What problem does it solve? Running AI agents as cloud-hosted or continuously running services introduces operational risks that single CLI sessions never face: uncontrolled rollouts, missing audit trails, runaway retries, and unclear failure recovery. This Skill provides the operational controls needed to run agent workloads safely in production. ## Core Features & Use Cases - Runtime Lifecycle Management: Start, pause, stop, and restart agent workloads with immutable deployment artifacts and least-privilege credentials. - Observability & Metrics: Track success rate, mean retries per task, time to recovery, cost per successful task, and failure class distribution across logs, metrics, and traces. - Safety & Change Management: Enforce scopes, permissions, kill switches, timeout and retry budgets, plus rollout, rollback, and audit logging for high-risk actions. - Use Case: When failure rates spike on a production agent, follow the incident pattern: freeze new rollouts, capture representative traces, isolate the failing route, patch with the smallest safe change, run regression and security checks, then resume gradually. ## Quick Start Ask the agent to apply the enterprise-agent-ops baseline controls and incident response pattern to your continuously running agent deployment.
