Your First Agent as a Service: Deploying a Support Agent on AgentCore

Description and use

A lightweight support agent deployed on AgentCore: deploy, invoke, observe, and delete in a single day.

Problem

A three‑person support desk needs a quick, trustworthy automation that can be deployed, invoked, observed, and deleted in a day.

Outcome

Deployment completed in 1 m 51 s, runtime ARN exposed, invoke latency ~2 s, trace visible in X‑Ray, and cleanup possible with a single command.

Workflow

• Verify toolchain and AWS environment with make doctor
• Run local development server with make dev to test invoke latency
• Deploy the runtime to AWS with make deploy
• Invoke the deployed agent via the provided endpoint
• Observe the OpenTelemetry trace in CloudWatch X‑Ray
• Delete the deployment with make delete

Evidence

The whole agent surface fits in one screen of main.py

Source

Deploy to AWS ..................... 1 m 27 s Total duration: 1 m 51 s

Source

make doctor ... 15 passed, 0 warned, 0 failed

Source