Introduction
crewAI is an emerging, powerful framework specifically designed for orchestrating role-playing, autonomous AI agents. It simplifies the process of creating multi-agent setups.
Core Concepts in crewAI
- Agents: You define an agent by giving it a Role, a Goal, and a Backstory.
- Tasks: Specific assignments you give to the agents, including the expected output format.
- Tools: Capabilities (like web search or file reading) assigned to specific agents.
- Crew: The overarching container that manages the agents and tasks, dictating the process (e.g., sequential or hierarchical execution).
from crewai import Agent, Task, Crew
# Define an agent
researcher = Agent(
role='Senior Research Analyst',
goal='Uncover cutting-edge developments in AI',
backstory='You work at a leading tech think tank.',
verbose=True
)
# Define a task
task = Task(
description='Research the latest advancements in quantum computing.',
expected_output='A 3-paragraph executive summary.',
agent=researcher
)
# Form a crew
crew = Crew(agents=[researcher], tasks=[task])
result = crew.kickoff()Assignment
- Read the crewAI documentation.
- Install
crewaiand build a simple 2-agent crew: a Researcher who finds information on the web, and a Writer who turns that information into a blog post.