amazon-bedrock-agentcore-samples/03-integrations/observability/honeycomb

AgentCore + Honeycomb observability

Deploy a Strands travel agent to AgentCore runtime with traces sent to Honeycomb via OTLP HTTP.

Architecture

AgentCore runtime → travel_agent.py
  └── Custom OTelTracerProvider + BedrockInstrumentor
        └── OTLPSpanExporter → https://api.honeycomb.io/v1/traces
              headers: x-honeycomb-team, x-honeycomb-dataset
                └── Honeycomb → your dataset → Traces

DISABLE_ADOT_OBSERVABILITY=true bypasses the default CloudWatch ADOT pipeline. OTEL_SEMCONV_STABILITY_OPT_IN=gen_ai_latest_experimental enables GenAI semantic conventions. BedrockInstrumentor(capture_content=True) adds Bedrock-specific span attributes.

Prerequisites

• Python 3.10+, uv
• AWS credentials configured
• Honeycomb account with API key

Quick Start

pip install bedrock-agentcore boto3 python-dotenv
cp .env.example .env
# Edit .env: set HONEYCOMB_API_KEY
python deploy.py
python invoke.py
# View traces: https://ui.honeycomb.io → your dataset → Traces
python cleanup.py

Files

File	Description
utils/travel_agent.py	Agent with Honeycomb OTel TracerProvider + BedrockInstrumentor
deploy.py	Deploys to AgentCore runtime with Honeycomb env vars
invoke.py	Invokes the deployed agent
cleanup.py	Deletes all created AWS resources

Additional Resources

• Honeycomb LLM Documentation
• AgentCore observability