Strands agents

Monitor applications built with the Strands Agents SDK using Coralogix AI Observability. The Strands integration enriches the built-in OpenTelemetry tracing provided by the SDK with GenAI semantic convention attributes, so you can track prompt content, completions, tool calls, and user identity across every agent run.

Overview

Strands Agents SDK already includes built-in OpenTelemetry tracing. This instrumentation enriches those existing spans with additional GenAI semantic convention attributes, optimized to support large language model (LLM) application development with streamlined integration, detailed production tracing, and effective debugging capabilities.

Requirements

• Python 3.10–3.13.
• Coralogix API keys.

Installation

Run the following command.

pip install "llm-tracekit-strands"

Authentication

Authentication data is passed during OTel Span Exporter definition:

1. Choose the ingress.:443 endpoint that corresponds to your Coralogix domain using the domain selector at the top of the page.
2. Use your customized API key in the authorization request header.
3. Provide the application and subsystem names.

from llm_tracekit.strands import setup_export_to_coralogix

setup_export_to_coralogix(
    coralogix_token=<your_coralogix_token>,
    coralogix_endpoint="ingress.:443",
    service_name="ai-service",
    application_name="ai-application",
    subsystem_name="ai-subsystem",
    capture_content=True,
)

Usage

This section describes how to set up instrumentation for Strands Agents.

Set up tracing

Automatic

Use the setup_export_to_coralogix function to set up tracing and export traces to Coralogix. See the code snippet in the Authentication section.

Manual

Alternatively, you can set up tracing manually.

from opentelemetry import trace
from opentelemetry.exporter.otlp.proto.grpc.trace_exporter import OTLPSpanExporter
from opentelemetry.sdk.resources import SERVICE_NAME, Resource
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.sdk.trace.export import SimpleSpanProcessor

tracer_provider = TracerProvider(
    resource=Resource.create({SERVICE_NAME: "ai-service"}),
)
exporter = OTLPSpanExporter()
span_processor = SimpleSpanProcessor(exporter)
tracer_provider.add_span_processor(span_processor)
trace.set_tracer_provider(tracer_provider)

Instrument

To instrument Strands, call the instrument method.

from llm_tracekit.strands import StrandsInstrumentor

StrandsInstrumentor().instrument()

Uninstrument

To uninstrument clients, call the uninstrument method.

StrandsInstrumentor().uninstrument()

Full example

from llm_tracekit.strands import StrandsInstrumentor, setup_export_to_coralogix
from strands import Agent

# Optional: Configure sending spans to Coralogix
# Reads Coralogix connection details from the following environment variables:
# - CX_TOKEN
# - CX_ENDPOINT
setup_export_to_coralogix(
    service_name="ai-service",
    application_name="ai-application",
    subsystem_name="ai-subsystem",
    capture_content=True,
)

# Activate instrumentation
StrandsInstrumentor().instrument()

# Strands usage example
agent = Agent(system_prompt="You are a helpful assistant.")
response = agent("Write a short poem on open telemetry.")

Enable message content capture

By default, message content — prompt contents, completions, function arguments, and return values — is not captured. To capture message content as span attributes:

• Pass capture_content=True when calling setup_export_to_coralogix.
• Set the environment variable OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT to true.

Most Coralogix AI evaluations require message contents to function properly, so enabling message capture is strongly recommended.

Pass user identity

To capture the end-user identifier, pass the user parameter in your model's params configuration:

from strands.models.openai import OpenAIModel

model = OpenAIModel(
    model_id="gpt-4o",
    params={"user": "[email protected]"},
)

agent = Agent(model=model)

Semantic conventions

Next steps

Once your integration is set up, explore the AI Center Overview to monitor performance, costs, quality issues, and security across all your AI applications — and to set up Guardrails for real-time policy enforcement.