# Quick Start

Get from zero to a running agent in under a minute.

## 1. Create a Project Directory

```bash
mkdir my-squadron && cd my-squadron
```

## 2. Launch Squadron

```bash
squadron engage
```

On a fresh directory, `engage` runs an interactive wizard that:

- Asks which LLM provider you want (Anthropic, OpenAI, or Gemini)
- Prompts for your API key and stores it in an encrypted vault
- Optionally generates a starter mission (`hn_research`) that fetches the Hacker News front page, researches the top 3 stories in parallel, and produces an executive summary

When the wizard finishes, Squadron runs in the background and opens the command center UI in your browser at `http://localhost:8080`. Squadron also installs a system service (launchd on macOS, systemd on Linux) so it starts automatically on boot.

## 3. Try It Out

From the command center UI you can:

- Run missions with input forms and watch execution live
- Edit HCL config files in the browser (with validation)
- Manage variables
- Browse mission run history with task-level detail

Or run missions from the terminal:

```bash
squadron mission hn_research
```

## 4. Stopping

Run from the same directory you engaged from:

```bash
squadron disengage
```

This stops the daemon and removes the system service so it no longer starts on boot.

## What's Next

- [engage](/cli/engage) — How the daemon works, all flags, remote command center setup
- [Agents](/config/agents) — Configure agent behavior
- [Tools](/config/tools) — Add custom tools
- [Missions](/missions/overview) — Build multi-step pipelines
- [Docker](/getting-started/docker) — Run Squadron in containers