ops/stop_services.py
• Legacy scripts are kept in scripts/archive/ for traceability and are not part of the recommended workflow.