orchestrate_local

One-shot local execution of a code string in the Monty sandbox.

Warning: Experimental feature: alpha — APIs may change without notice.

    Sends the code + inputs to Monty and returns the result directly,
    without creating a `TaskTemplate` or going through the controller.

    The **last expression** in *source* becomes the return value::

        result = await sandbox.orchestrate_local(
            "add(x, y) * 2",
            inputs=\{"x": 1, "y": 2\},
            tasks=[add],
        )
        # → 6

    Parameters
    ----------
    source:
        Python code string to execute in the sandbox.
    inputs:
        Mapping of input names to their values.
    tasks:
        List of external functions (tasks, durable ops) available inside the
        sandbox. Each item's `__name__` is used as the key.
    timeout_ms:
        Sandbox execution timeout in milliseconds.

async def orchestrate_local(
    source: str,
    *,
    inputs: Dict[str, Any],
    tasks: Optional[List[Any]] = None,
    timeout_ms: int = 30_000,
) -> Any: ...

One-shot local execution of a code string in the Monty sandbox. Sends the code + inputs to Monty and returns the result directly, without creating a TaskTemplate or going through the controller.

Parameters

Name	Type	Description
source	str	Python code string to execute in the sandbox.
inputs	Dict[str, Any]	Mapping of input names to their values.
tasks	Optional[List[Any]] = None	List of external functions (tasks, durable ops) available inside the sandbox. Each item's name is used as the key.
timeout_ms	int = 30000	Sandbox execution timeout in milliseconds.

Returns

Type	Description
Any	The evaluated result of the last expression in the provided source code.