// "Help craft efficient Mimir/Prometheus queries, troubleshoot metric issues, avoid high-cardinality problems, and recommend best practices for aggregation, recording rules, and performance."
| name | mimir-prometheus-troubleshoot |
| description | Help craft efficient Mimir/Prometheus queries, troubleshoot metric issues, avoid high-cardinality problems, and recommend best practices for aggregation, recording rules, and performance. |
Use this skill whenever a user needs help with:
Use labels such as:
job, instance, service, cluster, namespace, envAvoid:
user_id, session_id, request_id, raw UUIDsPrefer "5m", "15m", "1h".
rate() for counterssum by (...) for groupinghistogram_quantile() for latency| User Request | PromQL |
|---|---|
| "Error rate for payments in prod" | sum by (job) (rate(http_requests_total{job="payments", env="prod", status=~"5.."}[5m])) |
| "Latency p95 for frontend" | histogram_quantile(0.95, sum by (le) (rate(http_request_duration_seconds_bucket{app="frontend"}[5m]))) |
For:
โ Recommend Tempo + Loki correlations.