guardrails_ai.sdk
=================

.. py:module:: guardrails_ai.sdk


Submodules
----------

.. toctree::
   :maxdepth: 1

   /autoapi/guardrails_ai/sdk/abstract_client/index
   /autoapi/guardrails_ai/sdk/chat_completions_api/index
   /autoapi/guardrails_ai/sdk/guardrails_ai_client/index
   /autoapi/guardrails_ai/sdk/guards_api/index
   /autoapi/guardrails_ai/sdk/methods/index


Classes
-------

.. autoapisummary::

   guardrails_ai.sdk.GuardrailsAI


Package Contents
----------------

.. py:class:: GuardrailsAI(*, api_key: str, base_url: str = 'http://localhost:8000', timeout: float | None = None, max_retries: int = 5, headers: Mapping[str, str] | None = None, http_client: httpx.AsyncClient | None = None)

   Bases: :py:obj:`guardrails_ai.sdk.abstract_client.Client`


   Main Guardrails AI SDK client with namespaced API access.

   Example::

       client = GuardrailsAI(api_key="your-api-key")
       guard = await client.guards.retrieve(name="my-guard")
       result = await client.guards.validate("my-guard", "some text")

   Initialize the Guardrails AI client.

   :param api_key: Guardrails AI API key for authentication.
   :param base_url: Base URL for the Guardrails API. Defaults to ``http://localhost:8000``.
   :param timeout: HTTP request timeout in seconds. Defaults to None (no timeout).
   :param max_retries: Maximum number of retry attempts for failed requests. Defaults to 5.
   :param headers: Additional HTTP headers to include with every request.
   :param http_client: A pre-configured ``httpx.AsyncClient`` to use instead of the
                       default. If provided, the required auth headers will be merged into its
                       headers.


   .. py:attribute:: guards
      :type:  guardrails_ai.sdk.guards_api.GuardsApi


   .. py:attribute:: headers
      :type:  dict[str, str]


   .. py:attribute:: http_client
      :type:  httpx.AsyncClient


   .. py:attribute:: max_retries
      :type:  int