langgraph 多智能体 Multi-agent supervisor

1. 工具定义

1.1网络搜索工具

from typing import Annotated
import os
from langchain_community.tools.tavily_search import TavilySearchResults
from langchain_core.tools import tool
from langchain_experimental.utilities import PythonREPL

os.environ["TAVILY_API_KEY"] = "your api key"
tavily_tool = TavilySearchResults(max_results=3)

1.2 代码执行工具

from autogen_core.base import CancellationToken
from autogen_core.components.tools import PythonCodeExecutionTool
from autogen_ext.code_executors import DockerCommandLineCodeExecutor

# Create the tool.
code_executor = DockerCommandLineCodeExecutor()
await code_executor.start()
code_execution_tool = PythonCodeExecutionTool(code_executor)
cancellation_token = CancellationToken()


from langgraph.prebuilt import ToolNode


@tool
async def code_executor(
    code: Annotated[str, "The python code to execute."],                        
):
    """Use this to execute python code. If you want to see the output of a value,
    you should print it out with `print(...)`. This is visible to the user."""
    
    try:
        result = await code_execution_tool.run_json({"code": code}, cancellation_token)
    except BaseException as e:
        return f"Failed to execute. Error: {repr(e)}"
    
    
    result= code_execution_tool.return_value_as_string(result)
    
    result_str = f"The result is:{result}"
    return result_str

2. graph中各节点定义

from langgraph.graph import MessagesState

# The agent state is the input to each node in the graph
class AgentState(MessagesState):
    # The 'next' field indicates where to route to next
    next: str

from typing import Literal
from typing_extensions import TypedDict

from langchain_openai import ChatOpenAI

### Edges
llm = ChatOpenAI(
    temperature=0,
    model="GLM-4-plus",
    openai_api_key="your api key",
    openai_api_base="https://open.bigmodel.cn/api/paas/v4/"
)


members = ["researcher", "coder"]
# Our team supervisor is an LLM node. It just picks the next agent to process
# and decides when the work is completed
options = members + ["FINISH"]

system_prompt = (
    "You are a supervisor tasked with managing a conversation between the"
    f" following workers: {members}. Given the following user request,"
    " respond with the worker to act next. Each worker will perform a"
    " task and respond with their results and status. When finished,"
    " respond with FINISH."
)

from langchain_core.messages import HumanMessage
from langgraph.graph import StateGraph, START, END
from langgraph.prebuilt import create_react_agent


class Router(TypedDict):
    """Worker to route to next. If no workers needed, response 'FINISH'."""

    next: Literal[*options]



async def supervisor_node(state: AgentState) -> AgentState:
    messages = [
        {"role": "system", "content": system_prompt},
    ] + state["messages"]
    response = await llm.with_structured_output(Router).ainvoke(messages)
    next_ = response["next"]
    if "FINISH" in next_:
        next_ = END

    return {"next": next_}

research_agent = create_react_agent(
    llm, tools=[tavily_tool], state_modifier="You only provide information. Don't do the math."
)


async def research_node(state: AgentState) -> AgentState:
    result = await research_agent.ainvoke(state)
    return {
        "messages": [
            HumanMessage(content=result["messages"][-1].content, name="researcher")
        ]
    }


# NOTE: THIS PERFORMS ARBITRARY CODE EXECUTION, WHICH CAN BE UNSAFE WHEN NOT SANDBOXED
code_agent = create_react_agent(llm, tools=[code_executor],
                state_modifier="""
                when you use the code tool,it only detect the code block in markdown format.
                """             
            )


async def code_node(state: AgentState) -> AgentState:
    result = await code_agent.ainvoke(state)
    return {
        "messages": [HumanMessage(content=result["messages"][-1].content, name="coder")]
    }


builder = StateGraph(AgentState)
builder.add_edge(START, "supervisor")
builder.add_node("supervisor", supervisor_node)
builder.add_node("researcher", research_node)
builder.add_node("coder", code_node)

<langgraph.graph.state.StateGraph at 0x23c52d56c60>

3. graph中的边定义

for member in members:
    # We want our workers to ALWAYS "report back" to the supervisor when done
    builder.add_edge(member, "supervisor")

# The supervisor populates the "next" field in the graph state
# which routes to a node or finishes
builder.add_conditional_edges("supervisor", lambda state: state["next"])
# Finally, add entrypoint
builder.add_edge(START, "supervisor")

graph = builder.compile()

from IPython.display import display, Image
display(Image(graph.get_graph().draw_mermaid_png()))

请添加图片描述

4. 示例

4.1 第一个例子

问题： What's the square root of 42?

async for s in graph.astream(
    {"messages": [("user", "What's the square root of 42?")]}, subgraphs=True
):
    print(s)
    print("----")

((), {'supervisor': {'next': 'coder'}})
----
(('coder:7c74ccb6-c024-13da-c40d-83d4f5bb3251',), {'agent': {'messages': [AIMessage(content='', additional_kwargs={'tool_calls': [{'id': 'call_-9181532423403291248', 'function': {'arguments': '{"code": "import math\\nprint(math.sqrt(42))"}', 'name': 'code_executor'}, 'type': 'function', 'index': 0}], 'refusal': None}, response_metadata={'token_usage': {'completion_tokens': 18, 'prompt_tokens': 190, 'total_tokens': 208, 'completion_tokens_details': None, 'prompt_tokens_details': None}, 'model_name': 'GLM-4-plus', 'system_fingerprint': None, 'finish_reason': 'tool_calls', 'logprobs': None}, id='run-d809acc4-d589-4227-ae66-96a4908305bc-0', tool_calls=[{'name': 'code_executor', 'args': {'code': 'import math\nprint(math.sqrt(42))'}, 'id': 'call_-9181532423403291248', 'type': 'tool_call'}], usage_metadata={'input_tokens': 190, 'output_tokens': 18, 'total_tokens': 208, 'input_token_details': {}, 'output_token_details': {}})]}})
----
(('coder:7c74ccb6-c024-13da-c40d-83d4f5bb3251',), {'tools': {'messages': [ToolMessage(content='The result is:6.48074069840786\n', name='code_executor', id='9127536d-111b-4de9-9ecf-769a369e619d', tool_call_id='call_-9181532423403291248')]}})
----
(('coder:7c74ccb6-c024-13da-c40d-83d4f5bb3251',), {'agent': {'messages': [AIMessage(content='The square root of 42 is approximately 6.4807.', additional_kwargs={'refusal': None}, response_metadata={'token_usage': {'completion_tokens': 16, 'prompt_tokens': 225, 'total_tokens': 241, 'completion_tokens_details': None, 'prompt_tokens_details': None}, 'model_name': 'GLM-4-plus', 'system_fingerprint': None, 'finish_reason': 'stop', 'logprobs': None}, id='run-b7554cd1-c355-4ad4-8482-5a27ad2afa9a-0', usage_metadata={'input_tokens': 225, 'output_tokens': 16, 'total_tokens': 241, 'input_token_details': {}, 'output_token_details': {}})]}})
----
((), {'coder': {'messages': [HumanMessage(content='The square root of 42 is approximately 6.4807.', additional_kwargs={}, response_metadata={}, name='coder')]}})
----
((), {'supervisor': {'next': '__end__'}})
----

4.2 第二个例子

问题：Find the latest GDP of New York and California, then calculate the average

async for s in graph.astream(
    {
        "messages": [
            (
                "user",
                "Find the latest GDP of New York and California, then calculate the average",
            )
        ]
    },
    subgraphs=True,
):
    print(s)
    print("----")

((), {'supervisor': {'next': 'researcher'}})
----
(('researcher:1c6c3cb9-b2d8-3632-e480-a307b53f1d77',), {'agent': {'messages': [AIMessage(content='', additional_kwargs={'tool_calls': [{'id': 'call_-9181536546572673495', 'function': {'arguments': '{"query": "latest GDP of New York 2023"}', 'name': 'tavily_search_results_json'}, 'type': 'function', 'index': 0}, {'id': 'call_-9181536546572673494', 'function': {'arguments': '{"query": "latest GDP of California 2023"}', 'name': 'tavily_search_results_json'}, 'type': 'function', 'index': 1}], 'refusal': None}, response_metadata={'token_usage': {'completion_tokens': 41, 'prompt_tokens': 170, 'total_tokens': 211, 'completion_tokens_details': None, 'prompt_tokens_details': None}, 'model_name': 'GLM-4-plus', 'system_fingerprint': None, 'finish_reason': 'tool_calls', 'logprobs': None}, id='run-7553ea17-b477-45ff-973b-15c3584bfca9-0', tool_calls=[{'name': 'tavily_search_results_json', 'args': {'query': 'latest GDP of New York 2023'}, 'id': 'call_-9181536546572673495', 'type': 'tool_call'}, {'name': 'tavily_search_results_json', 'args': {'query': 'latest GDP of California 2023'}, 'id': 'call_-9181536546572673494', 'type': 'tool_call'}], usage_metadata={'input_tokens': 170, 'output_tokens': 41, 'total_tokens': 211, 'input_token_details': {}, 'output_token_details': {}})]}})
----
(('researcher:1c6c3cb9-b2d8-3632-e480-a307b53f1d77',), {'tools': {'messages': [ToolMessage(content='[{"url": "https://countryeconomy.com/gdp/usa-states/new-york", "content": "New York - GDP at market prices 2023 | countryeconomy.com US States GDP at market prices New York - GDP at market prices | 2023 | €2,008,702M | $2,172,010M | 1.5% | | 2022 | €1,949,439M | $2,052,759M | 1.7% | | 2021 | €1,626,290M | $1,923,413M | 4.8% | | 2020 | €1,552,591M | $1,773,370M | -2.8% | | 2019 | €1,596,669M | $1,787,471M | 2.4% | Evolution: GDP growth rate at constant prices New York Argentina: Fitch:Long Term Foreign currency Sovereign rating Barbados: Fitch:Long Term Foreign currency Sovereign rating Croatia: Fitch:Long Term Foreign currency Sovereign rating Egypt: Fitch:Long Term Foreign currency Sovereign rating Mongolia: Fitch:Long Term Foreign currency Sovereign rating Tunisia: Fitch:Long Term Foreign currency Sovereign rating"}, {"url": "https://www.bls.gov/regions/northeast/summary/blssummary_newyorkcity.pdf", "content": "New York City Economic Summary. Updated November 01, 2024. ... May 2023 New York area United States $39.45 $31.48 Prices paid by urban consumers for selected categories, over-the-year change Selling prices received by producers for selected industries nationwide, over-the-year change"}, {"url": "https://fred.stlouisfed.org/series/NYNGSP", "content": "Gross Domestic Product: All Industry Total in New York (NYNGSP) | FRED | St. Louis Fed FRED   About FRED About FRED What is FRED Gross Domestic Product: All Industry Total in New York (NYNGSP) Add data series to graph: U.S. Bureau of Economic Analysis, Gross Domestic Product: All Industry Total in New York [NYNGSP], retrieved from FRED, Federal Reserve Bank of St. Louis; https://fred.stlouisfed.org/series/NYNGSP, November 18, 2024. Real Gross Domestic Product: All Industry Total in New York Gross Domestic Product: All Industry Total in New Jersey Gross Domestic Product: All Industry Total in New York Gross State Product New York State Bureau of Economic Analysis Industry Gross Domestic Product Annual Public Domain: Citation Requested Not Seasonally Adjusted United States of America"}]', name='tavily_search_results_json', id='9ed353fe-b950-4b5f-9df1-4bc94514c3c4', tool_call_id='call_-9181536546572673495', artifact={'query': 'latest GDP of New York 2023', 'follow_up_questions': None, 'answer': None, 'images': [], 'results': [{'title': 'New York - GDP at market prices 2023 | countryeconomy.com', 'url': 'https://countryeconomy.com/gdp/usa-states/new-york', 'content': 'New York - GDP at market prices 2023 | countryeconomy.com US States GDP at market prices New York - GDP at market prices | 2023 | €2,008,702M | $2,172,010M | 1.5% | | 2022 | €1,949,439M | $2,052,759M | 1.7% | | 2021 | €1,626,290M | $1,923,413M | 4.8% | | 2020 | €1,552,591M | $1,773,370M | -2.8% | | 2019 | €1,596,669M | $1,787,471M | 2.4% | Evolution: GDP growth rate at constant prices New York Argentina: Fitch:Long Term Foreign currency Sovereign rating Barbados: Fitch:Long Term Foreign currency Sovereign rating Croatia: Fitch:Long Term Foreign currency Sovereign rating Egypt: Fitch:Long Term Foreign currency Sovereign rating Mongolia: Fitch:Long Term Foreign currency Sovereign rating Tunisia: Fitch:Long Term Foreign currency Sovereign rating', 'score': 0.99852777, 'raw_content': None}, {'title': 'PDF', 'url': 'https://www.bls.gov/regions/northeast/summary/blssummary_newyorkcity.pdf', 'content': 'New York City Economic Summary. Updated November 01, 2024. ... May 2023 New York area United States $39.45 $31.48 Prices paid by urban consumers for selected categories, over-the-year change Selling prices received by producers for selected industries nationwide, over-the-year change', 'score': 0.98862445, 'raw_content': None}, {'title': 'Gross Domestic Product: All Industry Total in New York', 'url': 'https://fred.stlouisfed.org/series/NYNGSP', 'content': 'Gross Domestic Product: All Industry Total in New York (NYNGSP) | FRED | St. Louis Fed FRED   About FRED About FRED What is FRED Gross Domestic Product: All Industry Total in New York (NYNGSP) Add data series to graph: U.S. Bureau of Economic Analysis, Gross Domestic Product: All Industry Total in New York [NYNGSP], retrieved from FRED, Federal Reserve Bank of St. Louis; https://fred.stlouisfed.org/series/NYNGSP, November 18, 2024. Real Gross Domestic Product: All Industry Total in New York Gross Domestic Product: All Industry Total in New Jersey Gross Domestic Product: All Industry Total in New York Gross State Product New York State Bureau of Economic Analysis Industry Gross Domestic Product Annual Public Domain: Citation Requested Not Seasonally Adjusted United States of America', 'score': 0.85452086, 'raw_content': None}], 'response_time': 1.76}), ToolMessage(content='[{"url": "https://www.gov.ca.gov/2024/04/16/california-remains-the-worlds-5th-largest-economy/", "content": "California remains the 5th largest economy in the world since 2017. California is the 5th largest economy in the world for the seventh consecutive year, with a nominal GDP of nearly $3.9 trillion in 2023 and a growth rate of 6.1% since the year prior, according to the U.S. Bureau of Economic Analysis (BEA). On a per capita basis, California is"}, {"url": "https://www.statista.com/statistics/187834/gdp-of-the-us-federal-state-of-california-since-1997/", "content": "Industry Overview\\nDigital & Trend reports\\nOverview and forecasts on trending topics\\nIndustry & Market reports\\nIndustry and market insights and forecasts\\nCompanies & Products reports\\nKey figures and rankings about companies and products\\nConsumer & Brand reports\\nConsumer and brand insights and preferences in various industries\\nPolitics & Society reports\\nDetailed information about political and social topics\\nCountry & Region reports\\nAll key figures about countries and regions\\nMarket forecast and expert KPIs for 1000+ markets in 190+ countries & territories\\nInsights on consumer attitudes and behavior worldwide\\nBusiness information on 100m+ public and private companies\\nExplore Company Insights\\nDetailed information for 39,000+ online stores and marketplaces\\nDirectly accessible data for 170 industries from 150+ countries\\nand over 1\xa0Mio. facts.\\n Statistics on\\n\\"\\nCalifornia\\n\\"\\nOther statistics that may interest you California\\nPopulation\\nEconomy\\nEmployment & Earnings\\nState & Local Government\\nMetro Areas\\nFurther related statistics\\nFurther Content: You might find this interesting as well\\nStatistics\\nTopics Other statistics on the topicCalifornia\\nEconomy\\nU.S. leading companies headquartered in California 2023, by number of employees\\nEconomy\\nU.S. average annual wages in California 2018-2026\\nEconomy\\nU.S. California fastest growing private companies 2023, by three year growth rate\\nResidential Real Estate\\nHourly wages needed to afford a two-bedroom apartment in California 2021-23, by metro\\nYou only have access to basic statistics.\\n Additional Information\\nShow sources information\\nShow publisher information\\nUse Ask Statista Research Service\\nMarch 2023\\nUnited States\\n2000 to 2022\\nData presented here is in 2012 chained U.S. dollars.\\n Transforming data into design:\\nStatista Content & Design\\nStrategy and business building for the data-driven economy:\\nU.S. real GDP of California 2000-2022\\nReal gross domestic product of California in the United States from 2000 to 2022\\n(in billion U.S. dollars)\\n"}, {"url": "https://www.bea.gov/news/2024/gross-domestic-product-fourth-quarter-and-year-2023-advance-estimate", "content": "Gross Domestic Product, Fourth Quarter and Year 2023 (Advance Estimate) | U.S. Bureau of Economic Analysis (BEA) Real gross domestic product (GDP) increased at an annual rate of 3.3 percent in the fourth quarter of 2023 (table 1), according to the \\"advance\\" estimate released by the Bureau of Economic Analysis. The price index for gross domestic purchases increased 1.9 percent in the fourth quarter, compared with an increase of 2.9 percent in the third quarter (table 4). BEA releases three vintages of the current quarterly estimate for GDP. Unlike GDP, advance current quarterly estimates of GDI and corporate profits are not released because data on domestic profits and net interest of domestic industries are not available."}]', name='tavily_search_results_json', id='ae678022-d845-45fc-8273-e6d145ee952d', tool_call_id='call_-9181536546572673494', artifact={'query': 'latest GDP of California 2023', 'follow_up_questions': None, 'answer': None, 'images': [], 'results': [{'title': "California Remains the World's 5th Largest Economy", 'url': 'https://www.gov.ca.gov/2024/04/16/california-remains-the-worlds-5th-largest-economy/', 'content': 'California remains the 5th largest economy in the world since 2017. California is the 5th largest economy in the world for the seventh consecutive year, with a nominal GDP of nearly $3.9 trillion in 2023 and a growth rate of 6.1% since the year prior, according to the U.S. Bureau of Economic Analysis (BEA). On a per capita basis, California is', 'score': 0.9961755, 'raw_content': None}, {'title': 'Real GDP California U.S. 2023 | Statista', 'url': 'https://www.statista.com/statistics/187834/gdp-of-the-us-federal-state-of-california-since-1997/', 'content': 'Industry Overview\nDigital & Trend reports\nOverview and forecasts on trending topics\nIndustry & Market reports\nIndustry and market insights and forecasts\nCompanies & Products reports\nKey figures and rankings about companies and products\nConsumer & Brand reports\nConsumer and brand insights and preferences in various industries\nPolitics & Society reports\nDetailed information about political and social topics\nCountry & Region reports\nAll key figures about countries and regions\nMarket forecast and expert KPIs for 1000+ markets in 190+ countries & territories\nInsights on consumer attitudes and behavior worldwide\nBusiness information on 100m+ public and private companies\nExplore Company Insights\nDetailed information for 39,000+ online stores and marketplaces\nDirectly accessible data for 170 industries from 150+ countries\nand over 1\xa0Mio. facts.\n Statistics on\n"\nCalifornia\n"\nOther statistics that may interest you California\nPopulation\nEconomy\nEmployment & Earnings\nState & Local Government\nMetro Areas\nFurther related statistics\nFurther Content: You might find this interesting as well\nStatistics\nTopics Other statistics on the topicCalifornia\nEconomy\nU.S. leading companies headquartered in California 2023, by number of employees\nEconomy\nU.S. average annual wages in California 2018-2026\nEconomy\nU.S. California fastest growing private companies 2023, by three year growth rate\nResidential Real Estate\nHourly wages needed to afford a two-bedroom apartment in California 2021-23, by metro\nYou only have access to basic statistics.\n Additional Information\nShow sources information\nShow publisher information\nUse Ask Statista Research Service\nMarch 2023\nUnited States\n2000 to 2022\nData presented here is in 2012 chained U.S. dollars.\n Transforming data into design:\nStatista Content & Design\nStrategy and business building for the data-driven economy:\nU.S. real GDP of California 2000-2022\nReal gross domestic product of California in the United States from 2000 to 2022\n(in billion U.S. dollars)\n', 'score': 0.60323083, 'raw_content': None}, {'title': 'Gross Domestic Product, Fourth Quarter and Year 2023 (Advance Estimate)', 'url': 'https://www.bea.gov/news/2024/gross-domestic-product-fourth-quarter-and-year-2023-advance-estimate', 'content': 'Gross Domestic Product, Fourth Quarter and Year 2023 (Advance Estimate) | U.S. Bureau of Economic Analysis (BEA) Real gross domestic product (GDP) increased at an annual rate of 3.3 percent in the fourth quarter of 2023 (table 1), according to the "advance" estimate released by the Bureau of Economic Analysis. The price index for gross domestic purchases increased 1.9 percent in the fourth quarter, compared with an increase of 2.9 percent in the third quarter (table 4). BEA releases three vintages of the current quarterly estimate for GDP. Unlike GDP, advance current quarterly estimates of GDI and corporate profits are not released because data on domestic profits and net interest of domestic industries are not available.', 'score': 0.5940803, 'raw_content': None}], 'response_time': 2.92})]}})
----
(('researcher:1c6c3cb9-b2d8-3632-e480-a307b53f1d77',), {'agent': {'messages': [AIMessage(content="The latest GDP of New York in 2023 is approximately $2,172,010 million, and the latest GDP of California in 2023 is nearly $3.9 trillion, which is $3,900,000 million.\n\nTo calculate the average GDP of the two states, you would add these two figures together and then divide by 2. \n\nPlease note that for accurate calculations, it's important to ensure that the units are consistent (both in millions or both in trillions). The figures provided here are in millions. \n\nYou can proceed with the calculation using these figures.", additional_kwargs={'refusal': None}, response_metadata={'token_usage': {'completion_tokens': 121, 'prompt_tokens': 1587, 'total_tokens': 1708, 'completion_tokens_details': None, 'prompt_tokens_details': None}, 'model_name': 'GLM-4-plus', 'system_fingerprint': None, 'finish_reason': 'stop', 'logprobs': None}, id='run-752b03c7-f372-447d-9440-97b513f0ae2b-0', usage_metadata={'input_tokens': 1587, 'output_tokens': 121, 'total_tokens': 1708, 'input_token_details': {}, 'output_token_details': {}})]}})
----
((), {'researcher': {'messages': [HumanMessage(content="The latest GDP of New York in 2023 is approximately $2,172,010 million, and the latest GDP of California in 2023 is nearly $3.9 trillion, which is $3,900,000 million.\n\nTo calculate the average GDP of the two states, you would add these two figures together and then divide by 2. \n\nPlease note that for accurate calculations, it's important to ensure that the units are consistent (both in millions or both in trillions). The figures provided here are in millions. \n\nYou can proceed with the calculation using these figures.", additional_kwargs={}, response_metadata={}, name='researcher')]}})
----
((), {'supervisor': {'next': 'coder'}})
----
(('coder:08b90861-7717-a01e-8178-a2245ce71495',), {'agent': {'messages': [AIMessage(content='', additional_kwargs={'tool_calls': [{'id': 'call_-9181530190019420575', 'function': {'arguments': '{\n    "code": "def calculate_average_gdp(gdp_new_york, gdp_california):\\n    return (gdp_new_york + gdp_california) / 2\\n\\n# GDP values in millions\\nlatest_gdp_new_york = 2172010\\nlatest_gdp_california = 3900000\\n\\naverage_gdp = calculate_average_gdp(latest_gdp_new_york, latest_gdp_california)\\nprint(f\\"The average GDP of New York and California is: {average_gdp} million\\")\\n"\n}', 'name': 'code_executor'}, 'type': 'function', 'index': 0}], 'refusal': None}, response_metadata={'token_usage': {'completion_tokens': 123, 'prompt_tokens': 316, 'total_tokens': 439, 'completion_tokens_details': None, 'prompt_tokens_details': None}, 'model_name': 'GLM-4-plus', 'system_fingerprint': None, 'finish_reason': 'tool_calls', 'logprobs': None}, id='run-ccf3ab4e-8e24-480c-8935-b09236974eeb-0', tool_calls=[{'name': 'code_executor', 'args': {'code': 'def calculate_average_gdp(gdp_new_york, gdp_california):\n    return (gdp_new_york + gdp_california) / 2\n\n# GDP values in millions\nlatest_gdp_new_york = 2172010\nlatest_gdp_california = 3900000\n\naverage_gdp = calculate_average_gdp(latest_gdp_new_york, latest_gdp_california)\nprint(f"The average GDP of New York and California is: {average_gdp} million")\n'}, 'id': 'call_-9181530190019420575', 'type': 'tool_call'}], usage_metadata={'input_tokens': 316, 'output_tokens': 123, 'total_tokens': 439, 'input_token_details': {}, 'output_token_details': {}})]}})
----
(('coder:08b90861-7717-a01e-8178-a2245ce71495',), {'tools': {'messages': [ToolMessage(content='The result is:The average GDP of New York and California is: 3036005.0 million\n', name='code_executor', id='63ae3dcf-e81d-49f9-a705-1d87091554e8', tool_call_id='call_-9181530190019420575')]}})
----
(('coder:08b90861-7717-a01e-8178-a2245ce71495',), {'agent': {'messages': [AIMessage(content='The average GDP of New York and California is approximately 3,036,005 million. This figure represents the midpoint of their respective GDPs, providing a general sense of the economic scale of these two states combined.', additional_kwargs={'refusal': None}, response_metadata={'token_usage': {'completion_tokens': 47, 'prompt_tokens': 459, 'total_tokens': 506, 'completion_tokens_details': None, 'prompt_tokens_details': None}, 'model_name': 'GLM-4-plus', 'system_fingerprint': None, 'finish_reason': 'stop', 'logprobs': None}, id='run-df630065-cd91-45ba-9fab-376ec79f9f60-0', usage_metadata={'input_tokens': 459, 'output_tokens': 47, 'total_tokens': 506, 'input_token_details': {}, 'output_token_details': {}})]}})
----
((), {'coder': {'messages': [HumanMessage(content='The average GDP of New York and California is approximately 3,036,005 million. This figure represents the midpoint of their respective GDPs, providing a general sense of the economic scale of these two states combined.', additional_kwargs={}, response_metadata={}, name='coder')]}})
----
((), {'supervisor': {'next': '__end__'}})
----

参考链接：https://langchain-ai.github.io/langgraph/tutorials/multi_agent/agent_supervisor/

如果有任何问题，欢迎在评论区提问。