# LLM Health

> Your private, own-risk health concierge for labs, wearables, records, timelines, reviews, diagnostic gaps, test candidates, category consults, and research queues. Alias-only local HUB, own-risk agreement gate, and the repackaged health-v2 analytics/dashboard stack. Claude Code, OpenAI Codex, OpenCode, Pi, and portable LLM agents are supported interfaces, not the product story.

Author: nvk · Source: https://github.com/nvk/llm-health · License: MIT

## Critical disclaimer

LLM Health is experimental software, not medical advice, not a diagnosis, not a prescription, not an order, not an emergency service and not a clinician relationship. Users are responsible for all decisions and omissions. Every local HUB requires explicit own-risk acceptance with `health agreement show` and `health agreement accept --own-risk` or `health config hub-path ~/health --init --accept-risk`.

## Install

```sh
brew tap nvk/tap
brew install llm-health
health agreement show
health config hub-path ~/health --init --accept-risk
health doctor
health welcome
```

## Interfaces

- Terminal: `health` and `health-v2` are the deterministic core.
- Dashboard: `health-v2` timelines, charts, categories, and exports.
- Chat: Claude Code `/health`, Codex `@health`, OpenCode/Pi instruction files.
- Portable agents: `AGENTS.health.md`.

## Common CLI

`health result`, `health review`, `health close-gaps`, `health test-battery`, `health consult`, `health data-wishlist`, `health dr-visit`, `health plugin-paths`, `health-v2`.