CrewAI

AgentOps and CrewAI teamed up to make monitoring Crew agents dead simple.

Crew has comprehensive documentation available as well as a great quickstart.

​

Adding AgentOps to Crew agents

1

Install the AgentOps SDK

pip

poetry

pip install agentops

Give us a star on GitHub while you’re at it (you may be our 2,000th 😊)

2

Install Crew from the AgentOps fork

At this time, Crew with AgentOps integration is in pre-release. In the interim, an official fork has been made available. After official support is released, this fork will be deprecated and developers will be encouraged to use the core Crew SDK.

pip

poetry

pip install git+https://github.com/AgentOps-AI/crewAI.git@main

3

Add 3 lines of code

1. Before calling the Crew() constructor in your code, call agentops.init()
2. At the end of your Crew run, call agentops.end_session("Success")

python

import agentops

# Beginning of program (i.e. main.py, __init__.py)
# IMPORTANT: Must be before calling the `Crew()` constructor
agentops.init(<INSERT YOUR API KEY HERE>)
...
# End of program (e.g. main.py)
agentops.end_session("Success") # Success|Fail|Indeterminate

Instantiating the AgentOps client will automatically instrument Crew, meaning you will be able to see all of your sessions on the AgentOps Dashboard along with the full LLM chat histories, cost, token counts, etc.

For more features see our Usage section.

4

Set your API key

Retrieve an API Key from your Settings > Projects & API Keys page.

Settings > Projects & API Keys

API keys are tied to individual projects.
A Default Project has been created for you, so just click Copy API Key

Set this API Key in your environment variables

.env

AGENTOPS_API_KEY=<YOUR API KEY>

5

Run your crew

Execute your program and visit app.agentops.ai/drilldown to observe your Crew! 🕵️

After your run, AgentOps prints a clickable url to console linking directly to your session in the Dashboard

Clickable link to session

​

Special Considerations with Crew

The Crew framework is capable of determining when all tasks have been accomplished and to halt execution. AgentOps will automatically end your active session when this determination is made. If you don’t want your AgentOps session to end at this time, add an optional parameter to your agentops.init() call.

agentops.init(skip_auto_end_session=True)

​

Crew + AgentOps Examples