Summary

This skill provides high-level operational guidance for managing long-lived agent systems in production environments (PM2, systemd, containers). It documents lifecycle controls, observability patterns, safety scopes, and change management best practices with incident response playbooks. However, the skill is primarily conceptual documentation of principles rather than actionable instructions—it lacks concrete implementation details, code examples, specific commands, or step-by-step procedures an agent could execute.

Detected Capabilities

Trigger Keywords

Phrases that agents use to match this skill to user intent.

Use Cases

• Manage agent runtime lifecycle in production fleets
• Monitor and track agent logs, metrics, and traces
• Implement least-privilege access controls and kill switches
• Execute safe rollout and rollback procedures with audit trails
• Respond to production incidents with standardized playbooks
• Track operational metrics (success rate, cost, recovery time)

Quality Notes

• +: Clear high-level operational domains and principles documented
• +: Incident response pattern provided with structured steps
• +: Mentions important safety controls (immutable artifacts, least-privilege, timeouts, audit logs)
• -: No concrete implementation examples (bash commands, config files, API calls)
• -: No specific tool/platform integration details despite mentioning PM2, systemd, containers
• -: Lacks step-by-step procedures or decision trees for deployment gates
• -: No error handling guidance or failure mode documentation
• -: Missing deployment artifact versioning or secret management specifics
• -: No validation patterns for 'least-privilege credentials' or scope boundaries
• -: Metrics tracking listed but no guidance on instrumentation or tooling
• -: No supporting files (example configs, scripts, monitoring queries)