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

# Output & Async

> How the CLI formats output, its exit codes, cost preflight, and non-blocking polling.

## JSON output

The CLI prints JSON on stdout by default, so it composes with `jq` and shell pipelines.

```bash theme={null}
contenthero model list | jq -r '.[].modelId'
BAL=$(contenthero account balance | jq -r '.credits')
```

## Exit codes

| Code | Meaning                                          |
| ---- | ------------------------------------------------ |
| `0`  | Success                                          |
| `1`  | General error                                    |
| `2`  | Usage error (bad flags or arguments)             |
| `3`  | Auth error (no or invalid credentials)           |
| `4`  | Timeout (work accepted but not finished in time) |

Exit code `4` is not a failure: the generation was accepted and is still rendering. Poll it.

## Cost preflight

Preview any cost for free. The preview takes the same parameters as the real call, so the quote matches the charge.

```bash theme={null}
contenthero generate image --prompt "..." --model nano-banana-2 --cost
```

## Async and polling

`generate` runs a smart-wait of about 50s, then returns the finished asset or an `outputId` if the render is still going. To return immediately and poll yourself:

```bash theme={null}
contenthero generate video --prompt "..." --model kling-3.0 --no-wait   # return an id now
contenthero generation wait <outputId>                                 # block until done
contenthero generation status <outputId>                               # snapshot
```

The `outputId` is always returned, so you can always keep polling. A `status` response includes a `pollAfterSeconds` hint telling you when to check again.