How to beginner · 3 min read

Fix AgentOps not tracking calls

Quick answer

To fix AgentOps not tracking calls, ensure you call agentops.init(api_key=os.environ["AGENTOPS_API_KEY"]) before any LLM client usage. This initializes automatic instrumentation. Also, verify your API key is correctly set in the environment and that you are using the latest agentops package.

PREREQUISITES

Python 3.8+
AgentOps API key
pip install agentops
OpenAI API key (if using OpenAI LLMs)

Setup

Install the agentops package and set your API keys as environment variables. This enables automatic tracking of LLM calls.

Install AgentOps SDK: pip install agentops
Set environment variables in your shell or .env file:

export AGENTOPS_API_KEY="your_agentops_api_key"
export OPENAI_API_KEY="your_openai_api_key"

bash

pip install agentops

output

Collecting agentops
  Downloading agentops-1.0.0-py3-none-any.whl (10 kB)
Installing collected packages: agentops
Successfully installed agentops-1.0.0

Step by step

Initialize agentops at the start of your application to enable automatic tracking. Then create and use your LLM client as usual.

python

import os
from openai import OpenAI
import agentops

# Initialize AgentOps for automatic tracking
agentops.init(api_key=os.environ["AGENTOPS_API_KEY"])

# Create OpenAI client
client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

# Make a chat completion call
response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello, AgentOps!"}]
)
print("Response:", response.choices[0].message.content)

# Optionally, manually start and end a session
session = agentops.start_session(tags=["example-session"])
# ... your code here ...
agentops.end_session("Success")

output

Response: Hello, AgentOps!
# (No errors, AgentOps tracks this call automatically)

Common variations

You can use agentops with other LLM providers by initializing before client creation. For async code, agentops.init() remains the same. To manually track sessions, use start_session() and end_session(). Ensure you do not create LLM clients before calling agentops.init().

python

import asyncio
import agentops
from openai import OpenAI

async def main():
    agentops.init(api_key=os.environ["AGENTOPS_API_KEY"])
    client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])
    response = await client.chat.completions.create(
        model="gpt-4o",
        messages=[{"role": "user", "content": "Async call with AgentOps"}]
    )
    print("Async response:", response.choices[0].message.content)

asyncio.run(main())

output

Async response: Async call with AgentOps
# (Tracked automatically by AgentOps)

Troubleshooting

If calls are not tracked: Confirm agentops.init() is called before any LLM client instantiation.
Check API key: Ensure AGENTOPS_API_KEY is set correctly in your environment.
Package version: Upgrade agentops to the latest version with pip install --upgrade agentops.
Conflicts: Avoid multiple initializations or creating clients before agentops.init().

✅

Key Takeaways

Always call agentops.init() before creating any LLM clients to enable automatic tracking.
Set your AGENTOPS_API_KEY environment variable correctly to authenticate tracking.
Use agentops.start_session() and agentops.end_session() for manual session control.
Keep the agentops package updated to avoid bugs affecting tracking.
Do not instantiate LLM clients before initializing agentops.

Verified 2026-04 · gpt-4o

Verify ↗