OpenInference Strands Agents Instrumentation

Python instrumentation library for Strands Agents.

This package provides a span processor that transforms Strands' native OpenTelemetry spans to OpenInference format. The traces are fully OpenTelemetry compatible and can be sent to an OpenTelemetry collector for viewing, such as arize-phoenix.

Installation

pip install openinference-instrumentation-strands-agents

Quickstart

In this example we will instrument a small Strands agent program and observe the traces via arize-phoenix.

Install packages.

pip install openinference-instrumentation-strands-agents strands-agents arize-phoenix opentelemetry-sdk opentelemetry-exporter-otlp

Start the phoenix server so that it is ready to collect traces. The Phoenix server runs entirely on your machine and does not send data over the internet.

python -m phoenix.server.main serve

In a python file, setup the Strands telemetry and add the StrandsAgentsToOpenInferenceProcessor to transform spans.

import os

from strands import Agent, tool
from strands.models.openai import OpenAIModel
from strands.telemetry import StrandsTelemetry

from openinference.instrumentation.strands_agents import StrandsAgentsToOpenInferenceProcessor

# Setup Strands native telemetry
telemetry = StrandsTelemetry()
telemetry.setup_otlp_exporter(endpoint="http://127.0.0.1:6006/v1/traces")

# Add OpenInference processor to transform spans
telemetry.tracer_provider.add_span_processor(StrandsAgentsToOpenInferenceProcessor())


@tool
def get_weather(city: str) -> dict:
    """Get the current weather for a city.

    Args:
        city: The name of the city
    """
    return {
        "status": "success",
        "content": [{"text": f"The weather in {city} is sunny and 72°F."}],
    }


if __name__ == "__main__":
    # Create and run agent
    model = OpenAIModel(model_id="gpt-4o-mini")
    agent = Agent(
        name="WeatherAgent",
        model=model,
        tools=[get_weather],
        system_prompt="You are a helpful weather assistant.",
    )

    result = agent("What's the weather in San Francisco?")
    print(result)

Since we are using OpenAI, we must set the OPENAI_API_KEY environment variable to authenticate with the OpenAI API.

export OPENAI_API_KEY=your-api-key

Now simply run the python file and observe the traces in Phoenix.

python your_file.py

Important: Processor Ordering

The StrandsAgentsToOpenInferenceProcessor mutates spans in-place. This means the order in which you add span processors matters.

Add the StrandsAgentsToOpenInferenceProcessor before any exporters that should receive the transformed OpenInference spans:

from opentelemetry.sdk.trace.export import BatchSpanProcessor
from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter

# Correct order: processor first, then exporter
telemetry.tracer_provider.add_span_processor(StrandsAgentsToOpenInferenceProcessor())
telemetry.tracer_provider.add_span_processor(BatchSpanProcessor(OTLPSpanExporter()))

If you need to export both the original GenAI spans and transformed OpenInference spans to different destinations, you'll need to set up the processors carefully or consider using separate tracer providers.

More Info

• More info on OpenInference and Phoenix
• How to customize spans to track sessions, metadata, etc.
• How to account for private information and span payload customization