# build_goal_continuation

> **Function** in `deepagents_code`

📖 [View in docs](https://reference.langchain.com/python/deepagents-code/goal_state_notice/build_goal_continuation)

Build a one-time goal continuation.

## Signature

```python
build_goal_continuation(
    transition: GoalTransition,
    *,
    unsaved_objective: str | None = None,
    unsaved_criteria: str | None = None,
    event_id: str | None = None,
) -> HumanMessage
```

## Parameters

| Name | Type | Required | Description |
|------|------|----------|-------------|
| `transition` | `GoalTransition` | Yes | Goal lifecycle transition that should resume work. |
| `unsaved_objective` | `str \| None` | No | Accepted objective supplied directly when creation state could not be persisted. (default: `None`) |
| `unsaved_criteria` | `str \| None` | No | Accepted acceptance criteria supplied alongside `unsaved_objective`. Carried here because the state notice, the model's only channel to the criteria, was never written for this transition, so omitting them leaves the model working toward a goal whose criteria it cannot obtain by any other means. (default: `None`) |
| `event_id` | `str \| None` | No | Optional stable identifier for deterministic tests. (default: `None`) |

## Returns

`HumanMessage`

Internal `HumanMessage` for the next agent turn.

---

[View source on GitHub](https://github.com/langchain-ai/deepagents/blob/e4c1ffe9ee63d7854147de490ca27495ef157a7b/libs/code/deepagents_code/goal_state_notice.py#L262)