it-operations_skill

Overview

This skill manages IT infrastructure operations to ensure service reliability, observability, and rapid incident response. It provides frameworks for monitoring, alerting, automation, capacity planning, change control, and ITIL-aligned service management. The focus is on practical workflows, measurable SLIs/SLOs, and continuous operational improvement.

How this skill works

The skill inspects service health via defined SLIs and collects metrics, logs, and traces to feed dashboards and alerting rules. It enforces incident workflows: severity classification, on-call escalation, diagnosis, mitigation, and post-incident reviews with tracked action items. Change and capacity processes evaluate risk, schedule windows, and validate outcomes. Automation patterns convert runbooks into reproducible scripts and IaC to reduce toil and improve consistency.

When to use it

• Design or validate monitoring and alerting for a new or existing service
• Establish or refine incident response, on-call rotations, and post-mortem practices
• Implement Infrastructure as Code, runbook automation, or self-healing remediation
• Create or review change management and risk assessment processes
• Build capacity forecasts, cost optimization plans, or runbook coverage audits

Best practices

• Define SLIs/SLOs tied to business impact and limit noisy alerts to actionable pages
• Automate repetitive tasks using IaC and runbook automation to keep toil < 30%
• Maintain blameless post-mortems with tracked action items and measurable follow-up
• Keep runbooks current: service overview, troubleshooting steps, rollback plans, and dashboard links
• Run regular alert tuning reviews and measure MTTA, MTTR, false positive rate, and on-call load

Example use cases

• Onboard a new microservice: set SLIs, dashboards, alerts, runbook, and deployment guardrails
• Respond to a P1 outage: follow severity matrix, engage responders, runbook-guided mitigation, post-incident review
• Migrate monitoring to a new platform: select tools by cost, metrics, logs, traces, and deployment model
• Reduce alert fatigue: categorize alerts, aggregate related signals, add runbook context and throttle noise
• Implement change governance: risk-scoring, CAB reviews, rollback plans, and post-change validation