A private, AI-powered health management system built using Obsidian templates and prompts. It addresses the fragmentation of medical records and privacy concerns associated with commercial health apps by enabling users to create a local, AI-analyzed health archive for themselves and their families. The target audience includes tech-savvy individuals and families prioritizing data privacy and seeking a centralized, intelligent approach to managing health information.

How It Works

This project provides a framework of templates and prompts designed to be used with AI models like Claude, ChatGPT, or Gemini. Users clone the repository and open the vault/ folder within Obsidian. Medical reports, medication photos, or health data exports are then fed to an AI, which uses the provided prompts or auto-loaded Claude skills to extract structured data, identify medications, analyze health trends, and generate personalized health plans. The core advantage is keeping sensitive health data locally within Obsidian, with AI serving as an analysis tool rather than a data repository.

Quick Start & Requirements

• Primary install/run command: Fork the repository or download the ZIP. Open the vault/ folder in Obsidian.
• Non-default prerequisites: Obsidian application, access to an AI model (e.g., Claude, ChatGPT, Gemini). Claude Code users benefit from auto-loaded skills in the .claude/skills/ directory.
• Estimated setup time: Approximately 30 minutes.
• Relevant pages: Repository: https://github.com/runesleo/ai-health-vault.git, Detailed Tutorial: guides/快速开始.md.

Highlighted Details

• Local-first privacy: All health data is stored locally within the Obsidian vault, with no automatic uploading to external services.
• AI-driven insights: Utilizes a comprehensive set of 8 prompts and 8 Claude Code skills for tasks like report extraction, medication recognition, trend analysis, and health plan generation.
• Versatile prompt design: Prompts are standalone and can be copy-pasted into any AI conversation, ensuring broad compatibility beyond specific AI tools.
• Family-oriented features: Includes templates and prompts for creating family member health archives and presenting information in plain language for parents.

Maintenance & Community

The project is maintained by Leo (@runes_leo), an independent builder focused on AI and crypto. Contributions are welcomed via Issues and Pull Requests for new prompts, template improvements, reference data, and translations. No specific community channels (e.g., Discord, Slack) or sponsorship details are listed in the README.

Licensing & Compatibility

The project is licensed under the MIT license. This permissive license allows for commercial use and integration into closed-source projects without significant restrictions.

Limitations & Caveats

This is not a standalone application but a set of templates and guides requiring user interaction with AI tools. Data sent for analysis passes through third-party AI provider servers unless a local AI model is configured and used. Users must have access to and be comfortable using AI services for the system to function.