---
id: openinference-instrumentation-mcp
version: "2.0.5"
license: Apache-2.0
license_treatment: permissive
maintenance: active
---

# openinference-instrumentation-mcp — OpenInference MCP Instrumentation

License: permissive · Maintenance: active · Downloads: 193.0K/mo

## What it is and what it does

This package adds OpenTelemetry instrumentation to the MCP Python SDK, enabling distributed tracing by propagating trace context from the caller into MCP tool executions. It integrates with the OpenInference framework and observability platforms like Arize Phoenix and Arize AX. Currently, it focuses on context propagation rather than generating new telemetry data itself—the primary use case is connecting existing spans across MCP boundaries so that tool calls and their execution can be viewed as part of a single trace.

The package depends on openinference-instrumentation, opentelemetry-api, opentelemetry-instrumentation, and wrapt. It is marked as Production/Stable and actively maintained, with support for modern Python versions.

Use it for:

- Connect traces from an LLM application into MCP tool executions for end-to-end observability.
- Debug MCP tool call latency and failures by viewing them in the context of the parent request trace.
- Correlate MCP tool behavior with upstream application logic in a distributed tracing platform.
- Monitor MCP tool call patterns and performance metrics through OpenTelemetry-compatible backends.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Auto-instruments MCP tool calls to propagate OpenTelemetry context between the caller and the tool execution, enabling distributed tracing across MCP interactions.

Yes, if you are already using MCP with Python and need distributed tracing or observability. The low install friction, active maintenance, permissive license, and zero known vulnerabilities make it a safe addition. Install it only if you have an OpenTelemetry collector or observability platform in place to consume the context propagation it enables.

## Install

pip install openinference-instrumentation-mcp
uv add openinference-instrumentation-mcp
poetry add openinference-instrumentation-mcp

## Installing openinference-instrumentation-mcp

Before you install: Low friction installation with a pure-Python wheel. Actively maintained as of 2026-08-14 with recent release history; no known vulnerabilities.

License in practice: Apache-2.0 permissive license allows use in commercial and proprietary projects with minimal restrictions.

Quickstart:

pip install openinference-instrumentation-mcp

from openinference.instrumentation.mcp import MCP_INSTRUMENTATION
from opentelemetry import trace

# Instrumentation enables context propagation for MCP tool calls
MCP_INSTRUMENTATION.instrument()

Requires Python 3.10 or later (supports up to 3.14); OpenTelemetry API and instrumentation packages must be installed.

Verify before relying:

- Whether context propagation works with all MCP SDK versions or has version constraints beyond the package metadata.
- Performance overhead of instrumentation in high-throughput MCP tool call scenarios.
- Compatibility with non-OpenTelemetry tracing backends or observability platforms.

## Package facts

- License: Apache-2.0 (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 193.0K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

mcp instrumentation, opentelemetry mcp, distributed tracing mcp, context propagation mcp, mcp observability, openinference mcp, observability, distributed-tracing, mcp

[View on SkillFed](https://skillfed.io/packages/openinference-instrumentation-mcp) · [View on PyPI](https://pypi.org/project/openinference-instrumentation-mcp/)