Ai Health Vault

用 AI + Obsidian 为自己和家人搭建私有化健康管理系统。模板 + Prompt + 搭建指南。

312 Stars
GitHub

AI Health Vault

English | 中文

Build a private, AI-powered health management system for yourself and your family using Obsidian.

Not an app. Not a code project. It's a set of templates + prompts + setup guides. Clone it, let your AI (Claude / ChatGPT / Gemini) read it, and it will guide you through building your family's health archive step by step.

Why

  • Medical reports scattered across hospitals — no one tracks trends for you
  • Your parents can't remember their own medication, dosages, or follow-up dates
  • Commercial health apps require uploading your family's private health data
  • You just need local templates where the data stays on your machine

30-Minute Setup

1. Fork this repo (or download ZIP)
2. Open the vault/ folder in Obsidian
3. Fill in family member names in "健康管理中心.md"
4. Take a photo of a medical report → send to AI → AI fills in the template
5. Done. Repeat step 4 for every checkup, doctor visit, or new medication.

Quick Start: Claude Code Users

If you use Claude Code, just clone and start talking:

git clone https://github.com/runesleo/ai-health-vault.git
cd ai-health-vault
claude
# "Help me analyze this checkup report" (attach photo)

The .claude/skills/ directory contains 8 pre-built skills that activate automatically — no manual prompt copy-paste needed.

Repository Structure

ai-health-vault/
├── .claude/skills/                 # Claude Code skills (auto-loaded)
│   ├── health-report-extract.md    # Checkup report → structured data
│   ├── medication-recognize.md     # Pill box photo → drug info
│   ├── health-trend-analysis.md    # Multi-checkup trend analysis
│   ├── medical-visit-prep.md       # Pre-visit checklist generator
│   ├── apple-watch-analysis.md     # Apple Watch health data analysis
│   ├── family-friendly-health.md   # Plain language for parents
│   ├── checkup-calendar.md         # Follow-up calendar + alerts
│   └── daily-health-plan.md        # Personalized daily health plan
├── vault/                          # Obsidian Vault templates (ready to use)
│   ├── 健康管理中心.md              # Hub + workflow instructions
│   ├── 家庭成员健康档案.md           # Family overview
│   ├── 就医记录.md                  # Medical visits / surgery / chronic conditions
│   ├── 成员模板/
│   │   └── 体检档案-模板.md         # Per-person checkup archive (incl. Apple Watch)
│   ├── tracking/
│   │   ├── 用药打卡.csv             # Daily medication log
│   │   ├── 饮食记录.csv             # Diet tracking
│   │   ├── 运动记录.csv             # Exercise tracking
│   │   └── 体检指标.csv             # Key checkup metrics over time
│   └── 知识库/
│       └── 常见指标参考.md           # Common health metrics reference
├── prompts/                        # Prompt collection (works with any AI)
│   ├── 01-体检报告提取.md            # Photo → structured data
│   ├── 02-药盒识别.md               # Medication photo → drug list
│   ├── 03-趋势分析.md               # Historical comparison + anomaly detection
│   ├── 04-就医准备.md               # Pre-visit checklist generator
│   ├── 05-Apple-Watch数据分析.md    # Health data export → analysis
│   ├── 06-微信版口语化.md            # Convert to plain language for parents
│   ├── 07-复查日历生成.md            # Follow-up calendar + overdue alerts
│   └── 08-日常管理方案.md            # Diet / exercise / medication / warning signs
├── guides/
│   └── 快速开始.md                  # Detailed setup tutorial
└── LICENSE

Prompts

Each file in prompts/ is a standalone prompt — copy-paste into any AI conversation:

PromptPurposeInput
Checkup Report ExtractionPhoto/PDF → structured tableReport photo
Medication RecognitionPill box photo → drug name + dosage + frequencyMedication photo
Trend AnalysisCompare multiple checkups, flag anomaliesCheckup archive
Pre-Visit PrepGenerate questions for your doctorDepartment + history
Apple Watch AnalysisExport health data → full reportexport.zip
WeChat VersionConvert analysis to plain language for parentsAny analysis result
Follow-up CalendarGenerate follow-up schedule with overdue alertsFamily archives
Daily Health PlanDiet + exercise + medication + warning signsPersonal archive

Who This Is For

  • Tech-savvy people who want to build health archives for their parents
  • Anyone who cares about health data but doesn't want to use third-party apps
  • Apple Watch / smartband owners who want more value from their data

Who This Is NOT For

  • People who don't use AI tools (requires Claude / ChatGPT / Gemini)
  • People who want fully automated solutions (this is a template — you feed data to AI)
  • People who are fine with commercial health apps handling their data

Privacy

All health data is stored locally in Obsidian — nothing is uploaded automatically. However, when you send data to AI for analysis, the content passes through their servers:

  • Claude Code (API mode) — Anthropic does not use API inputs for model training (privacy policy)
  • ChatGPT — Turn off "Improve the model for everyone" in Settings → Data Controls
  • Gemini — Check your data sharing settings in Google AI Studio
  • Maximum privacy — Use a local model (e.g., Ollama, LM Studio) for analysis

Obsidian is your local vault. AI is your analyst. The vault never leaves your machine — but the analysis step involves cloud APIs unless you use a local model.

Contributing

Issues and PRs welcome:

  • New prompts (e.g., specialized extraction for specific test types)
  • Improve existing template structures
  • Add reference ranges for common health metrics
  • Translations

Star History

Star History Chart

Author

Leo (@runes_leo) — AI × Crypto independent builder. Quantitative trading on Polymarket, building data analysis and automation systems with Claude Code. Shipping code, building products, documenting lessons learned. More → leolabs.me

License

MIT

How to Install

  1. Download the template file from GitHub
  2. Move it anywhere in your vault
  3. Open it in Obsidian — done!

Stats

Stars

312

Forks

72

License

MIT

Last updated 25d ago