AI Agent开发学习系列 - LangGraph(6): 有多个节点的Sequential Graph(练习解答)

在AI Agent开发学习系列 - LangGraph(5): 有多个节点的Sequential Graph中，我们学习了如何创建有多个节点的Sequential Graph。为了巩固学习，我们来做一个练习。

用LangGraph创建如下图的一个Agent:

要求：

1. 输入你的名字
2. 输入你的年龄
3. 输入你的技能列表，如[Java, Python, JS]
4. 在第一个节点中，返回结果<名字> welcmoe to the system!
5. 在第二个节点中，返回第一个节点的结果加上You are <年龄> years old!
6. 第三个节点中中，返回第二个节点的结果加上You have skills in <技能>, <技能> and <技能>.
7. 最终输出图和最终消息结果。

解答：

from typing import TypedDict
from langgraph.graph import StateGraphclass AgentState(TypedDict):name: strage: strskills: listmessage: strdef first_note(state: AgentState) -> AgentState:"""This is the first node of our sequence"""state["message"] = f"{state["name"]}, welcmoe to the system! "return statedef second_node(state: AgentState) -> AgentState:"""This is the second node of our sequence"""state["message"] = state["message"] + f"You are {state["age"]} years old! "return statedef third_node(state: AgentState) -> AgentState:"""This is the third node of our sequence"""state["message"] = state["message"] + f"You have skills in {", ".join(state["skills"][0:-1])} and {state["skills"][-1]}."return stategraph = StateGraph(AgentState)graph.add_node("first_note", first_note)
graph.add_node("second_node", second_node)
graph.add_node("third_node", third_node)graph.set_entry_point("first_note")
graph.add_edge("first_note", "second_node")
graph.add_edge("second_node", "third_node")
graph.set_finish_point("third_node")app = graph.compile()from IPython.display import Image, display
display(Image(app.get_graph().draw_mermaid_png()))result = app.invoke({"name": "Alex", "age": "41", "skills": ["Python", "AI Agent", "LangGraph"]})
print(result["message"])

输出结果：

Alex, welcmoe to the system! You are 41 years old! You have skills in Python, AI Agent and LangGraph.