Extending the solution

The solution has two extension points, which serve different purposes:

1. Extending detection (what conditions are caught)

• Event-driven path: the webhook bridge Lambda drops Info-level events and forwards all Warn and Error level HyperPod events to DevOps Agent. This typically does not need modification - it already catches all actionable events.
• Polling-based path: the periodic-audit Lambda checks Kubernetes state. To detect additional conditions (e.g. GPU allocation below a threshold or specific Pod labels stuck in error states), add that logic to the Lambda code.

2. Extending reasoning (how the agent investigates and classifies)

Edit the plain-English skill definitions. For example, you can teach the RCA skill new classification rules, add domain-specific context about your workload's expected behavior, or adjust the recurrence thresholds.

Detection is code; reasoning is natural language. Both are in the repo and designed to be customized independently. See IMPLEMENTATION.md in the source repo for the full skill-authoring guide.

DevOps Agent APIs used by this solution

For readers interested in the programmatic integration, here are the key DevOps Agent APIs this solution calls:

Component	API	Purpose
Webhook provisioner (deployment)	register_service	Register the generic webhook service with DevOps Agent
Webhook provisioner (deployment)	associate_service	Associate the webhook with the Agent Space
Skill uploader (deployment)	list_assets	Check if a skill already exists
Skill uploader (deployment)	create_asset / update_asset	Upload or update the triage and RCA skill definitions
Email notifier (runtime)	get_backlog_task	Retrieve task metadata (title, priority, timestamps)
Email notifier (runtime)	list_journal_records	Retrieve findings, symptoms, and gaps from the investigation journal
Teardown	disassociate_service / deregister_service / delete_asset	Clean up on stack deletion

What's next

From here you can customize the solution for your environment:

• Adjust the CloudFormation parameters - tune the periodic-audit schedule, CrashLoopBackOff thresholds, NotReady node percentages, namespace filtering, and email recipients - all without code changes. See the parameter reference.
• Customize the verdict thresholds - the recovery time budgets and recurrence thresholds live in the RCA skill (sourced from the HyperPod mental model bundled with the skill). Edit and redeploy.
• Extend detection - modify the periodic-audit Lambda to check for additional Kubernetes conditions specific to your workloads (e.g. GPU allocation below a threshold, specific Pod labels stuck in error states).
• Extend reasoning - edit the triage or RCA skill definitions to adjust classification rules, add domain context about your expected cluster behavior, or tune the recurrence thresholds.
• Add notification channels - connect Slack, ServiceNow, PagerDuty, or Microsoft Teams via DevOps Agent's built-in integrations, or via a sibling EventBridge rule on the same aws.aidevops event stream that the email notifier uses.

The skills are plain English. Iterate on them the same way you'd iterate on a runbook.