# shelbi orchestrate

Make sure the project's orchestrator session is running, starting it if needed.

```text
shelbi orchestrate [OPTIONS]
```

`shelbi orchestrate` makes sure the current project's
[orchestrator](/docs/concepts/orchestrator) is running as the project's
orchestrator session. It starts the hub daemon on demand if it isn't up yet,
then starts the orchestrator agent if it isn't already running. It prints
`✓ orchestrator started (…)` or `orchestrator already running (…)`, followed by
a reminder to open the project with `shelbi` to see it.

You rarely need it: opening a project with `shelbi` starts the orchestrator,
and the daemon relaunches it if it dies while the project is open. It is
useful from a script or a plain shell when you want the orchestrator up
without opening a client.

## Flags

| Flag | Type | Default | Description |
| --- | --- | --- | --- |
| `--status` | flag | off | Reserved. It currently behaves exactly like a plain `shelbi orchestrate`: it still starts the orchestrator if it isn't running, and prints the same lines. |
| `--root <PATH>` | string | install-time default / `$SHELBI_ROOT` / `~/.shelbi` | Override the Shelbi root directory. The flag wins over both the env var and the compile-time default. |
| `-p, --project <PROJECT>` | string | env / cwd lookup | Project whose orchestrator to start. |

## Examples

Start the orchestrator for the project in the current directory:

```bash
shelbi orchestrate
```

## See also

- [Orchestrator](/docs/concepts/orchestrator): what the orchestrator does and
  how it is supervised.
- [`shelbi reload`](/docs/cli/reload): relaunch the orchestrator (and the rest
  of the project) on a freshly installed binary.
- [`shelbi orchestrator events`](/docs/cli/orchestrator): the event queue the
  orchestrator drains.