opentelemetry-instrumentation-bedrock

What it is and what it does

This package integrates OpenTelemetry tracing into AWS Bedrock API calls made via Boto3, automatically capturing request and response data as distributed trace spans. It lets you observe what prompts and completions flow through your LLM application, making it easier to debug issues and evaluate output quality.

By default, the instrumentation logs full prompt and completion text to span attributes, which provides visibility but may expose sensitive user data. You can disable this logging by setting the TRACELOOP_TRACE_CONTENT environment variable to false to reduce trace size or protect privacy. The package depends on the OpenTelemetry API, instrumentation framework, and semantic conventions libraries.

Use it for:

• Debug LLM application behavior by inspecting actual prompts and completions sent to Bedrock models.
• Monitor Bedrock API usage patterns and latency across your distributed system using OpenTelemetry exporters.
• Evaluate output quality by correlating trace data with application metrics and user feedback.
• Audit sensitive data flow in production by selectively disabling content logging while keeping call metadata.
• Integrate Bedrock observability into existing OpenTelemetry deployments without additional instrumentation code.

Worth the install?

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

Adds distributed tracing to AWS Bedrock API calls made through Boto3, capturing prompts, completions, and embeddings as span attributes for observability and debugging.

Yes, if you use Bedrock with Boto3 and have an OpenTelemetry deployment in place. The package is actively maintained, has no known vulnerabilities, and low install friction. Be aware that it logs sensitive prompt/completion data by default—set TRACELOOP_TRACE_CONTENT=false if privacy is a concern.

Install

opentelemetry-instrumentation-bedrock on PyPI

pip install opentelemetry-instrumentation-bedrock

uv add opentelemetry-instrumentation-bedrock

poetry add opentelemetry-instrumentation-bedrock

Installing opentelemetry-instrumentation-bedrock

Before you install

Low friction install with four runtime dependencies, all from the OpenTelemetry ecosystem. Active maintenance with a recent release (4 days old) and a well-established repository.

License in practice

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

Quickstart

pip install opentelemetry-instrumentation-bedrock

from opentelemetry.instrumentation.bedrock import BedrockInstrumentor

BedrockInstrumentor().instrument()

Requires Python 3.10 or later; Boto3 and an OpenTelemetry exporter must be configured separately to send traces.

Verify before relying

• Whether the instrumentation works with all Bedrock model types or only a subset
• Performance overhead of tracing on high-volume Bedrock API calls
• Compatibility with different OpenTelemetry exporters and backends

Package facts

Evidence: opentelemetry_instrumentation_bedrock-0.62.3-py3-none-any.whl

Tags

More Monitoring packages