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

# Auto, Plan, and effort

> Choose the agent's intent, sprint scope, and depth of work.

Three controls answer different questions: **Auto or Plan** sets the intent of your conversation, **sprints** define a unit of work, and **effort** controls how deeply a sprint approaches it. [Model selection](/coding-agent/building/pro-mode) chooses the models used for that work.

## Auto and Plan

| Mode | Behavior                                                                 | Command |
| ---- | ------------------------------------------------------------------------ | ------- |
| Auto | The agent can answer, ask for context, implement directly, or plan first | `/auto` |
| Plan | Messages shape requirements, user flows, roadmap, and architecture       | `/plan` |

Auto is the default. Plan mode returns to Auto when the plan is produced. Selecting Plan does not itself start a build: send the requirements you want to develop, review the artifacts, then request implementation.

## Choose a unit of work

For a small change, describe it in chat. Use a dedicated sprint for a defined feature:

```text theme={null}
/sprint add CSV export to the reports page with a progress indicator
```

`/fork <prompt>` starts another task in a separate session. In projects with a roadmap, **Build Next Task** works on the next unit of planned work and **Build on Autopilot** continues through the roadmap. Availability depends on the project's plan and active work.

Follow the session that owns the run. Parallel work, queuing, and admission limits depend on the client and account; submitting another message is not a guarantee that it starts immediately.

## Sprint effort

Set effort with `/effort` before launching the sprint.

| Setting  | Approach                                             |
| -------- | ---------------------------------------------------- |
| `auto`   | Choose an approach for the task                      |
| `low`    | A direct implementation loop for focused changes     |
| `medium` | A direct loop organized around a task list           |
| `high`   | A separate research, coding, and acceptance workflow |

Higher effort can involve more model work and credits. Effort controls the workflow; it does not promise a particular duration or result. A direct chat change does not necessarily run the full high-effort pipeline.

## Review the outcome

State the checks you need in the request: for example, type checking, a specific test suite, or a browser journey. Read the actual [verification evidence](/coding-agent/building/acceptance-criteria) and inspect the diff before shipping.