slo-design

What problem does it solve?

It prevents misleading service-level objectives by helping you define the right indicator, measurement window, and error-budget policy so the SLO reflects user experience and actually drives operational behavior.

Core Features & Use Cases

• SLI selection that matches user experience: choose the right ratio or latency percentile and clearly define what counts as a valid request.
• Percentile-based latency SLOs: replace average latency with p95/p99 reasoning to protect against tail-latency failures.
• Error budget and policy that changes behavior: specify what happens when budget is healthy, tight, or exhausted so teams don’t treat the SLO as theater.
• Burn-rate alert guidance: design fast-burn/slow-burn multi-window alerts so issues trigger during incidents, not after the fact.
• Common SLO anti-pattern avoidance: guard against average-latency SLOs, unrealistic targets, missing burn alerts, and missing enforcement.

Quick Start

Ask the AI to help you design an SLO by specifying what users experience, proposing an SLI (including valid-request boundaries), selecting an appropriate percentile and window, and defining a concrete error-budget policy plus burn-rate alert thresholds.