Agent Class — langchain Architecture

Source Code

libs/langchain/langchain_classic/agents/agent.py lines 704–980

class Agent(BaseSingleActionAgent):
    """Agent that calls the language model and deciding the action.

    This is driven by a LLMChain. The prompt in the LLMChain MUST include
    a variable called "agent_scratchpad" where the agent can put its
    intermediary work.
    """

    llm_chain: LLMChain
    """LLMChain to use for agent."""
    output_parser: AgentOutputParser
    """Output parser to use for agent."""
    allowed_tools: list[str] | None = None
    """Allowed tools for the agent. If `None`, all tools are allowed."""

    @override
    def dict(self, **kwargs: Any) -> builtins.dict:
        """Return dictionary representation of agent."""
        _dict = super().dict()
        del _dict["output_parser"]
        return _dict

    def get_allowed_tools(self) -> list[str] | None:
        """Get allowed tools."""
        return self.allowed_tools

    @property
    def return_values(self) -> list[str]:
        """Return values of the agent."""
        return ["output"]

    @property
    def _stop(self) -> list[str]:
        return [
            f"\n{self.observation_prefix.rstrip()}",
            f"\n\t{self.observation_prefix.rstrip()}",
        ]

    def _construct_scratchpad(
        self,
        intermediate_steps: list[tuple[AgentAction, str]],
    ) -> str | list[BaseMessage]:
        """Construct the scratchpad that lets the agent continue its thought process."""
        thoughts = ""
        for action, observation in intermediate_steps:
            thoughts += action.log
            thoughts += f"\n{self.observation_prefix}{observation}\n{self.llm_prefix}"
        return thoughts

    def plan(
        self,
        intermediate_steps: list[tuple[AgentAction, str]],
        callbacks: Callbacks = None,
        **kwargs: Any,
    ) -> AgentAction | AgentFinish:
        """Given input, decided what to do.

        Args:
            intermediate_steps: Steps the LLM has taken to date,
                along with observations.
            callbacks: Callbacks to run.
            **kwargs: User inputs.

        Returns:
            Action specifying what tool to use.
        """
        full_inputs = self.get_full_inputs(intermediate_steps, **kwargs)
        full_output = self.llm_chain.predict(callbacks=callbacks, **full_inputs)
        return self.output_parser.parse(full_output)

    async def aplan(
        self,
        intermediate_steps: list[tuple[AgentAction, str]],
        callbacks: Callbacks = None,
        **kwargs: Any,
    ) -> AgentAction | AgentFinish:
        """Async given input, decided what to do.

        Args:
            intermediate_steps: Steps the LLM has taken to date,
                along with observations.