1. The tool abstraction in LangChain associates a Python function with a schema that defines the function's name, description and expected arguments.
2. Tools can be passed to chat models that support tool calling allowing the model to request the execution of a specific function with specific inputs.

Key concepts​

• Tools are a way to encapsulate a function and its schema in a way that can be passed to a chat model.
• Create tools using the @tool decorator, which simplifies the process of tool creation, supporting the following:
  • Automatically infer the tool's name, description and expected arguments, while also supporting customization.
  • Defining tools that return artifacts (e.g. images, dataframes, etc.)
  • Hiding input arguments from the schema (and hence from the model) using injected tool arguments.

Tool interface

The key attributes that correspond to the tool's schema:

• name: The name of the tool.
• description: A description of what the tool does.
• args: Property that returns the JSON schema for the tool's arguments.

The key methods to execute the function associated with the tool:

• invoke: Invokes the tool with the given arguments.
• ainvoke: Invokes the tool with the given arguments, asynchronously.

Create tools using the @tool decorator

1. Way to create tools is using the @tool decorator.

   from langchain_core.tools import tool
   
   @tool
   def multiply(a: int, b: int) -> int:
      """Multiply two numbers."""
      return a * b
   

2. Once you have defined a tool, you can use it directly by calling the function.

   multiply.invoke({"a": 2, "b": 3})
   

3. You can also inspect the tool's schema and other properties:

   print(multiply.name) # multiply
   print(multiply.description) # Multiply two numbers.
   print(multiply.args) 
   # {
   # 'type': 'object', 
   # 'properties': {'a': {'type': 'integer'}, 'b': {'type': 'integer'}}, 
   # 'required': ['a', 'b']
   # }
   

Configuring the schema