简单Demo

这里们传入的是 model，而不是 model_with_tools。这是因为 create_react_agent 会在后台为我们调用 .bind_tools。

MemorySaver：工作流状态管理工具，基于内存，保存完整工作流状态。

{ "ts": "2023-01-01T12:00:00Z"

, "step": 5

, // 当前执行步骤 "data": { "messages": [...], // 对话消息

"tool_results": {...}, // 工具执行结果

"decision_path": [...] // 分支决策路径 } }
ChatMessageHistory (LangChain)：只是个简单存储的消息列表

[
  HumanMessage(content="Hello"),
  AIMessage(content="Hi there!"),
  HumanMessage(content="How's weather?")
]

# Import relevant functionality
from langchain_anthropic import ChatAnthropic
from langchain_community.tools.tavily_search import TavilySearchResults
from langchain_core.messages import HumanMessage
from langgraph.checkpoint.memory import MemorySaver
from langgraph.prebuilt import create_react_agent

# Create the agent
memory = MemorySaver()
model = ChatAnthropic(model_name="claude-3-sonnet-20240229")
search = TavilySearchResults(max_results=2)
tools = [search]
agent_executor = create_react_agent(model, tools, checkpointer=memory)

# Use the agent
config = {"configurable": {"thread_id": "abc123"}}
for chunk in agent_executor.stream(
    {"messages": [HumanMessage(content="hi im bob! and i live in sf")]}, config
):
    print(chunk)
    print("----")

for chunk in agent_executor.stream(
    {"messages": [HumanMessage(content="whats the weather where I live?")]}, config
):
    print(chunk)
    print("----")

流式Token

event["event"] 的值是​​由框架自动判别和填充​​的

# 可能捕获的所有事件类型（部分常见）
"on_llm_start"      # 语言模型调用开始
"on_llm_end"        # 语言模型调用结束
"on_chain_start"    # 链开始执行
"on_chain_end"      # 链结束执行
"on_tool_start"     # 工具调用开始
"on_tool_end"       # 工具调用结束
"on_chat_model_stream"  # 聊天模型流式输出
"on_retriever_start"   # 检索器开始工作
"on_retriever_end"     # 检索器结束工作

自动判别机制​​：

• 当Agent开始执行一个链（chain）时，会触发on_chain_start事件
• 当模型开始生成响应时，会触发on_chat_model_stream事件
• 当工具开始执行时，会触发on_tool_start事件
• 当对应操作完成时，会触发相应的结束事件（如on_chain_end，on_tool_end）
• 这些事件类型字符串由LangChain内部机制决定

async for event in agent_executor.astream_events(
    {"messages": [HumanMessage(content="whats the weather in sf?")]}, version="v1"
):
    kind = event["event"]
    if kind == "on_chain_start":
        if (
            event["name"] == "Agent"
        ):  # Was assigned when creating the agent with `.with_config({"run_name": "Agent"})`
            print(
                f"Starting agent: {event['name']} with input: {event['data'].get('input')}"
            )
    elif kind == "on_chain_end":
        if (
            event["name"] == "Agent"
        ):  # Was assigned when creating the agent with `.with_config({"run_name": "Agent"})`
            print()
            print("--")
            print(
                f"Done agent: {event['name']} with output: {event['data'].get('output')['output']}"
            )
    if kind == "on_chat_model_stream":
        content = event["data"]["chunk"].content
        if content:
            # Empty content in the context of OpenAI means
            # that the model is asking for a tool to be invoked.
            # So we only print non-empty content
            print(content, end="|")
    elif kind == "on_tool_start":
        print("--")
        print(
            f"Starting tool: {event['name']} with inputs: {event['data'].get('input')}"
        )
    elif kind == "on_tool_end":
        print(f"Done tool: {event['name']}")
        print(f"Tool output was: {event['data'].get('output')}")
        print("--")