Skip to main content
Automatically capture agent runs, LLM calls, and tool invocations within the Pydantic AI framework via its native OpenTelemetry instrumentation.

​
Setup

import traceroot
from traceroot import Integration

traceroot.initialize(integrations=[Integration.PYDANTIC_AI])

​
Usage

Once initialized, all Pydantic AI agent runs and tool invocations are captured automatically: 
import traceroot
from traceroot import Integration
from pydantic_ai import Agent

traceroot.initialize(integrations=[Integration.PYDANTIC_AI])

agent = Agent(
    "openai:gpt-4o-mini",
    system_prompt="You are a helpful assistant.",
)

result = agent.run_sync("What is the capital of France?")
print(result.output)

traceroot.flush()

​
What Gets Captured

AttributeDescription
Agent runsEach agent.run() / agent.run_sync() / agent.run_stream() invocation
Tool callsEach tool invocation with input arguments and results
LLM callsRaw completion requests with input messages, output messages, and model name
Tokens & CostInput and output token counts per LLM call; cost calculated from model pricing
Cache tokensCache read and write token counts when reported by the provider
LatencyDuration per agent run and per span

​
Run the example

Clone the repo and run a complete agent end-to-end.

Python

Run the Python example