Each trace can be enriched with Keywords AI params to better understand your AI products and workflows. Keywords AI params include:

  • customer_identifier, customer_name, customer_email
  • metadata
  • custom_identifier
  • thread_identifier
  • …

You can check the full list of Keywords AI params in the Keywords AI API reference.

​
Example code

This is an example of how to pass Keywords AI params to your traces in Python.

from keywordsai_tracing.contexts.span import keywordsai_span_attributes
from openai import OpenAI
from keywordsai_tracing import KeywordsAITelemetry
from keywordsai_tracing.decorators import workflow, task
import os
os.environ["KEYWORDSAI_BASE_URL"] = "https://api.keywordsai.co/api"
os.environ["KEYWORDSAI_API_KEY"] = os.getenv("KEYWORDSAI_API_KEY")

k_tl = KeywordsAITelemetry()
client = OpenAI()

@workflow(name="test_dynamic_attributes")
def test_dynamic_attributes():
    with keywordsai_span_attributes(
        keywordsai_params={
            "customer_identifier": "123",
            "customer_email": "test@test.com",
            "customer_name": "John Doe",
            "thread_identifier": "789",
            "custom_identifier": "101",
            "metadata": {"some_key": "some_value"},
        }
    ):
        response = client.chat.completions.create(
            model="gpt-4o",
            messages=[{"role": "user", "content": "Hello, world!"}],
        )
    return response

if __name__ == "__main__":
    test_dynamic_attributes()
QuickstartGroup traces