respan-instrumentation-strands-agents 0.1.0

Respan instrumentation plugin for Strands Agents

respan-instrumentation-strands-agents

Respan instrumentation plugin for Strands Agents.

This package consumes Strands Agents' native OpenTelemetry spans and maps them directly into the Respan span contract used by the OTLP pipeline. It does not require OpenInference at runtime.

Install

pip install respan-instrumentation-strands-agents

Quickstart

import os

from respan import Respan
from respan_instrumentation_strands_agents import StrandsAgentsInstrumentor
from strands import Agent, tool
from strands.models.openai import OpenAIModel

respan_api_key = os.environ["RESPAN_API_KEY"]
respan_base_url = os.getenv("RESPAN_BASE_URL", "https://api.respan.ai/api")

respan = Respan(
    api_key=respan_api_key,
    base_url=respan_base_url,
    instrumentations=[StrandsAgentsInstrumentor()],
)

model = OpenAIModel(
    model_id="gpt-4o-mini",
    client_args={"api_key": respan_api_key, "base_url": respan_base_url},
)


@tool
def get_weather(city: str) -> str:
    """Get the current weather for a city."""
    return f"The weather in {city} is sunny and 72F."


agent = Agent(
    name="WeatherAgent",
    model=model,
    tools=[get_weather],
    system_prompt="You are a concise weather assistant.",
)

result = agent("What is the weather in Seattle?")
print(result)

respan.flush()

Notes

• Initialize Respan(...) before running the agent so Strands uses the active Respan OpenTelemetry provider.
• The instrumentor can refresh an already-created Strands tracer singleton when possible, which helps when an agent was constructed before activation.
• Tool definitions are enabled by default via Strands' gen_ai_tool_definitions semantic-convention opt-in and are exported as canonical llm.request.functions attributes.