sre-reliability skill

Overview

This skill is an SRE Reliability Agent focused on improving service availability and operational resilience. It helps define and manage SLOs/SLIs, track error budgets, and guide performance and reliability improvements. The agent produces clear artifacts for monitoring and SLO governance to support engineering and ops teams.

How this skill works

The agent inspects service telemetry, SLI measurements, and incident history to recommend SLO targets and error budget policies. It synthesizes monitoring gaps, proposes alerts and dashboards, and suggests remediation or capacity changes to reduce downtime. Outputs include SLO definitions, monitoring configurations, and prioritized reliability actions.

When to use it

• When launching or revising service-level objectives (SLOs/SLIs)
• After repeated incidents or when error budgets are near exhaustion
• During capacity planning or before major releases
• To establish or improve monitoring and observability practices
• When you need objective measures for reliability and operational risk

Best practices

• Start with a small set of meaningful SLIs tied to user experience rather than internal metrics
• Define clear, measurable SLO targets and document error budget policies for risk-based decision making
• Automate SLI collection and alerting to ensure objective, timely signals
• Use post-incident analysis to refine SLOs and remediation runbooks
• Prioritize corrective work based on error budget depletion and customer impact

Example use cases

• Create SLOs for latency and availability for a customer-facing API and store definitions under docs/slo/
• Monitor request error rates and latency trends, and update monitoring assets under monitoring/
• Recommend throttling or rollback policies when error budget burn rate is high
• Drive a performance engineering plan to reduce tail latency and improve capacity headroom
• Report periodic reliability metrics to product and engineering leadership using SLO-backed dashboards

FAQ