---
id: opentelemetry-instrumentation-crewai
version: "0.62.3"
license: Apache-2.0
license_treatment: permissive
maintenance: active
---

# opentelemetry-instrumentation-crewai — OpenTelemetry crewAI instrumentation

License: permissive · Maintenance: active · Downloads: 6.9M/mo

## What it is and what it does

This package integrates OpenTelemetry tracing into crewAI agent workflows, automatically capturing LLM calls, agent decisions, and task execution as distributed traces. It wraps crewAI components to emit span events that record prompts, completions, and embeddings by default, giving visibility into how agents behave and make decisions.

The instrumentation is enabled with a single call to CrewAIInstrumentor().instrument() and respects the TRACELOOP_TRACE_CONTENT environment variable to disable sensitive content logging for privacy. It depends on the OpenTelemetry API, instrumentation framework, and AI-specific semantic conventions to standardize how agent behavior is represented in traces.

Use it for:

- Debug crewAI agent behavior by examining LLM prompts and responses in distributed traces.
- Monitor production agent workflows to detect performance bottlenecks or unexpected LLM outputs.
- Audit agent decisions and data flow for compliance by logging prompts and completions to traces.
- Correlate agent activity with other system components using OpenTelemetry's trace context propagation.
- Evaluate agent output quality by analyzing completions and embeddings captured in spans.

## Worth the install?

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

Adds OpenTelemetry tracing to crewAI agentic workflows, capturing LLM interactions and agent behavior as spans with optional content logging.

Yes, if you are running crewAI agents in production or development and need observability. The low install friction, active maintenance, permissive license, and zero known vulnerabilities make it a safe choice. Install only if you have an OpenTelemetry collector or exporter configured to receive traces; otherwise traces will be generated but discarded.

## Install

pip install opentelemetry-instrumentation-crewai
uv add opentelemetry-instrumentation-crewai
poetry add opentelemetry-instrumentation-crewai

## Installing opentelemetry-instrumentation-crewai

Before you install: Low friction install with four stable OpenTelemetry dependencies. Package is actively maintained with recent releases and no known vulnerabilities.

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

Quickstart:

pip install opentelemetry-instrumentation-crewai

from opentelemetry.instrumentation.crewai import CrewAIInstrumentor

CrewAIInstrumentor().instrument()

Requires Python 3.10 or later. Requires a running OpenTelemetry collector or exporter to receive traces.

Verify before relying:

- Whether the package works with all crewAI versions or has specific compatibility constraints.
- Performance overhead of instrumentation on large-scale agent deployments.
- Whether custom span processors or exporters are easily configurable beyond environment variables.

## Package facts

- License: Apache-2.0 (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 6.9M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

opentelemetry crewai tracing, agentic workflow instrumentation, llm agent observability, crewai span tracing, agent behavior monitoring, opentelemetry ai instrumentation, crewai debugging traces, observability, agent-tracing, llm-instrumentation

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