Find the series explosion without corrupting your data.
metrics troubleshooting
Diagnoses active Prometheus cardinality problems and identifies safe remediation paths.
When to use it
Use when the user is currently experiencing a cardinality incident.
Give it a current Prometheus performance, memory, or cost problem; it returns a systematic diagnosis and remediation path.
What you provide
This skill
Prometheus HTTP API
Reads Prometheus cardinality data
The main triage workflow inspects an active Prometheus TSDB endpoint.
jq parses the TSDB and label API responses used during diagnosis.
Hands off prevention and future label-strategy design; it is not required for the current diagnosis.
Hands off post-ingest aggregation when metrics must be retained while reducing cost.