Prometheus Metrics & time-series
Ask plain-English questions about your own Prometheus metrics — what is being scraped, what a metric reads right now, and how it moved over time — answered from live data.
Connect your Prometheus and Sauble can query your real metrics without hand-writing PromQL. It reads which targets are up, what each metric actually means, its current value, and the full time series behind a spike.
What Sauble can do
5 capabilities, grouped by what you are trying to find out.
Query your metrics
execute_queryThe current value of any metric or PromQL expression — a point-in-time reading.execute_range_queryThe same query over a time window, stepped into a series — how a value moved, and when.Discover what is collected
list_metricsEvery metric name Prometheus is storing, searchable by pattern.get_metric_metadataWhat a metric means — its type, unit, and help text — so you read it correctly.Check what is being scraped
get_targetsEvery scrape target and whether it is up — so you know your data is actually flowing in.Full read APIexecute_queryruns any PromQL expression — the full query surface, not just these examples.
Plus health_check — a quick confirmation that the connector can reach your Prometheus before you rely on an answer.
Ask it things like
Natural language — Sauble picks the right tools for you.
How Prometheus organizes your data
The 5 things Sauble reads — and how they connect.
Targets
The endpoints Prometheus scrapes for metrics — each one either up or down.
Metrics
The named measurements collected from targets, each with a type, unit, and help text.
Labels
Key/value dimensions on a metric — host, job, instance — used to slice and filter it.
Series
A metric plus one unique set of label values — a single line you can chart over time.
Samples
The timestamped values within a series — the raw points behind any trend.