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

# quanuxctl CLI reference: all subcommands and usage

> quanuxctl is the primary CLI for managing QuanuX nodes, strategies, and infrastructure. Install it once and operate your entire cluster from one terminal.

`quanuxctl` is the single entrypoint for operating the entire QuanuX ecosystem. You use it to provision bare-metal servers into execution nodes, deploy and compile the C++ engine, run backtests, manage live risk parameters, and control observability and infrastructure — all without SSH-ing into individual machines. It ships as part of the standard QuanuX installation.

## Installation

`quanuxctl` is included with every QuanuX installation. If you installed QuanuX via Conda, the binary is already available in your environment:

```bash theme={null}
conda install quanux
quanuxctl --help
```

For a source installation, `quanuxctl` is available after completing the standard setup:

```bash theme={null}
git clone https://github.com/quantdiy/QuanuX.git
cd QuanuX
pip install -r requirements.txt
```

## Subcommands

| Subcommand | Purpose                                                                             |
| ---------- | ----------------------------------------------------------------------------------- |
| `habitat`  | Prepare OS dependencies and baseline infrastructure on a bare server                |
| `nest`     | Deploy and natively compile the C++ execution engine on a conditioned node          |
| `crucible` | Start, stop, monitor, and report on backtesting jobs                                |
| `engine`   | Tune the OS/kernel, initialize NATS JetStream, and start the Annex decoder services |
| `risk`     | View and update global risk state, notional caps, and force node re-hydration       |
| `obs`      | Configure exchange settlement epochs and trigger observability plane operations     |
| `cluster`  | Manage Raft elections, leader promotion, and STONITH fencing for HA clusters        |
| `infra`    | Provision and destroy zero-trust VPC infrastructure                                 |
| `query`    | Validate and dry-run DuckDB SQL queries against analytical targets                  |
| `secrets`  | Store and retrieve API keys and credentials from the OS keyring                     |

## Deployment order

`habitat` and `nest` have a strict dependency: you must run `habitat equip` before `nest drop`. Together they form the Two-Stage Immutable Deployment pattern — habitat conditions the server, nest compiles and installs the engine on the conditioned node.

<Note>
  Never use Docker or cross-compilation for execution node deployments. The C++ engine must be compiled natively on the target hardware with `-O3 -march=native` to achieve deterministic 59ns latency. `quanuxctl nest` enforces this automatically.
</Note>

## Explore the subcommands

<CardGroup cols={2}>
  <Card title="habitat" icon="server" href="/cli/habitat">
    Stage 1 deployment: condition a bare server into a QuanuX execution node.
  </Card>

  <Card title="nest" icon="cpu" href="/cli/nest">
    Stage 2 deployment: compile and install the C++ engine as a systemd service.
  </Card>

  <Card title="crucible" icon="flask-conical" href="/cli/crucible">
    Orchestrate backtesting jobs and pull L3 execution metrics.
  </Card>

  <Card title="engine" icon="bolt" href="/cli/engine">
    Tune the OS, start NATS JetStream, and control the live execution services.
  </Card>

  <Card title="risk" icon="shield-check" href="/cli/risk">
    View global notional exposure and update risk caps in real time.
  </Card>
</CardGroup>