​
Overview

Keywords AI parameters allow you to add contextual information to your traces, enabling better analytics, user tracking, and workflow understanding.

​
Available Keywords AI Parameters

ParameterTypeDescriptionExample
customer_identifierstringUnique customer ID"user_123"
customer_namestringCustomer display name"John Doe"
customer_emailstringCustomer email address"john@example.com"
metadataobjectCustom key-value pairs{"priority": "high"}
custom_identifierstringCustom operation ID"ticket_789"
thread_identifierstringConversation thread ID"thread_abc"
For the complete list of available parameters, see the Keywords AI API reference.

​
Compatibility

IntegrationSupportNotes
Keywords AI Native✅Full support for all Keywords AI params
OpenAI Agents SDK✅Not supported yet
Vercel AI SDK❌Not supported yet

​
Integration

Setup

Make sure you have everything ready before you start.

​
Implementation

​
Basic Keywords AI params usage

from keywordsai_tracing.decorators import workflow, task
from keywordsai_tracing.contexts.span import keywordsai_span_attributes
from openai import OpenAI

client = OpenAI()

@task(name="simple_task")
def simple_task():
    with keywordsai_span_attributes(
        keywordsai_params={
            "customer_params": {
                "customer_identifier": "user_123",
            },
            "metadata": {"task_type": "simple"}
        }
    ):
        response = client.chat.completions.create(
            model="gpt-3.5-turbo",
            messages=[{"role": "user", "content": "Hello!"}]
        )
        return response.choices[0].message.content
This workflow creates a trace hierarchy with Keywords AI params:
  • Parent workflow: customer_support_workflow with thread and session metadata
  • Individual tasks: user_query_processing, response_formatting with customer and operation metadata
  • Complete parameter tracking: customer info, custom identifiers, and contextual metadata
  • Enhanced analytics: filtering by customer, priority, department, and session

​
How to see this in the platform

Once you’ve implemented Keywords AI params in your traces, you can view and filter by them in the Keywords AI platform:

​
Accessing enriched traces

  1. Navigate to the Traces page in your Keywords AI dashboard
  2. You’ll see traces with additional metadata and customer information

​
Understanding the enhanced trace view

  • Customer tracking: Filter traces by customer identifier, name, or email
  • Custom metadata: Use metadata fields for advanced filtering and analytics
  • Thread grouping: View related conversations using thread identifiers
  • Custom identifiers: Track specific operations or tickets

​
Example trace with Keywords AI params

Traces with Keywords AI parameters showing customer and metadata information
The enhanced trace view shows:
  • Customer information in the trace details
  • Custom metadata for filtering and analysis
  • Thread and session grouping
  • Performance metrics with contextual data