test_tool_search() — langchain Function Reference

Architecture documentation for the test_tool_search() function in test_chat_models.py from the langchain codebase.

Function python LangChainCore LanguageModelBase

Entity Profile

LangChainCore→ LanguageModelBase→ test_tool_search() — langchain Function Reference

Dependency Diagram

graph TD
  3a5f3f74_d5f5_c742_63ae_34d6335862a2["test_tool_search()"]
  f27640dd_3870_5548_d153_f9504ae1021f["test_chat_models.py"]
  3a5f3f74_d5f5_c742_63ae_34d6335862a2 -->|defined in| f27640dd_3870_5548_d153_f9504ae1021f
  style 3a5f3f74_d5f5_c742_63ae_34d6335862a2 fill:#6366f1,stroke:#818cf8,color:#fff

Relationship Graph

Source Code

libs/partners/anthropic/tests/integration_tests/test_chat_models.py lines 2081–2147

def test_tool_search(output_version: str) -> None:
    """Test tool search with LangChain tools using extras parameter."""

    @tool(parse_docstring=True, extras={"defer_loading": True})
    def get_weather(location: str, unit: str = "fahrenheit") -> str:
        """Get the current weather for a location.

        Args:
            location: City name
            unit: Temperature unit (celsius or fahrenheit)
        """
        return f"The weather in {location} is sunny and 72°{unit[0].upper()}"

    @tool(parse_docstring=True, extras={"defer_loading": True})
    def search_files(query: str) -> str:
        """Search through files in the workspace.

        Args:
            query: Search query
        """
        return f"Found 3 files matching '{query}'"

    model = ChatAnthropic(
        model="claude-opus-4-5-20251101", output_version=output_version
    )

    agent = create_agent(  # type: ignore[var-annotated]
        model,
        tools=[
            {
                "type": "tool_search_tool_regex_20251119",
                "name": "tool_search_tool_regex",
            },
            get_weather,
            search_files,
        ],
    )

    # Test with actual API call
    input_message = {
        "role": "user",
        "content": "What's the weather in San Francisco? Find and use a tool.",
    }
    result = agent.invoke({"messages": [input_message]})
    first_response = result["messages"][1]
    content_types = [block["type"] for block in first_response.content]
    if output_version == "v0":
        assert content_types == [
            "text",
            "server_tool_use",
            "tool_search_tool_result",
            "text",
            "tool_use",
        ]
    else:
        # v1
        assert content_types == [
            "text",
            "server_tool_call",
            "server_tool_result",
            "text",
            "tool_call",
        ]

    answer = result["messages"][-1]
    assert not answer.tool_calls
    assert answer.text