1 初次使用 240109

1.1 项目地址

https://github.com/joaomdmoura/crewAI

1.2 功能

通过设置多个智能体,协同解决问题,以处理复杂任务;这种方法的实现方式是将一个任务的输出作为另一个任务的输入。它的优势在于小而有效,原理直观易懂,而且所需的调用代码也相当简单。

很多工作需要多次交互才能完成,不同角色 的 Agent 可设置成不同模型,不同辅助工具,非常好用。

1.3 代码分析

当前版本 python 代码 800 多行,但几乎是我看过最简单好用的多工具组合策略。

1.4 模型

除了 openai,还可以支持本地搭建的模型 ollama。

1.5 安装

1
$ pip install crewai

1.6 代码

1.6.1 示例

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
import os
from crewai import Agent, Task, Crew, Process

os.environ["OPENAI_API_KEY"] = "YOUR KEY"

from langchain.tools import DuckDuckGoSearchRun
search_tool = DuckDuckGoSearchRun()

# Define your agents with roles and goals
researcher = Agent(
role='Senior Research Analyst',
goal='Uncover cutting-edge developments in AI and data science in',
backstory="""You work at a leading tech think tank.
Your expertise lies in identifying emerging trends.
You have a knack for dissecting complex data and presenting
actionable insights.""",
verbose=True,
allow_delegation=False,
tools=[search_tool]
)
writer = Agent(
role='Tech Content Strategist',
goal='Craft compelling content on tech advancements',
backstory="""You are a renowned Content Strategist, known for
your insightful and engaging articles.
You transform complex concepts into compelling narratives.""",
verbose=True,
allow_delegation=True,
)

# Create tasks for your agents
task1 = Task(
description="""Conduct a comprehensive analysis of the latest advancements in AI in 2024.
Identify key trends, breakthrough technologies, and potential industry impacts.
Your final answer MUST be a full analysis report""",
agent=researcher
)

task2 = Task(
description="""Using the insights provided, develop an engaging blog
post that highlights the most significant AI advancements.
Your post should be informative yet accessible, catering to a tech-savvy audience.
Make it sound cool, avoid complex words so it doesn't sound like AI.
Your final answer MUST be the full blog post of at least 4 paragraphs.""",
agent=writer
)

# Instantiate your crew with a sequential process
crew = Crew(
agents=[researcher, writer],
tasks=[task1, task2],
verbose=2, # You can set it to 1 or 2 to different logging levels
)

# Get your crew to work!
result = crew.kickoff()

print("######################")
print(result)

1.6.2 代码分析

在上述代码中,定义了两个智能体:分析师和内容创作者。他们共同协作,完成了一篇关于预测 AI 2014 发展趋势的博文。

值得一提的是,在这里使用 Langchain 来调用 DuckDuckGo 搜索引擎,以便分析师智能体收集相关信息。具体来说,首先,分析师智能体会利用收集到的信息生成分析结果;然后,这些分析结果将被传递给内容创作者智能体,由他来撰写博文。

1.6.3 注意事项

模型默认调用 OpenAI GPT-4,测试花费比较高,如果介意,请在 Agent 中设置成便宜的版本

2 文档更新 241222

  • 核心功能:这款产品主打多 Agent 协作,当前的 LLM 能处理如此复杂的问题吗?
  • 使用感受:整体氛围更加自由。
  • 概念:
    • 项目将 " 做什么 " 和 " 怎么做 " 进行了区分。
    • Task:需要实现的目标。
    • Agent:所提供的能力。
    • 底层不同模型的差异交由 langchain 处理,项目内部通过 llm.py 来弥合这些差异,主要侧重于国外的大模型。
  • 支持的模型:包括 openai、anthropic、gemini、groq、ollama、watson、bedrock、azure、cerebras 等。
  • 代码:
    • 总计约 1.7 万行代码,其中 src 目录中约有 1 万行,其余为测试用例。
    • 例程与主项目分开,主要针对 tools 的示例,请查看 https://github.com/crewAIInc/crewAI-examples。
  • 用法:
    • 外接项目 crew-tools 提供了一些非常丰富的工具,可以从 https://github.com/crewAIInc/crewAI-tools 获取。
    • 该项目也支持使用 langchain 的工具,这一点设计得很巧妙。
    • 由于依赖于多种外部工具,最新版本需要安装的依赖库非常多,与当前的工具链 langchain 版本可能存在冲突。