slo-alert-tuner

Tuning noisy SLO alerts requires evaluating burn-rate logic and thresholds against repo conventions. This skill inspects monitoring diffs and configs to surface minimal, high-leverage rule changes that reduce alert fatigue while maintaining actionability.

Burn-rate logic determines how quickly your error budget is consumed, directly impacting alert quality. Suboptimal burn-rate thresholds often trigger noisy alerts or miss real issues. Analyzing these rules against expected guarantees helps prioritize minimal changes for actionable signals.

The best way to validate SLO alert rule changes is to compare them against known risk areas and repo conventions. This skill delivers ranked recommendations with clear validation steps and follow-up edits to reduce risk and ensure alert actionability.

Yes, you can analyze monitoring diffs and source files to identify alert regressions. The skill inspects relevant diffs, configs, and artifacts before forming a conclusion, comparing the current state against expected guarantees to surface improvement opportunities.

No specific dependencies are required to inspect monitoring configurations. The skill operates directly on your codebase artifacts, evaluating SLO-based thresholds and alert rules against existing repo conventions without needing external modules.

You should not use automated alert tuning when your monitoring configurations lack clear expected guarantees or established repo conventions. Without these baselines, comparing SLO thresholds and burn-rate logic against known risk areas to generate actionable recommendations is ineffective.