create_tool_call() — langchain Function Reference

Architecture documentation for the create_tool_call() function in content.py from the langchain codebase.

Function python LangChainCore ApiManagement

Entity Profile

LangChainCore→ ApiManagement→ create_tool_call() — langchain Function Reference

Dependency Diagram

graph TD
  ac6a4301_33e2_e3f2_52ef_1c8780d48bbc["create_tool_call()"]
  c7f11daf_e104_efbe_7225_f1d6da8e8630["content.py"]
  ac6a4301_33e2_e3f2_52ef_1c8780d48bbc -->|defined in| c7f11daf_e104_efbe_7225_f1d6da8e8630
  style ac6a4301_33e2_e3f2_52ef_1c8780d48bbc fill:#6366f1,stroke:#818cf8,color:#fff

Relationship Graph

Source Code

libs/core/langchain_core/messages/content.py lines 1318–1360

def create_tool_call(
    name: str,
    args: dict[str, Any],
    *,
    id: str | None = None,
    index: int | str | None = None,
    **kwargs: Any,
) -> ToolCall:
    """Create a `ToolCall`.

    Args:
        name: The name of the tool to be called.
        args: The arguments to the tool call.
        id: An identifier for the tool call.

            Generated automatically if not provided.
        index: Index of block in aggregate response.

            Used during streaming.

    Returns:
        A properly formatted `ToolCall`.

    !!! note

        The `id` is generated automatically if not provided, using a UUID4 format
        prefixed with `'lc_'` to indicate it is a LangChain-generated ID.
    """
    block = ToolCall(
        type="tool_call",
        name=name,
        args=args,
        id=ensure_id(id),
    )

    if index is not None:
        block["index"] = index

    extras = {k: v for k, v in kwargs.items() if v is not None}
    if extras:
        block["extras"] = extras

    return block