metrics_skill

Overview

This skill defines metrics standards for observability environments to help teams instrument, validate, and maintain reliable telemetry. It focuses on industry-standard patterns, secure defaults, and practical guidance to get production-grade metrics in place quickly. The goal is clear, testable, and maintainable metrics that support monitoring, alerting, and performance analysis.

How this skill works

The skill inspects metric design, implementation patterns, and operational controls. It recommends architectures for metric collection, error handling, and performance considerations, and it provides checklists and templates to accelerate adoption. Implementers use the guidance to validate inputs, add observability hooks, and integrate testing and documentation for metrics pipelines.

When to use it

• When adding metrics to a new service or feature to ensure consistency
• When auditing existing telemetry for gaps, security, or performance issues
• When defining SLIs, SLOs, and alerting rules based on reliable metrics
• When onboarding teams to shared observability standards and templates
• When preparing metrics for production deployment and long-term maintenance

Best practices

• Choose metric types and labels that reflect cardinality and query needs
• Validate and sanitize inputs before recording metrics to avoid noise
• Instrument error handling and edge cases to improve signal accuracy
• Write unit and integration tests for metric emission and aggregation
• Document metric semantics, units, and intended use for consumers

Example use cases

• Standardizing counter, gauge, and histogram usage across microservices
• Auditing a metrics pipeline to remove high-cardinality labels
• Implementing SLI-based alerts using well-defined service metrics
• Creating templates for consistent metric naming, units, and tags
• Adding observability hooks to background jobs and batch processes