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

# ralph

> Keep an agent continuously working on a prompt.

`ralph` is an archived example, not installed by `init`. Copy it from
`examples/init-pack/` with its dependency closure, or ask `create-workflow`
for a bounded variant. Once installed, it repeatedly runs the configured
smart agent on your prompt.

```bash theme={"theme":{"light":"github-light","dark":"github-dark"}}
bunx smthrs workflow run ralph --prompt "Continue reducing flaky tests"
```

<Frame caption="The ralph UI shows the single agent task re-running on your prompt in an unbounded loop.">
  <img src="https://mintcdn.com/smithers/py7ertgBTLCgQQDl/images/workflow-ui/ralph.png?fit=max&auto=format&n=py7ertgBTLCgQQDl&q=85&s=67726b975e5a51e221817427df97221f" alt="ralph workflow UI showing a single agent task in an unbounded loop" width="1280" height="832" data-path="images/workflow-ui/ralph.png" />
</Frame>

## Inputs

| Input    | Type   | Default                                   |
| -------- | ------ | ----------------------------------------- |
| `prompt` | string | `"Continue working on the current task."` |

## Use it when

* An open-ended prompt with no fixed finish line benefits from continuous iteration (reducing flaky tests, chipping at tech debt).

## Use something else when

* The work has a clear definition of done: use [`implement`](/workflows/implement) or [`mission`](/workflows/mission).
* Need review gates or validation each pass: use a validation-loop workflow such as [`implement`](/workflows/implement).

## Resume & retry

`ralph` won't stop on its own: cancel it with `bunx smthrs cancel RUN_ID` when enough work is done, or configure a `maxIterations` limit in your workflow file. Runs are durable and resume with `up --resume`.