ChatAgent Class — langchain Architecture

Architecture documentation for the ChatAgent class in base.py from the langchain codebase.

Class python

Entity Profile

Dependency Diagram

graph TD
  b3f40ce7_80ee_f7bd_df0e_7eb84f6efab7["ChatAgent"]
  37fcc3f3_2798_3648_915c_2bffdd19bff7["Agent"]
  b3f40ce7_80ee_f7bd_df0e_7eb84f6efab7 -->|extends| 37fcc3f3_2798_3648_915c_2bffdd19bff7
  356f1882_f802_ae8d_971b_9adf83ef86ea["base.py"]
  b3f40ce7_80ee_f7bd_df0e_7eb84f6efab7 -->|defined in| 356f1882_f802_ae8d_971b_9adf83ef86ea
  84505bd3_5cbc_2907_db15_266c1d471d88["observation_prefix()"]
  b3f40ce7_80ee_f7bd_df0e_7eb84f6efab7 -->|method| 84505bd3_5cbc_2907_db15_266c1d471d88
  61513a51_b63c_7f8d_7770_1d457e8d96a2["llm_prefix()"]
  b3f40ce7_80ee_f7bd_df0e_7eb84f6efab7 -->|method| 61513a51_b63c_7f8d_7770_1d457e8d96a2
  2afe5c6a_955b_c3e0_01d8_639bb3398225["_construct_scratchpad()"]
  b3f40ce7_80ee_f7bd_df0e_7eb84f6efab7 -->|method| 2afe5c6a_955b_c3e0_01d8_639bb3398225
  03335848_7c0b_88ca_e96f_eaa5a72dc146["_get_default_output_parser()"]
  b3f40ce7_80ee_f7bd_df0e_7eb84f6efab7 -->|method| 03335848_7c0b_88ca_e96f_eaa5a72dc146
  0e7041cb_6040_49fd_f4a2_f9c6892bdef1["_validate_tools()"]
  b3f40ce7_80ee_f7bd_df0e_7eb84f6efab7 -->|method| 0e7041cb_6040_49fd_f4a2_f9c6892bdef1
  c944b985_377f_3d2e_121a_6a399f0486f7["_stop()"]
  b3f40ce7_80ee_f7bd_df0e_7eb84f6efab7 -->|method| c944b985_377f_3d2e_121a_6a399f0486f7
  6ddf1017_f7b1_a745_71bb_6fcfb79c937b["create_prompt()"]
  b3f40ce7_80ee_f7bd_df0e_7eb84f6efab7 -->|method| 6ddf1017_f7b1_a745_71bb_6fcfb79c937b
  672a572f_b218_8d99_e7ff_59f1b43194b5["from_llm_and_tools()"]
  b3f40ce7_80ee_f7bd_df0e_7eb84f6efab7 -->|method| 672a572f_b218_8d99_e7ff_59f1b43194b5
  ac228751_ffdd_63a3_3ff6_8ebd49b23e65["_agent_type()"]
  b3f40ce7_80ee_f7bd_df0e_7eb84f6efab7 -->|method| ac228751_ffdd_63a3_3ff6_8ebd49b23e65

Relationship Graph

Source Code

libs/langchain/langchain_classic/agents/chat/base.py lines 36–178

class ChatAgent(Agent):
    """Chat Agent."""

    output_parser: AgentOutputParser = Field(default_factory=ChatOutputParser)
    """Output parser for the agent."""

    @property
    def observation_prefix(self) -> str:
        """Prefix to append the observation with."""
        return "Observation: "

    @property
    def llm_prefix(self) -> str:
        """Prefix to append the llm call with."""
        return "Thought:"

    def _construct_scratchpad(
        self,
        intermediate_steps: list[tuple[AgentAction, str]],
    ) -> str:
        agent_scratchpad = super()._construct_scratchpad(intermediate_steps)
        if not isinstance(agent_scratchpad, str):
            msg = "agent_scratchpad should be of type string."
            raise ValueError(msg)  # noqa: TRY004
        if agent_scratchpad:
            return (
                f"This was your previous work "
                f"(but I haven't seen any of it! I only see what "
                f"you return as final answer):\n{agent_scratchpad}"
            )
        return agent_scratchpad

    @classmethod
    @override
    def _get_default_output_parser(cls, **kwargs: Any) -> AgentOutputParser:
        return ChatOutputParser()

    @classmethod
    def _validate_tools(cls, tools: Sequence[BaseTool]) -> None:
        super()._validate_tools(tools)
        validate_tools_single_input(class_name=cls.__name__, tools=tools)

    @property
    def _stop(self) -> list[str]:
        return ["Observation:"]

    @classmethod
    def create_prompt(
        cls,
        tools: Sequence[BaseTool],
        system_message_prefix: str = SYSTEM_MESSAGE_PREFIX,
        system_message_suffix: str = SYSTEM_MESSAGE_SUFFIX,
        human_message: str = HUMAN_MESSAGE,
        format_instructions: str = FORMAT_INSTRUCTIONS,
        input_variables: list[str] | None = None,
    ) -> BasePromptTemplate:
        """Create a prompt from a list of tools.

        Args:
            tools: A list of tools.
            system_message_prefix: The system message prefix.
            system_message_suffix: The system message suffix.
            human_message: The `HumanMessage`.
            format_instructions: The format instructions.
            input_variables: The input variables.

        Returns:
            A prompt template.
        """
        tool_strings = "\n".join([f"{tool.name}: {tool.description}" for tool in tools])
        tool_names = ", ".join([tool.name for tool in tools])
        format_instructions = format_instructions.format(tool_names=tool_names)
        template = (
            f"{system_message_prefix}\n\n"
            f"{tool_strings}\n\n"
            f"{format_instructions}\n\n"
            f"{system_message_suffix}"
        )
        messages = [
            SystemMessagePromptTemplate.from_template(template),
            HumanMessagePromptTemplate.from_template(human_message),