builtin

Base class for server-side built-in tools.

Inherits from :class:dict so instances can be passed directly to model.bind_tools() without additional conversion. Subclasses build the underlying :mod:openai.types.responses TypedDict and pass it to super().__init__(**sdk_typed_dict) so the dict payload always matches the OpenAI SDK schema.

Some tools require extra HTTP headers to be sent with every inference API request. Subclasses that need this should set self._request_headers (a plain :class:dict) in their __init__. The :attr:request_headers property exposes these headers to bind_tools implementations in the model classes, which merge and forward them to the underlying API client.

Example – defining a custom built-in tool::

class MyTool(BuiltinTool):
    def __init__(self, option: str = "default") -> None:
        super().__init__(type="my_tool", option=option)

A tool that runs Python code server-side to help generate a response.

The model can write and execute Python code within a sandboxed container and include the results in its response.

Wraps :class:openai.types.responses.tool_param.CodeInterpreter.

Example::

from langchain.chat_models import init_chat_model
from langchain_azure_ai.tools.builtin import CodeInterpreterTool
from azure.identity import DefaultAzureCredential

credential = DefaultAzureCredential()
model = init_chat_model(model="azure_ai:gpt-4.1", credential=credential)
tool = CodeInterpreterTool()
model_with_code = model.bind_tools([tool])
response = model_with_code.invoke("Plot a sine wave using Python")

A tool that searches for relevant content from uploaded vector stores.

Wraps :class:openai.types.responses.FileSearchToolParam.

Example::

from langchain_azure_ai.tools.builtin import FileSearchTool

tool = FileSearchTool(vector_store_ids=["vs_abc123"])
model_with_search = model.bind_tools([tool])

A tool that generates or edits images using GPT image models.

Wraps :class:openai.types.responses.tool_param.ImageGeneration.

When model_deployment is specified the tool automatically injects an x-ms-oai-image-generation-deployment HTTP header into every inference API call made via :meth:model.bind_tools. This tells the Azure AI Foundry endpoint which model deployment to use for image generation.

Example::

from langchain_azure_ai.tools.builtin import ImageGenerationTool

tool = ImageGenerationTool(
    quality="high",
    size="1024x1024",
    model_deployment="my-gpt-image-1-deployment",
)
model_with_img = model.bind_tools([tool])

A tool that gives the model access to an external MCP server.

Allows the model to call tools exposed by a remote Model Context Protocol (MCP) server within a single conversational turn.

Wraps :class:openai.types.responses.tool_param.Mcp.

Example::

from langchain_azure_ai.tools.builtin import McpTool

tool = McpTool(
    server_label="my_server",
    server_url="https://my-mcp-server.example.com",
)
model_with_mcp = model.bind_tools([tool])

A tool that searches the internet for sources related to the prompt.

Wraps :class:openai.types.responses.WebSearchToolParam.

Example::

from langchain_azure_ai.tools.builtin import WebSearchTool

tool = WebSearchTool(search_context_size="high")
model_with_search = model.bind_tools([tool])