BaseTool.run does not pass kwargs to the actual function call

[kapicic]

Checked other resources

• I added a very descriptive title to this issue.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.
• I am sure that this is a bug in LangChain rather than my code.
• The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package).

Example Code

from langchain_core.tools import StructuredTool


def print_input_and_kwargs(input: str, **kwargs) -> None:
    print(f"Input: {input}")
    print(f"Kwargs: {kwargs}")


tool = StructuredTool.from_function(
    func=print_input_and_kwargs,
    name="print_input_and_kwargs",
    description="Print the input and kwargs.",
)

tool.run(
    "Hello, world!", kwarg1="First keyword argument.", kwarg2="Second keyword argument."
)

Actual

# Input: Hello, world!
# Kwargs: {}

Expected

# Input: Hello, world!
# Kwargs: {"kwarg1": "First keyword argument.", "kwarg2": "Second keyword argument."}

Error Message and Stack Trace (if applicable)

No response

Description

I am using langchain to create custom tools.
The tools are using kwargs to pass some data.
BaseTool.run accepts **kwargs as an argument and the docs clearly state:
kwargs: Additional arguments to pass to the tool

I expect that if I pass any kwargs to tool.run they will be available for me to use inside the function, which they are NOT.

System Info

System Information

OS: Darwin
OS Version: Darwin Kernel Version 23.5.0: Wed May 1 20:12:58 PDT 2024; root:xnu-10063.121.3~5/RELEASE_ARM64_T6000
Python Version: 3.12.3 (main, Apr 9 2024, 08:09:14) [Clang 15.0.0 (clang-1500.3.9.4)]

Package Information

langchain_core: 0.2.23
langchain: 0.2.7
langchain_community: 0.2.7
langsmith: 0.1.85
langchain_anthropic: 0.1.19
langchain_google_community: 1.0.6
langchain_google_genai: 1.0.7
langchain_groq: 0.1.6
langchain_mistralai: 0.1.10
langchain_openai: 0.1.17
langchain_text_splitters: 0.2.2
langchainhub: 0.1.20

Packages not installed (Not Necessarily a Problem)

The following packages were not found:

langgraph
langserve

[keenborder786]

I guess the Docstring for BaseTool need a little bit clarification but you can achieve your behaviour using the following code:

def print_input_and_kwargs(input: str, additional_kwargs:dict) -> None:
    print(f"Input: {input}")
    print(f"Kwargs: {additional_kwargs}")


tool = StructuredTool.from_function(
    func=print_input_and_kwargs,
    name="print_input_and_kwargs",
    description="Print the input and kwargs.",
)

tool.run(
    tool_input = {"input":"Hello, world!", "additional_kwargs": {"kwarg1":"First keyword argument.", "kwarg2":"Second keyword argument."}}
)