Welcome to the least informative post of the whole series! 😁 Just a small heads-up. The really complex topic is for tomorrow.
ℹ️ This post is part of the “Crew AI Caveats” series, which I create to fill in the gaps left by official courses and to help you master CrewAI faster and easier.
These two commands create entry points that are structured very differently:
crewai create crew <name>
crewai create flow <name>
create crew
version main.py
:
def run():
...
def train():
...
def replay():
...
def test():
...
# No `if __name__ == "__main__"` statement at all
create flow
version main.py
:
def kickoff():
poem_flow = PoemFlow()
poem_flow.kickoff()
def plot():
poem_flow = PoemFlow()
poem_flow.plot()
if __name__ == "__main__":
kickoff()
The difference is that the flow version is structured as a script entry point and the crew version is nonfunctional without the crewai run
command.
Additionally, the crew
version exports some interesting entrypoints, while the flow
version does not. Not sure why—maybe because they design that the crews should be trained independently.
Stay tuned
In the next post: CrewAI hurts TDD fans! Let’s invent a cure.
Top comments (0)