分层代理团队¶
在我们之前的示例（代理监督器）中，我们介绍了单个监督器节点的概念，用于在不同的工作节点之间分配工作。
但是，如果单个工作节点的任务变得过于复杂怎么办？如果工作节点的数量变得过多怎么办？
对于某些应用程序，如果工作以_分层_方式分配，系统可能会更有效。
你可以通过组合不同的子图并创建一个顶级监督器以及中级监督器来实现这一点。
为此，让我们构建一个简单的研究助手！该图大致如下所示：
本笔记本受到了 Wu 等人的论文 AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation 的启发。在本笔记本的其余部分，你将：
- 定义代理用于访问网络和写入文件的工具
- 定义一些有助于创建图和代理的实用工具
- 创建并定义每个团队（网络研究 + 文档撰写）
- 将所有内容组合在一起。
环境设置¶
首先，让我们安装所需的包并设置 API 密钥
%%capture --no-stderr
%pip install -U langgraph langchain_community langchain_anthropic langchain_experimental
import getpass
import os
def _set_if_undefined(var: str):
if not os.environ.get(var):
os.environ[var] = getpass.getpass(f"Please provide your {var}")
_set_if_undefined("OPENAI_API_KEY")
_set_if_undefined("TAVILY_API_KEY")
为 LangGraph 开发设置 LangSmith
注册 LangSmith，以便快速发现问题并提升你的 LangGraph 项目的性能。LangSmith 允许你使用跟踪数据来调试、测试和监控使用 LangGraph 构建的大语言模型应用程序 — 点击 此处 了解更多关于如何开始使用的信息。
创建工具¶
每个团队将由一个或多个智能体组成，每个智能体配备一个或多个工具。下面，请定义不同团队将使用的所有工具。
我们先从研究团队开始。
研究团队工具
研究团队可以使用搜索引擎和 URL 抓取器在网络上查找信息。你可以在下面随意添加额外的功能来提升团队的表现！
from typing import Annotated, List
from langchain_community.document_loaders import WebBaseLoader
from langchain_community.tools.tavily_search import TavilySearchResults
from langchain_core.tools import tool
tavily_tool = TavilySearchResults(max_results=5)
@tool
def scrape_webpages(urls: List[str]) -> str:
"""Use requests and bs4 to scrape the provided web pages for detailed information."""
loader = WebBaseLoader(urls)
docs = loader.load()
return "\n\n".join(
[
f'<Document name="{doc.metadata.get("title", "")}">\n{doc.page_content}\n</Document>'
for doc in docs
]
)
API Reference: WebBaseLoader | TavilySearchResults | tool
文档编写团队工具
接下来，我们将提供一些供文档编写团队使用的工具。 我们在下面定义了一些基础的文件访问工具。
请注意，这会让智能体（agents）访问你的文件系统，这可能存在安全风险。此外，我们尚未针对性能对工具描述进行优化。
from pathlib import Path
from tempfile import TemporaryDirectory
from typing import Dict, Optional
from langchain_experimental.utilities import PythonREPL
from typing_extensions import TypedDict
_TEMP_DIRECTORY = TemporaryDirectory()
WORKING_DIRECTORY = Path(_TEMP_DIRECTORY.name)
@tool
def create_outline(
points: Annotated[List[str], "List of main points or sections."],
file_name: Annotated[str, "File path to save the outline."],
) -> Annotated[str, "Path of the saved outline file."]:
"""Create and save an outline."""
with (WORKING_DIRECTORY / file_name).open("w") as file:
for i, point in enumerate(points):
file.write(f"{i + 1}. {point}\n")
return f"Outline saved to {file_name}"
@tool
def read_document(
file_name: Annotated[str, "File path to read the document from."],
start: Annotated[Optional[int], "The start line. Default is 0"] = None,
end: Annotated[Optional[int], "The end line. Default is None"] = None,
) -> str:
"""Read the specified document."""
with (WORKING_DIRECTORY / file_name).open("r") as file:
lines = file.readlines()
if start is None:
start = 0
return "\n".join(lines[start:end])
@tool
def write_document(
content: Annotated[str, "Text content to be written into the document."],
file_name: Annotated[str, "File path to save the document."],
) -> Annotated[str, "Path of the saved document file."]:
"""Create and save a text document."""
with (WORKING_DIRECTORY / file_name).open("w") as file:
file.write(content)
return f"Document saved to {file_name}"
@tool
def edit_document(
file_name: Annotated[str, "Path of the document to be edited."],
inserts: Annotated[
Dict[int, str],
"Dictionary where key is the line number (1-indexed) and value is the text to be inserted at that line.",
],
) -> Annotated[str, "Path of the edited document file."]:
"""Edit a document by inserting text at specific line numbers."""
with (WORKING_DIRECTORY / file_name).open("r") as file:
lines = file.readlines()
sorted_inserts = sorted(inserts.items())
for line_number, text in sorted_inserts:
if 1 <= line_number <= len(lines) + 1:
lines.insert(line_number - 1, text + "\n")
else:
return f"Error: Line number {line_number} is out of range."
with (WORKING_DIRECTORY / file_name).open("w") as file:
file.writelines(lines)
return f"Document edited and saved to {file_name}"
# Warning: This executes code locally, which can be unsafe when not sandboxed
repl = PythonREPL()
@tool
def python_repl_tool(
code: Annotated[str, "The python code to execute to generate your chart."],
):
"""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 = repl.run(code)
except BaseException as e:
return f"Failed to execute. Error: {repr(e)}"
return f"Successfully executed:\n\`\`\`python\n{code}\n\`\`\`\nStdout: {result}"
辅助工具函数¶
我们将创建几个工具函数，以便在以下场景中使代码更加简洁：
- 创建一个工作代理。
- 为子图创建一个监管器。
这些工具函数将在最后简化我们的图组合代码，让我们更容易理解代码的逻辑。
from typing import List, Optional, Literal
from langchain_core.language_models.chat_models import BaseChatModel
from langgraph.graph import StateGraph, MessagesState, START, END
from langgraph.types import Command
from langchain_core.messages import HumanMessage, trim_messages
class State(MessagesState):
next: str
def make_supervisor_node(llm: BaseChatModel, members: list[str]) -> str:
options = ["FINISH"] + members
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."
)
class Router(TypedDict):
"""Worker to route to next. If no workers needed, route to FINISH."""
next: Literal[*options]
def supervisor_node(state: State) -> Command[Literal[*members, "__end__"]]:
"""An LLM-based router."""
messages = [
{"role": "system", "content": system_prompt},
] + state["messages"]
response = llm.with_structured_output(Router).invoke(messages)
goto = response["next"]
if goto == "FINISH":
goto = END
return Command(goto=goto, update={"next": goto})
return supervisor_node
API Reference: BaseChatModel | HumanMessage | trim_messages
定义代理团队¶
现在我们可以开始定义我们的层级团队了。“选择你的玩家！”
研究团队¶
研究团队将有一个搜索代理和一个网页抓取“研究代理”作为两个工作节点。让我们创建这些代理以及团队主管。
from langchain_core.messages import HumanMessage
from langchain_openai import ChatOpenAI
from langgraph.prebuilt import create_react_agent
llm = ChatOpenAI(model="gpt-4o")
search_agent = create_react_agent(llm, tools=[tavily_tool])
def search_node(state: State) -> Command[Literal["supervisor"]]:
result = search_agent.invoke(state)
return Command(
update={
"messages": [
HumanMessage(content=result["messages"][-1].content, name="search")
]
},
# We want our workers to ALWAYS "report back" to the supervisor when done
goto="supervisor",
)
web_scraper_agent = create_react_agent(llm, tools=[scrape_webpages])
def web_scraper_node(state: State) -> Command[Literal["supervisor"]]:
result = web_scraper_agent.invoke(state)
return Command(
update={
"messages": [
HumanMessage(content=result["messages"][-1].content, name="web_scraper")
]
},
# We want our workers to ALWAYS "report back" to the supervisor when done
goto="supervisor",
)
research_supervisor_node = make_supervisor_node(llm, ["search", "web_scraper"])
API Reference: HumanMessage
既然我们已经创建了必要的组件，那么定义它们之间的交互就很容易了。将节点添加到团队图中，并定义边，边决定了转换条件。
research_builder = StateGraph(State)
research_builder.add_node("supervisor", research_supervisor_node)
research_builder.add_node("search", search_node)
research_builder.add_node("web_scraper", web_scraper_node)
research_builder.add_edge(START, "supervisor")
research_graph = research_builder.compile()
from IPython.display import Image, display
display(Image(research_graph.get_graph().draw_mermaid_png()))
我们可以直接将这项团队工作分配给他们。在下面试试看。
for s in research_graph.stream(
{"messages": [("user", "when is Taylor Swift's next tour?")]},
{"recursion_limit": 100},
):
print(s)
print("---")
{'supervisor': {'next': 'search'}}
---
{'search': {'messages': [HumanMessage(content="Taylor Swift's next tour is The Eras Tour, which includes both U.S. and international dates. She announced additional U.S. dates for 2024. You can find more details about the tour and ticket information on platforms like Ticketmaster and official announcements.", additional_kwargs={}, response_metadata={}, name='search', id='4df8687b-50a8-4342-aad5-680732c4a10f')]}}
---
{'supervisor': {'next': 'web_scraper'}}
---
{'web_scraper': {'messages': [HumanMessage(content='Taylor Swift\'s next tour is "The Eras Tour." Here are some of the upcoming international dates for 2024 that were listed on Ticketmaster:\n\n1. **Toronto, ON, Canada** at Rogers Centre\n - November 21, 2024\n - November 22, 2024\n - November 23, 2024\n\n2. **Vancouver, BC, Canada** at BC Place\n - December 6, 2024\n - December 7, 2024\n - December 8, 2024\n\nFor the most current information and additional dates, you can check platforms like Ticketmaster or Taylor Swift\'s [official website](https://www.taylorswift.com/events).', additional_kwargs={}, response_metadata={}, name='web_scraper', id='27524ebc-d179-4733-831d-ee10a58a2528')]}}
---
{'supervisor': {'next': '__end__'}}
---
文档编写团队¶
使用类似的方法创建下面的文档编写团队。这一次，我们将为每个智能体提供不同的文件编写工具。
请注意，我们在这里为智能体提供了文件系统访问权限，这在某些情况下并不安全。
llm = ChatOpenAI(model="gpt-4o")
doc_writer_agent = create_react_agent(
llm,
tools=[write_document, edit_document, read_document],
prompt=(
"You can read, write and edit documents based on note-taker's outlines. "
"Don't ask follow-up questions."
),
)
def doc_writing_node(state: State) -> Command[Literal["supervisor"]]:
result = doc_writer_agent.invoke(state)
return Command(
update={
"messages": [
HumanMessage(content=result["messages"][-1].content, name="doc_writer")
]
},
# We want our workers to ALWAYS "report back" to the supervisor when done
goto="supervisor",
)
note_taking_agent = create_react_agent(
llm,
tools=[create_outline, read_document],
prompt=(
"You can read documents and create outlines for the document writer. "
"Don't ask follow-up questions."
),
)
def note_taking_node(state: State) -> Command[Literal["supervisor"]]:
result = note_taking_agent.invoke(state)
return Command(
update={
"messages": [
HumanMessage(content=result["messages"][-1].content, name="note_taker")
]
},
# We want our workers to ALWAYS "report back" to the supervisor when done
goto="supervisor",
)
chart_generating_agent = create_react_agent(
llm, tools=[read_document, python_repl_tool]
)
def chart_generating_node(state: State) -> Command[Literal["supervisor"]]:
result = chart_generating_agent.invoke(state)
return Command(
update={
"messages": [
HumanMessage(
content=result["messages"][-1].content, name="chart_generator"
)
]
},
# We want our workers to ALWAYS "report back" to the supervisor when done
goto="supervisor",
)
doc_writing_supervisor_node = make_supervisor_node(
llm, ["doc_writer", "note_taker", "chart_generator"]
)
在创建好对象本身之后，我们就可以构建图了。
# Create the graph here
paper_writing_builder = StateGraph(State)
paper_writing_builder.add_node("supervisor", doc_writing_supervisor_node)
paper_writing_builder.add_node("doc_writer", doc_writing_node)
paper_writing_builder.add_node("note_taker", note_taking_node)
paper_writing_builder.add_node("chart_generator", chart_generating_node)
paper_writing_builder.add_edge(START, "supervisor")
paper_writing_graph = paper_writing_builder.compile()
from IPython.display import Image, display
display(Image(paper_writing_graph.get_graph().draw_mermaid_png()))
for s in paper_writing_graph.stream(
{
"messages": [
(
"user",
"Write an outline for poem about cats and then write the poem to disk.",
)
]
},
{"recursion_limit": 100},
):
print(s)
print("---")
{'supervisor': {'next': 'note_taker'}}
---
{'note_taker': {'messages': [HumanMessage(content='The outline for the poem about cats has been created and saved as "cats_poem_outline.txt".', additional_kwargs={}, response_metadata={}, name='note_taker', id='14a5d8ca-9092-416f-96ee-ba16686e8658')]}}
---
{'supervisor': {'next': 'doc_writer'}}
---
{'doc_writer': {'messages': [HumanMessage(content='The poem about cats has been written and saved as "cats_poem.txt".', additional_kwargs={}, response_metadata={}, name='doc_writer', id='c4e31a94-63ae-4632-9e80-1166f3f138b2')]}}
---
{'supervisor': {'next': '__end__'}}
---
添加层¶
在这个设计中，我们正在实施自上而下的规划策略。我们已经创建了两个图，但我们必须决定如何在这两个图之间分配工作。
我们将创建一个“第三个”图来编排前两个图，并添加一些连接器来定义如何在不同的图之间共享这个顶级状态。
from langchain_core.messages import BaseMessage
llm = ChatOpenAI(model="gpt-4o")
teams_supervisor_node = make_supervisor_node(llm, ["research_team", "writing_team"])
API Reference: BaseMessage
def call_research_team(state: State) -> Command[Literal["supervisor"]]:
response = research_graph.invoke({"messages": state["messages"][-1]})
return Command(
update={
"messages": [
HumanMessage(
content=response["messages"][-1].content, name="research_team"
)
]
},
goto="supervisor",
)
def call_paper_writing_team(state: State) -> Command[Literal["supervisor"]]:
response = paper_writing_graph.invoke({"messages": state["messages"][-1]})
return Command(
update={
"messages": [
HumanMessage(
content=response["messages"][-1].content, name="writing_team"
)
]
},
goto="supervisor",
)
# Define the graph.
super_builder = StateGraph(State)
super_builder.add_node("supervisor", teams_supervisor_node)
super_builder.add_node("research_team", call_research_team)
super_builder.add_node("writing_team", call_paper_writing_team)
super_builder.add_edge(START, "supervisor")
super_graph = super_builder.compile()
from IPython.display import Image, display
display(Image(super_graph.get_graph().draw_mermaid_png()))
for s in super_graph.stream(
{
"messages": [
("user", "Research AI agents and write a brief report about them.")
],
},
{"recursion_limit": 150},
):
print(s)
print("---")
{'supervisor': {'next': 'research_team'}}
---
{'research_team': {'messages': [HumanMessage(content="**AI Agents Overview 2023**\n\nAI agents are sophisticated technologies that automate and enhance various processes across industries, becoming increasingly integral to business operations. In 2023, these agents are notable for their advanced capabilities in communication, data visualization, and language processing.\n\n**Popular AI Agents in 2023:**\n1. **Auto GPT**: This agent is renowned for its seamless integration abilities, significantly impacting industries by improving communication and operational workflows.\n2. **ChartGPT**: Specializing in data visualization, ChartGPT enables users to interact with data innovatively, providing deeper insights and comprehension.\n3. **LLMops**: With advanced language capabilities, LLMops is a versatile tool seeing widespread use across multiple sectors.\n\n**Market Trends:**\nThe AI agents market is experiencing rapid growth, with significant advancements anticipated by 2030. There's a growing demand for AI agents in personalized interactions, particularly within customer service, healthcare, and marketing sectors. This trend is fueled by the need for more efficient and tailored customer experiences.\n\n**Key Players:**\nLeading companies such as Microsoft, IBM, Google, Oracle, and AWS are key players in the AI agents market, highlighting the widespread adoption and investment in these technologies.\n\n**Technological Innovations:**\nAI agents are being developed alongside simulation technologies for robust testing and deployment environments. Innovations in generative AI are accelerating, supported by advancements in large language models and platforms like ChatGPT.\n\n**Applications in Healthcare:**\nIn healthcare, AI agents are automating routine tasks, allowing medical professionals to focus more on patient care. They're poised to significantly enhance healthcare delivery and efficiency.\n\n**Future Prospects:**\nThe future of AI agents is promising, with continued evolution and integration into various platforms and ecosystems, offering more seamless and intelligent interactions. As these technologies advance, they are expected to redefine business operations and customer interactions.", additional_kwargs={}, response_metadata={}, name='research_team', id='5f6606e0-838c-406c-b50d-9f9f6a076322')]}}
---
{'supervisor': {'next': 'writing_team'}}
---
{'writing_team': {'messages': [HumanMessage(content="Here are the contents of the documents:\n\n### AI Agents Overview 2023\n\n**AI Agents Overview 2023**\n\nAI agents are sophisticated technologies that automate and enhance various processes across industries, becoming increasingly integral to business operations. In 2023, these agents are notable for their advanced capabilities in communication, data visualization, and language processing.\n\n**Popular AI Agents in 2023:**\n1. **Auto GPT**: This agent is renowned for its seamless integration abilities, significantly impacting industries by improving communication and operational workflows.\n2. **ChartGPT**: Specializing in data visualization, ChartGPT enables users to interact with data innovatively, providing deeper insights and comprehension.\n3. **LLMops**: With advanced language capabilities, LLMops is a versatile tool seeing widespread use across multiple sectors.\n\n**Market Trends:**\nThe AI agents market is experiencing rapid growth, with significant advancements anticipated by 2030. There's a growing demand for AI agents in personalized interactions, particularly within customer service, healthcare, and marketing sectors. This trend is fueled by the need for more efficient and tailored customer experiences.\n\n**Key Players:**\nLeading companies such as Microsoft, IBM, Google, Oracle, and AWS are key players in the AI agents market, highlighting the widespread adoption and investment in these technologies.\n\n**Technological Innovations:**\nAI agents are being developed alongside simulation technologies for robust testing and deployment environments. Innovations in generative AI are accelerating, supported by advancements in large language models and platforms like ChatGPT.\n\n**Applications in Healthcare:**\nIn healthcare, AI agents are automating routine tasks, allowing medical professionals to focus more on patient care. They're poised to significantly enhance healthcare delivery and efficiency.\n\n**Future Prospects:**\nThe future of AI agents is promising, with continued evolution and integration into various platforms and ecosystems, offering more seamless and intelligent interactions. As these technologies advance, they are expected to redefine business operations and customer interactions.\n\n### AI_Agents_Overview_2023_Outline\n\n1. Introduction to AI Agents in 2023\n2. Popular AI Agents: Auto GPT, ChartGPT, LLMops\n3. Market Trends and Growth\n4. Key Players in the AI Agents Market\n5. Technological Innovations: Simulation and Generative AI\n6. Applications of AI Agents in Healthcare\n7. Future Prospects of AI Agents", additional_kwargs={}, response_metadata={}, name='writing_team', id='851bd8a6-740e-488c-8928-1f9e05e96ea0')]}}
---
{'supervisor': {'next': 'writing_team'}}
---
{'writing_team': {'messages': [HumanMessage(content='The documents have been successfully created and saved:\n\n1. **AI_Agents_Overview_2023.txt** - Contains the detailed overview of AI agents in 2023.\n2. **AI_Agents_Overview_2023_Outline.txt** - Contains the outline of the document.', additional_kwargs={}, response_metadata={}, name='writing_team', id='c87c0778-a085-4a8e-8ee1-9b43b9b0b143')]}}
---
{'supervisor': {'next': '__end__'}}
---