如何将配置传递给工具¶

先决条件

本指南假设您熟悉以下内容

• 可运行接口
• 工具调用代理
• 工具
• 流式传输
• 聊天模型

在运行时，您可能需要将值传递给工具，例如用户 ID，出于安全原因，这应该由应用程序逻辑设置，而不是由 LLM 控制。LLM 应该只管理其预期的参数。

LangChain 工具使用 Runnable 接口，其中 invoke 等方法通过 RunnableConfig 参数接受运行时信息。

在下面的示例中，我们将设置一个带有工具的代理来管理用户的宠物收藏——添加、读取和删除条目，同时通过应用程序逻辑修复用户 ID 并让聊天模型控制其他参数

设置¶

首先，让我们安装所需的软件包并设置我们的 API 密钥

在 [1]

%%capture --no-stderr
%pip install --quiet -U langgraph langchain_anthropic

%%capture --no-stderr %pip install --quiet -U langgraph langchain_anthropic

在 [2]

import getpass
import os


def _set_env(var: str):
    if not os.environ.get(var):
        os.environ[var] = getpass.getpass(f"{var}: ")


_set_env("ANTHROPIC_API_KEY")

import getpass import os def _set_env(var: str): if not os.environ.get(var): os.environ[var] = getpass.getpass(f"{var}: ") _set_env("ANTHROPIC_API_KEY")

为 LangGraph 开发设置 LangSmith

注册 LangSmith 以快速发现问题并提高 LangGraph 项目的性能。LangSmith 允许您使用跟踪数据来调试、测试和监控使用 LangGraph 构建的 LLM 应用程序——阅读有关如何开始的更多信息 此处。

定义工具和模型¶

在 [3]

from typing import List

from langchain_core.tools import tool
from langchain_core.runnables.config import RunnableConfig

from langgraph.prebuilt import ToolNode

user_to_pets = {}


@tool(parse_docstring=True)
def update_favorite_pets(
    # NOTE: config arg does not need to be added to docstring, as we don't want it to be included in the function signature attached to the LLM
    pets: List[str],
    config: RunnableConfig,
) -> None:
    """Add the list of favorite pets.

    Args:
        pets: List of favorite pets to set.
    """
    user_id = config.get("configurable", {}).get("user_id")
    user_to_pets[user_id] = pets


@tool
def delete_favorite_pets(config: RunnableConfig) -> None:
    """Delete the list of favorite pets."""
    user_id = config.get("configurable", {}).get("user_id")
    if user_id in user_to_pets:
        del user_to_pets[user_id]


@tool
def list_favorite_pets(config: RunnableConfig) -> None:
    """List favorite pets if any."""
    user_id = config.get("configurable", {}).get("user_id")
    return ", ".join(user_to_pets.get(user_id, []))

from typing import List from langchain_core.tools import tool from langchain_core.runnables.config import RunnableConfig from langgraph.prebuilt import ToolNode user_to_pets = {} @tool(parse_docstring=True) def update_favorite_pets( # NOTE: config 参数不需要添加到文档字符串中，因为我们不希望它包含在附加到 LLM 的函数签名中 pets: List[str], config: RunnableConfig, ) -> None: """添加宠物列表。参数：pets: 要设置的宠物列表。""" user_id = config.get("configurable", {}).get("user_id") user_to_pets[user_id] = pets @tool def delete_favorite_pets(config: RunnableConfig) -> None: """删除宠物列表。""" user_id = config.get("configurable", {}).get("user_id") if user_id in user_to_pets: del user_to_pets[user_id] @tool def list_favorite_pets(config: RunnableConfig) -> None: """列出宠物（如果有）。""" user_id = config.get("configurable", {}).get("user_id") return ", ".join(user_to_pets.get(user_id, []))

在 [4]

tools = [update_favorite_pets, delete_favorite_pets, list_favorite_pets]
tool_node = ToolNode(tools)

tools = [update_favorite_pets, delete_favorite_pets, list_favorite_pets] tool_node = ToolNode(tools)

在我们的示例中，我们将使用 Anthropic 的小型聊天模型。要将聊天模型与工具调用一起使用，我们需要首先确保模型了解可用的工具。我们通过在 ChatAnthropic 模型上调用 .bind_tools 方法来做到这一点

在 [5]

from langchain_anthropic import ChatAnthropic
from langgraph.graph import StateGraph, MessagesState
from langgraph.prebuilt import ToolNode


model_with_tools = ChatAnthropic(
    model="claude-3-haiku-20240307", temperature=0
).bind_tools(tools)

from langchain_anthropic import ChatAnthropic from langgraph.graph import StateGraph, MessagesState from langgraph.prebuilt import ToolNode model_with_tools = ChatAnthropic( model="claude-3-haiku-20240307", temperature=0 ).bind_tools(tools)

ReAct 代理¶

让我们设置 ReAct 代理 的图实现。此代理将某些查询作为输入，然后重复调用工具，直到拥有足够的信息来解决查询。我们将使用我们刚刚定义的预构建 ToolNode 和带有工具的 Anthropic 模型。

在 [6]

from typing import Literal

from langgraph.graph import StateGraph, MessagesState, START, END


def should_continue(state: MessagesState):
    messages = state["messages"]
    last_message = messages[-1]
    if last_message.tool_calls:
        return "tools"
    return END


def call_model(state: MessagesState):
    messages = state["messages"]
    response = model_with_tools.invoke(messages)
    return {"messages": [response]}


builder = StateGraph(MessagesState)

# Define the two nodes we will cycle between
builder.add_node("agent", call_model)
builder.add_node("tools", tool_node)

builder.add_edge(START, "agent")
builder.add_conditional_edges("agent", should_continue, ["tools", END])
builder.add_edge("tools", "agent")

graph = builder.compile()

from typing import Literal from langgraph.graph import StateGraph, MessagesState, START, END def should_continue(state: MessagesState): messages = state["messages"] last_message = messages[-1] if last_message.tool_calls: return "tools" return END def call_model(state: MessagesState): messages = state["messages"] response = model_with_tools.invoke(messages) return {"messages": [response]} builder = StateGraph(MessagesState) # 定义我们将循环遍历的两个节点 builder.add_node("agent", call_model) builder.add_node("tools", tool_node) builder.add_edge(START, "agent") builder.add_conditional_edges("agent", should_continue, ["tools", END]) builder.add_edge("tools", "agent") graph = builder.compile()

在 [7]

from IPython.display import Image, display

try:
    display(Image(graph.get_graph().draw_mermaid_png()))
except Exception:
    # This requires some extra dependencies and is optional
    pass

from IPython.display import Image, display try: display(Image(graph.get_graph().draw_mermaid_png())) except Exception: # 这需要一些额外的依赖项，并且是可选的 pass

开始使用！¶

在 [8]

from langchain_core.messages import HumanMessage

user_to_pets.clear()  # Clear the state

print(f"User information prior to run: {user_to_pets}")

inputs = {"messages": [HumanMessage(content="my favorite pets are cats and dogs")]}
for chunk in graph.stream(
    inputs, {"configurable": {"user_id": "123"}}, stream_mode="values"
):
    chunk["messages"][-1].pretty_print()

print(f"User information after the run: {user_to_pets}")

from langchain_core.messages import HumanMessage user_to_pets.clear() # 清除状态 print(f"运行前的用户信息：{user_to_pets}") inputs = {"messages": [HumanMessage(content="我最喜欢的宠物是猫和狗")]} for chunk in graph.stream( inputs, {"configurable": {"user_id": "123"}}, stream_mode="values" ): chunk["messages"][-1].pretty_print() print(f"运行后的用户信息：{user_to_pets}")

User information prior to run: {}
================================ Human Message =================================

my favorite pets are cats and dogs
================================== Ai Message ==================================

[{'text': "Okay, let's update your favorite pets:", 'type': 'text'}, {'id': 'toolu_01SU6vhbKDjSsPj2z86QA3wy', 'input': {'pets': ['cats', 'dogs']}, 'name': 'update_favorite_pets', 'type': 'tool_use'}]
Tool Calls:
  update_favorite_pets (toolu_01SU6vhbKDjSsPj2z86QA3wy)
 Call ID: toolu_01SU6vhbKDjSsPj2z86QA3wy
  Args:
    pets: ['cats', 'dogs']
================================= Tool Message =================================
Name: update_favorite_pets

null
================================== Ai Message ==================================

Your favorite pets have been updated to cats and dogs.
User information after the run: {'123': ['cats', 'dogs']}

在 [9]

from langchain_core.messages import HumanMessage

print(f"User information prior to run: {user_to_pets}")

inputs = {"messages": [HumanMessage(content="what are my favorite pets")]}
for chunk in graph.stream(
    inputs, {"configurable": {"user_id": "123"}}, stream_mode="values"
):
    chunk["messages"][-1].pretty_print()

print(f"User information prior to run: {user_to_pets}")

from langchain_core.messages import HumanMessage print(f"运行前的用户信息：{user_to_pets}") inputs = {"messages": [HumanMessage(content="我最喜欢的宠物是什么")]} for chunk in graph.stream( inputs, {"configurable": {"user_id": "123"}}, stream_mode="values" ): chunk["messages"][-1].pretty_print() print(f"运行前的用户信息：{user_to_pets}")

User information prior to run: {'123': ['cats', 'dogs']}
================================ Human Message =================================

what are my favorite pets
================================== Ai Message ==================================

[{'id': 'toolu_01DdpiqiCxzbR4RjQdEoR6mJ', 'input': {}, 'name': 'list_favorite_pets', 'type': 'tool_use'}]
Tool Calls:
  list_favorite_pets (toolu_01DdpiqiCxzbR4RjQdEoR6mJ)
 Call ID: toolu_01DdpiqiCxzbR4RjQdEoR6mJ
  Args:
================================= Tool Message =================================
Name: list_favorite_pets

cats, dogs
================================== Ai Message ==================================

Based on the list_favorite_pets tool, your favorite pets are cats and dogs.
User information prior to run: {'123': ['cats', 'dogs']}

在 [10]

print(f"User information prior to run: {user_to_pets}")

inputs = {
    "messages": [
        HumanMessage(content="please forget what i told you about my favorite animals")
    ]
}
for chunk in graph.stream(
    inputs, {"configurable": {"user_id": "123"}}, stream_mode="values"
):
    chunk["messages"][-1].pretty_print()

print(f"User information prior to run: {user_to_pets}")

print(f"运行前的用户信息：{user_to_pets}") inputs = { "messages": [ HumanMessage(content="请忘记我告诉你的关于我最喜欢的动物的信息") ] } for chunk in graph.stream( inputs, {"configurable": {"user_id": "123"}}, stream_mode="values" ): chunk["messages"][-1].pretty_print() print(f"运行前的用户信息：{user_to_pets}")

User information prior to run: {'123': ['cats', 'dogs']}
================================ Human Message =================================

please forget what i told you about my favorite animals
================================== Ai Message ==================================

[{'id': 'toolu_013TXG6yTxvuWiugbdKGTKSF', 'input': {}, 'name': 'delete_favorite_pets', 'type': 'tool_use'}]
Tool Calls:
  delete_favorite_pets (toolu_013TXG6yTxvuWiugbdKGTKSF)
 Call ID: toolu_013TXG6yTxvuWiugbdKGTKSF
  Args:
================================= Tool Message =================================
Name: delete_favorite_pets

null
================================== Ai Message ==================================

I have deleted the information about your favorite pets. The list of favorite pets has been cleared.
User information prior to run: {}

评论