Google ADK

Google ADK (Agent Development Kit) is Google’s framework for building AI agents powered by Gemini models. It provides native OpenTelemetry support for tracing agent runs.

Prerequisites

• SideSeat running locally (sideseat)
• Python 3.9+
• Google Cloud credentials configured

Quick Start

1. Start SideSeat

   npx sideseat

2. Install dependencies

   pip

   pip install google-adk sideseat

   uv

   uv add google-adk sideseat

3. Add telemetry

   from sideseat import SideSeat, Frameworks
   from google.adk.agents import Agent
   from google.adk.runners import Runner
   
   SideSeat(framework=Frameworks.GoogleADK)
   
   agent = Agent(
       model="gemini-2.5-flash",
       name="assistant",
       instruction="You are a helpful assistant."
   )

4. View runs

   Open http://localhost:5388 to see your runs.

Without SideSeat SDK

1. Start SideSeat

   npx sideseat

2. Set the endpoint

   export OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:5388/otel/default

3. Install dependencies

   pip

   pip install google-adk opentelemetry-sdk opentelemetry-exporter-otlp

   uv

   uv add google-adk opentelemetry-sdk opentelemetry-exporter-otlp

4. Add telemetry

   from opentelemetry import trace
   from opentelemetry.sdk.trace import TracerProvider
   from opentelemetry.sdk.trace.export import BatchSpanProcessor
   from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter
   
   provider = TracerProvider()
   provider.add_span_processor(BatchSpanProcessor(OTLPSpanExporter()))
   trace.set_tracer_provider(provider)

5. View runs

   Open http://localhost:5388 to see your runs.

Tools

Define tools for your agent:

from google.adk.agents import Agent

def get_weather(city: str) -> dict:
    """Get weather for a city."""
    return {"city": city, "forecast": "sunny"}

agent = Agent(
    model="gemini-2.5-flash",
    name="weather_assistant",
    instruction="Use the get_weather tool to provide forecasts.",
    tools=[get_weather]
)

SideSeat shows:

• Parent span for the agent run
• Child spans for each Gemini call
• Child spans for each tool execution

Sessions

ADK supports session management:

import asyncio
from google.adk.apps import App
from google.adk.runners import Runner
from google.adk.sessions import InMemorySessionService
from google.genai import types

app = App(name="my_app", root_agent=agent)
session_service = InMemorySessionService()

async def run():
    session = await session_service.create_session(
        app_name="my_app",
        user_id="user-123",
        session_id="session-456"
    )

    runner = Runner(app=app, session_service=session_service)

    message = types.Content(
        role="user",
        parts=[types.Part(text="Hello!")]
    )

    async for event in runner.run_async(
        session_id=session.id,
        user_id="user-123",
        new_message=message
    ):
        if event.content and event.content.parts:
            for part in event.content.parts:
                if hasattr(part, "text") and part.text:
                    print(part.text)

asyncio.run(run())

Extracted Attributes

Attribute	Description
gen_ai.system	google_adk
gen_ai.request.model	Gemini model ID
gen_ai.usage.input_tokens	Input token count
gen_ai.usage.output_tokens	Output token count

Next Steps

• Google Gemini — provider details
• Python SDK — SDK reference
• Core Concepts — understanding runs, steps, and messages