> ## Documentation Index
> Fetch the complete documentation index at: https://docs.harmstack.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Introduction

> Benchmark medical AI model endpoints with Harmstack using a simple CLI and REST API.

Harmstack helps teams evaluate medical AI quality with repeatable, structured benchmarks. Point Harmstack at your model endpoint, run a benchmark, and review results through the `harmstack` CLI or API.

<CardGroup cols={2}>
  <Card title="Quick Start" icon="bolt" href="/quickstart">
    Install the CLI with Homebrew or a release archive, then run your first benchmark in minutes.
  </Card>

  <Card title="CLI Reference" icon="terminal" href="/cli/overview">
    Explore every command and flag in the `harmstack` CLI.
  </Card>

  <Card title="API Authentication" icon="key" href="/api/authentication">
    Learn how to authenticate API requests with your API key.
  </Card>

  <Card title="API Endpoints" icon="code" href="/api/overview">
    Browse the full public REST API reference.
  </Card>
</CardGroup>

## How it works

<Steps>
  <Step title="Get your API key">
    Obtain your `HARMSTACK_API_KEY` from Vetted Medical. This key authenticates all CLI commands and API requests.
  </Step>

  <Step title="Install the CLI">
    Prebuilt binaries are published on GitHub:

    <a href="https://github.com/vettedmedical/harmstack-install-v0/releases/latest">vettedmedical/harmstack-install-v0 — latest release</a>

    **macOS (Apple Silicon)** and **Linux (amd64):** install with [Homebrew](https://brew.sh/) or download the `.tar.gz` for your platform. **Windows (amd64):** download the `.zip` and use `harmstack.exe`. Full steps are in [Quick Start](/quickstart) and [CLI overview](/cli/overview).
  </Step>

  <Step title="Configure your model endpoint">
    Set your model endpoint and credentials in the wizard, or export defaults so you do not need to re-enter them each time:

    ```bash theme={null}
    export TARGET_MODEL_ENDPOINT_URL="https://api.openai.com/v1/responses"
    export TARGET_MODEL_API_KEY="sk-proj-1234567"
    ```

    Harmstack supports multiple provider shapes, including `openai`, `openai_responses`, `gemini`, and `raw`.
  </Step>

  <Step title="Run and review benchmarks">
    Choose a benchmark, submit a run, and inspect job-level scores and aggregate stats from the CLI or API.
  </Step>
</Steps>

## What you can do

* **Submit benchmarks** - Run one-off or repeated evaluations against your model endpoint
* **Inspect results** - Review scores, metadata, and run details for each job
* **Compare runs** - Measure quality changes across model or prompt revisions
* **Track trends** - Monitor benchmark performance over time
* **Automate in CI/CD** - Run non-interactively with `--consentandskip`

<Note>
  Admin-only API endpoints are intentionally excluded from this documentation. If you need admin access, contact Vetted Medical support.
</Note>

<Tip>
  Start with <a href="/quickstart">Quick Start</a> for installation options and a complete non-interactive command example.
</Tip>