Auto-instrument every Google Agent Development Kit run with zero code changes. Get full agent session traces, per-step tool call spans, and Gemini model costs attributed to each agent turn.
Free tier · No credit card · 2-minute setup
Full visibility into ADK agent runs, every step, tool call, and model response is a traced span, giving you a complete picture of what your agent did and how long it took.
Each ADK tool call is a child span with the function name, input arguments, response, and latency, essential for debugging agents that call many tools per turn.
Underlying Gemini API calls made by ADK agents are traced and costed at Google's published Gemini rates, no separate instrumentation needed.
import fluiq
fluiq.instrument(api_key="fl_...") # patches Google ADK automatically
from google.adk.agents import LlmAgent
from google.adk.runners import Runner
from google.adk.sessions import InMemorySessionService
def get_weather(city: str) -> dict:
return {"temperature": 22, "condition": "Sunny"}
agent = LlmAgent(
name="weather-agent",
model="gemini-1.5-pro",
instruction="You are a helpful weather assistant.",
tools=[get_weather],
)
session_service = InMemorySessionService()
runner = Runner(agent=agent, app_name="weather-app",
session_service=session_service)
# Agent run fully traced with tool spans:
response = runner.run(user_id="u1", session_id="s1",
new_message=types.Content(role="user",
parts=[types.Part(text="Weather in Tokyo?")]))Every call to these methods is automatically traced, no decorators, no wrappers, no manual spans.
Runner.run() / run_async()LlmAgent._run_async_impl()Agent tool function callsgoogle.adk.models.Gemini.generate_content()BaseSessionService.create_session()BaseMemoryService.search_memory()Free tier. No credit card. Full traces, security scanning, and evals on your first Google ADK call.
50,000 free traces / month · 1,000 evals / month · 14-day retention