# CrewAI — Multi-Agent Orchestration Framework

> Python framework for orchestrating role-playing AI agents that collaborate on complex tasks. Define agents with roles, goals, and tools, then let them work together autonomously. 25,000+ stars.

## Install

Save as a script file and run:

## Quick Use

```bash
pip install crewai crewai-tools
```

```python
from crewai import Agent, Task, Crew

researcher = Agent(
    role="Senior Researcher",
    goal="Find the latest AI trends",
    backstory="Expert at analyzing tech trends",
    tools=[search_tool]
)

writer = Agent(
    role="Tech Writer",
    goal="Write engaging content about AI",
    backstory="Skilled at making complex topics accessible"
)

research_task = Task(
    description="Research the top 5 AI coding tools in 2026",
    agent=researcher
)

write_task = Task(
    description="Write a blog post based on the research",
    agent=writer
)

crew = Crew(agents=[researcher, writer], tasks=[research_task, write_task])
result = crew.kickoff()
```

---

## Intro

CrewAI is a Python framework for orchestrating role-playing AI agents that collaborate on complex tasks with 25,000+ GitHub stars. Define agents with specific roles, goals, backstories, and tools, then organize them into crews that work together autonomously — like assembling a virtual team where each member has specialized expertise. Best for developers building multi-agent systems for research, content creation, data analysis, and workflow automation. Works with: Claude, GPT-4, Gemini, Ollama, any LangChain-supported model. Setup time: under 3 minutes.

---

## Core Concepts

### Agents
Each agent has a role, goal, and optional tools:

```python
analyst = Agent(
    role="Data Analyst",
    goal="Analyze datasets and find actionable insights",
    backstory="10 years of experience in data science",
    tools=[csv_tool, sql_tool, chart_tool],
    llm="claude-sonnet-4-20250514"
)
```

### Tasks
Tasks define what needs to be done:

```python
analysis_task = Task(
    description="Analyze Q1 sales data and identify top-performing regions",
    expected_output="Report with charts and recommendations",
    agent=analyst,
    output_file="q1_analysis.md"
)
```

### Crews
Crews orchestrate agents and tasks:

```python
crew = Crew(
    agents=[analyst, writer, reviewer],
    tasks=[analysis_task, write_task, review_task],
    process=Process.sequential  # or Process.hierarchical
)
```

### Process Types

| Process | How It Works |
|---------|-------------|
| `sequential` | Tasks run one after another |
| `hierarchical` | Manager agent delegates to workers |

### Built-In Tools

```python
from crewai_tools import (
    SerperDevTool,      # Web search
    ScrapeWebsiteTool,  # Web scraping
    FileReadTool,       # Read files
    DirectoryReadTool,  # Read directories
    CodeInterpreterTool # Execute code
)
```

### Real-World Example: Content Pipeline

```python
# Research → Write → Edit → Publish
researcher = Agent(role="Researcher", tools=[search_tool])
writer = Agent(role="Writer")
editor = Agent(role="Editor")

crew = Crew(
    agents=[researcher, writer, editor],
    tasks=[
        Task(description="Research topic X", agent=researcher),
        Task(description="Write 1000-word article", agent=writer),
        Task(description="Edit for grammar and clarity", agent=editor),
    ],
    process=Process.sequential
)
```

### Key Stats
- 25,000+ GitHub stars
- Role-based agent design
- Sequential and hierarchical processes
- 20+ built-in tools
- Memory and caching support

### FAQ

**Q: What is CrewAI?**
A: CrewAI is a Python framework for building teams of AI agents that collaborate on complex tasks, each with specialized roles, goals, and tools.

**Q: Is CrewAI free?**
A: Yes, open-source under MIT license. Enterprise edition available.

**Q: How is CrewAI different from LangGraph?**
A: CrewAI uses a role-playing metaphor — you define agents with roles and goals. LangGraph uses a graph-based approach with nodes and edges. CrewAI is more intuitive for team-like workflows.

---

## Source & Thanks

> Created by [CrewAI](https://github.com/crewAIInc). Licensed under MIT.
>
> [crewAI](https://github.com/crewAIInc/crewAI) — ⭐ 25,000+

Thanks to the CrewAI team for making multi-agent orchestration accessible.

---

<!-- ZH -->

## 快速使用

```bash
pip install crewai crewai-tools
```

```python
from crewai import Agent, Task, Crew

researcher = Agent(role="研究员", goal="找到最新 AI 趋势")
writer = Agent(role="作者", goal="撰写引人入胜的内容")

crew = Crew(
    agents=[researcher, writer],
    tasks=[Task(description="研究 AI 工具趋势", agent=researcher),
           Task(description="写一篇博客文章", agent=writer)]
)
result = crew.kickoff()
```

---

## 简介

CrewAI 是一个用于编排角色扮演 AI Agent 的 Python 框架，GitHub 25,000+ stars。定义具有角色、目标和工具的 Agent，组织成团队自主协作完成复杂任务。适合构建研究、内容创作、数据分析等多 Agent 系统。

---

## 来源与感谢

> Created by [CrewAI](https://github.com/crewAIInc). Licensed under MIT.
>
> [crewAI](https://github.com/crewAIInc/crewAI) — ⭐ 25,000+

---
Source: https://tokrepo.com/en/workflows/97fce2da-d60f-413c-a31d-527cd7aaa558
Author: Agent Toolkit