Summary

This skill provides operational guidance for long-lived, cloud-hosted agent workloads, emphasizing lifecycle management, observability, and safety controls. It outlines deployment patterns, incident response procedures, and best practices for running agents at enterprise scale, but lacks concrete implementation examples and tactical guidance for executing these operations.

Detected Capabilities

Trigger Keywords

Phrases that MCP clients use to match this skill to user intent.

Risk Signals

Use Cases

• Monitor and manage continuously running agent systems in production
• Implement lifecycle controls (start, pause, stop, restart) for long-lived workloads
• Establish observability practices (logs, metrics, traces) for agent operations
• Execute incident response and rollback procedures when agent failures occur
• Integrate agents with PM2, systemd, or container orchestration platforms
• Track operational metrics to optimize agent performance and cost efficiency

Quality Notes

• Skill provides high-level operational framework but lacks concrete implementation guidance — an agent cannot execute these instructions without supplementary documentation or code examples
• Good conceptual organization: domains are clearly named (lifecycle, observability, safety, change management) and incident response is logically structured
• Metrics section is well-chosen for operational assessment but lacks guidance on how to instrument or report these metrics
• Missing tactical examples: no sample PM2 configs, systemd service definitions, or code snippets for timeout/retry budgets
• No error handling patterns or edge cases documented (e.g., what happens if a freeze command fails, how to verify rollback success)
• Security baseline is well-intentioned but underdeveloped: terms like 'immutable deployment artifacts' and 'hard timeout' are not explained in sufficient detail for implementation