openinference-instrumentation-agno

What it is and what it does

This package adds automatic tracing to Agno Agent applications using the OpenTelemetry standard. When installed and initialized, it intercepts agent execution and emits structured trace spans that capture agent behavior, tool calls, and model interactions. Those spans can be exported to any OpenTelemetry-compatible backend—such as Arize Phoenix (a local-only observability UI), Arize AX, or Langfuse—for real-time monitoring and post-hoc analysis.

The instrumentation is designed to work with minimal code changes: a single call to `AgnoInstrumentor().instrument()` after setting up your tracer provider. It depends on the OpenTelemetry API and instrumentation framework, plus OpenInference semantic conventions, to ensure traces are structured consistently and interoperable across tools. The package targets Python 3.10 and later and is actively maintained.

Use it for:

• Monitor Agno agent execution in development to debug tool calls and model responses without manual logging.
• Export agent traces to Arize Phoenix for local, browser-based visualization of agent behavior and performance.
• Collect agent traces in production to Langfuse or Arize AX for analytics, debugging, and compliance auditing.
• Correlate agent traces with upstream application traces by integrating with existing OpenTelemetry infrastructure.
• Track agent session metadata and custom attributes for multi-turn conversations or batch agent runs.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Auto-instruments Agno Agent applications to emit OpenTelemetry-compatible traces for monitoring and observability, exportable to collectors like Arize Phoenix or Langfuse.

Yes. The package is actively maintained, has no known vulnerabilities, uses a permissive license, and integrates cleanly with standard OpenTelemetry tooling. Install it if you are building Agno agents and need observability; it adds minimal friction and works with existing monitoring infrastructure. Skip it only if you have no observability requirements or prefer manual logging.

Install

openinference-instrumentation-agno on PyPI

pip install openinference-instrumentation-agno

uv add openinference-instrumentation-agno

poetry add openinference-instrumentation-agno

Installing openinference-instrumentation-agno

Before you install

Low friction installation as a pure Python wheel. Active maintenance with a release 2 days old and steady upstream activity. Depends on established OpenTelemetry libraries and OpenInference conventions, all with permissive licensing.

License in practice

Apache-2.0 permissive license allows commercial and private use without restriction, with only attribution and license-notice requirements.

Quickstart

pip install openinference-instrumentation-agno

from openinference.instrumentation.agno import AgnoInstrumentor
from opentelemetry.sdk import trace as trace_sdk
from opentelemetry import trace as trace_api

tracer_provider = trace_sdk.TracerProvider()
trace_api.set_tracer_provider(tracer_provider)
AgnoInstrumentor().instrument()

Requires Python 3.10 or later; Agno agent framework must be installed separately; typically paired with an OpenTelemetry exporter and collector (e.g., Arize Phoenix).

Verify before relying

• Whether instrumentation overhead is negligible for production agent workloads.
• Support for custom span attributes and session tracking beyond the documented examples.
• Compatibility with Agno versions not explicitly tested in the fact sheet.

Package facts

Evidence: openinference_instrumentation_agno-1.0.3-py3-none-any.whl

Tags

More Monitoring packages