crewai-agents

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

CrewAI Multi-Agent Orchestration

CrewAI 多智能体编排

Build teams of autonomous AI agents that collaborate on complex tasks.

构建可协作完成复杂任务的自主AI智能体团队。

When to Use

使用场景

Building multi-agent systems with specialized roles
Need autonomous collaboration between agents
Want role-based task delegation (researcher, writer, analyst)
Require sequential or hierarchical process execution
Building production workflows with memory and observability

构建具有特定角色的多智能体系统
需要智能体之间的自主协作
希望实现基于角色的任务分配（研究员、撰稿人、分析师）
需要执行顺序或分层流程
构建具备记忆和可观测性的生产级工作流

Quick Start

快速开始

bash

pip install crewai
pip install 'crewai[tools]'  # With 50+ built-in tools

bash

pip install crewai
pip install 'crewai[tools]'  # 包含50+内置工具

Simple Crew

简单协作组示例

python

from crewai import Agent, Task, Crew, Process

python

from crewai import Agent, Task, Crew, Process

Define agents

定义智能体

researcher = Agent( role="Senior Research Analyst", goal="Discover cutting-edge developments in AI", backstory="You are an expert analyst with a keen eye for trends.", verbose=True )

writer = Agent( role="Technical Writer", goal="Create clear, engaging content about technical topics", backstory="You excel at explaining complex concepts.", verbose=True )

researcher = Agent( role="Senior Research Analyst", goal="Discover cutting-edge developments in AI", backstory="You are an expert analyst with a keen eye for trends.", verbose=True )

writer = Agent( role="Technical Writer", goal="Create clear, engaging content about technical topics", backstory="You excel at explaining complex concepts.", verbose=True )

Define tasks

定义任务

research_task = Task( description="Research the latest developments in {topic}. Find 5 key trends.", expected_output="A detailed report with 5 bullet points.", agent=researcher )

write_task = Task( description="Write a blog post based on the research findings.", expected_output="A 500-word blog post in markdown format.", agent=writer, context=[research_task] # Uses research output )

research_task = Task( description="Research the latest developments in {topic}. Find 5 key trends.", expected_output="A detailed report with 5 bullet points.", agent=researcher )

Create and run crew

创建并运行协作组

crew = Crew( agents=[researcher, writer], tasks=[research_task, write_task], process=Process.sequential, verbose=True )

result = crew.kickoff(inputs={"topic": "AI Agents"}) print(result.raw)

undefined

crew = Crew( agents=[researcher, writer], tasks=[research_task, write_task], process=Process.sequential, verbose=True )

result = crew.kickoff(inputs={"topic": "AI Agents"}) print(result.raw)

undefined

Process Types

流程类型

Sequential

顺序流程

Tasks execute in order:

python

crew = Crew(
    agents=[researcher, writer],
    tasks=[research_task, write_task],
    process=Process.sequential  # Task 1 -> Task 2 -> Task 3
)

任务按顺序执行：

python

crew = Crew(
    agents=[researcher, writer],
    tasks=[research_task, write_task],
    process=Process.sequential  # 任务1 -> 任务2 -> 任务3
)

Hierarchical

分层流程

Auto-creates a manager agent that delegates:

python

crew = Crew(
    agents=[researcher, writer, analyst],
    tasks=[research_task, write_task, analyze_task],
    process=Process.hierarchical,
    manager_llm="gpt-4o"
)

自动创建管理智能体进行任务分配：

python

crew = Crew(
    agents=[researcher, writer, analyst],
    tasks=[research_task, write_task, analyze_task],
    process=Process.hierarchical,
    manager_llm="gpt-4o"
)

Using Tools

工具使用

python

from crewai_tools import (
    SerperDevTool,       # Web search
    ScrapeWebsiteTool,   # Web scraping
    FileReadTool,        # Read files
    PDFSearchTool,       # Search PDFs
)

researcher = Agent(
    role="Researcher",
    goal="Find accurate information",
    backstory="Expert at finding data online.",
    tools=[SerperDevTool(), ScrapeWebsiteTool()]
)

python

from crewai_tools import (
    SerperDevTool,       # 网页搜索
    ScrapeWebsiteTool,   # 网页爬取
    FileReadTool,        # 文件读取
    PDFSearchTool,       # PDF搜索
)

researcher = Agent(
    role="Researcher",
    goal="Find accurate information",
    backstory="Expert at finding data online.",
    tools=[SerperDevTool(), ScrapeWebsiteTool()]
)

Custom Tools

自定义工具

python

from crewai.tools import BaseTool

class CalculatorTool(BaseTool):
    name: str = "Calculator"
    description: str = "Performs mathematical calculations."

    def _run(self, expression: str) -> str:
        try:
            return f"Result: {eval(expression)}"
        except Exception as e:
            return f"Error: {str(e)}"

python

from crewai.tools import BaseTool

class CalculatorTool(BaseTool):
    name: str = "Calculator"
    description: str = "Performs mathematical calculations."

    def _run(self, expression: str) -> str:
        try:
            return f"Result: {eval(expression)}"
        except Exception as e:
            return f"Error: {str(e)}"

YAML Configuration

YAML配置

agents.yaml

yaml

researcher:
  role: "{topic} Senior Data Researcher"
  goal: "Uncover cutting-edge developments in {topic}"
  backstory: >
    You're a seasoned researcher with a knack for uncovering
    the latest developments in {topic}.

reporting_analyst:
  role: "Reporting Analyst"
  goal: "Create detailed reports based on research data"
  backstory: >
    You're a meticulous analyst who transforms raw data into
    actionable insights.

yaml

researcher:
  role: "{topic} Senior Data Researcher"
  goal: "Uncover cutting-edge developments in {topic}"
  backstory: >
    You're a seasoned researcher with a knack for uncovering
    the latest developments in {topic}.

reporting_analyst:
  role: "Reporting Analyst"
  goal: "Create detailed reports based on research data"
  backstory: >
    You're a meticulous analyst who transforms raw data into
    actionable insights.

tasks.yaml

yaml

research_task:
  description: >
    Conduct thorough research about {topic}.
    Find the most relevant information for {year}.
  expected_output: >
    A list with 10 bullet points of the most relevant
    information about {topic}.
  agent: researcher

reporting_task:
  description: >
    Review the research and create a comprehensive report.
  expected_output: >
    A detailed report in markdown format.
  agent: reporting_analyst
  output_file: report.md

yaml

research_task:
  description: >
    Conduct thorough research about {topic}.
    Find the most relevant information for {year}.
  expected_output: >
    A list with 10 bullet points of the most relevant
    information about {topic}.
  agent: researcher

reporting_task:
  description: >
    Review the research and create a comprehensive report.
  expected_output: >
    A detailed report in markdown format.
  agent: reporting_analyst
  output_file: report.md

Memory System

记忆系统

python

crew = Crew(
    agents=[researcher],
    tasks=[research_task],
    memory=True,  # Enable memory
    embedder={
        "provider": "openai",
        "config": {"model": "text-embedding-3-small"}
    }
)

python

crew = Crew(
    agents=[researcher],
    tasks=[research_task],
    memory=True,  # 启用记忆功能
    embedder={
        "provider": "openai",
        "config": {"model": "text-embedding-3-small"}
    }
)

LLM Providers

大语言模型提供商

python

from crewai import LLM

llm = LLM(model="gpt-4o")  # OpenAI
llm = LLM(model="claude-sonnet-4-5-20250929")  # Anthropic
llm = LLM(model="ollama/llama3.1", base_url="http://localhost:11434")  # Local

agent = Agent(role="Analyst", goal="Analyze data", llm=llm)

python

from crewai import LLM

llm = LLM(model="gpt-4o")  # OpenAI
llm = LLM(model="claude-sonnet-4-5-20250929")  # Anthropic
llm = LLM(model="ollama/llama3.1", base_url="http://localhost:11434")  # 本地模型

agent = Agent(role="Analyst", goal="Analyze data", llm=llm)

Best Practices

最佳实践

Clear roles - Each agent should have a distinct specialty
YAML config - Better organization for larger projects
Enable memory - Improves context across tasks
Set max_iter - Prevent infinite loops (default 15)
Limit tools - 3-5 tools per agent max
Rate limiting - Set max_rpm to avoid API limits

明确角色 - 每个智能体应具备独特的专业领域
YAML配置 - 大型项目的更佳组织方式
启用记忆 - 提升跨任务的上下文连贯性
设置max_iter - 防止无限循环（默认值为15）
限制工具数量 - 每个智能体最多使用3-5个工具
速率限制 - 设置max_rpm以避免API调用超限

Common Issues

常见问题

Agent stuck in loop:

python

agent = Agent(
    role="...",
    max_iter=10,
    max_rpm=5
)

Task not using context:

python

task2 = Task(
    description="...",
    context=[task1],  # Explicitly pass context
    agent=writer
)

智能体陷入循环：

python

agent = Agent(
    role="...",
    max_iter=10,
    max_rpm=5
)

任务未使用上下文：

python

task2 = Task(
    description="...",
    context=[task1],  # 显式传递上下文
    agent=writer
)

vs Alternatives

与竞品对比

Feature	CrewAI	LangChain	LangGraph
Best for	Multi-agent teams	General LLM apps	Stateful workflows
Learning curve	Low	Medium	Higher
Agent paradigm	Role-based	Tool-based	Graph-based

特性	CrewAI	LangChain	LangGraph
适用场景	多智能体团队	通用LLM应用	有状态工作流
学习曲线	低	中等	较高
智能体范式	基于角色	基于工具	基于图结构

Resources

资源

GitHub: https://github.com/crewAIInc/crewAI
Docs: https://docs.crewai.com
Tools: https://github.com/crewAIInc/crewAI-tools

GitHub: https://github.com/crewAIInc/crewAI
文档: https://docs.crewai.com
工具: https://github.com/crewAIInc/crewAI-tools