How is listing metrics different from querying them?

Listing returns the metric names available; querying returns a metric's values over time. An agent lists first to discover the exact series name, then queries that series for the number. Asking for a value without the right name is how a query comes back empty.

Observability metrics or product metrics?

Both, depending on the server. Prometheus and Axiom list observability metrics (latency, error rate, throughput); Mixpanel lists product-analytics metrics defined in a project. The list call is the same idea on each: return the metric names so an agent can pick one to query.

MCP servers that can list metrics

3 verified servers expose a tool that can list available metrics

Before an agent can query a metric, it has to know which metrics exist. Listing metrics returns the available names, so the agent picks the right series, request latency, signups, error rate, rather than guessing at a metric that may not be there.

These verified servers let an agent list the available metrics.

Top pick

Prometheus

pab1it0

Community

A maintained Prometheus MCP server: run PromQL instant and range queries, discover metrics and metadata, and inspect scrape targets from your agent.

monitoring-observability450

Tool:

list_metrics

Prometheus's list_metrics returns the metric names it scrapes, with pagination and filtering, the discovery step before an agent runs PromQL.

Pick 2

Axiom

Official

Axiom's official remote MCP server lets agents query logs, traces, and metrics with APL and manage datasets, monitors, and dashboards over OAuth.

monitoring-observability

Tool:

listMetrics

On Axiom, listMetrics returns the metric names in a dataset, so an agent can find the series to query or chart.

Pick 3

Mixpanel

Official

Mixpanel's official hosted MCP server lets an agent query events, funnels, retention, and dashboards and manage metrics, experiments, and feature flags.

data-analytics

Tool:

List-Metrics

In Mixpanel, List-Metrics returns the metrics defined in a project, the product-analytics series an agent reports on.

What to know

A metric list is the discovery step before any query. A monitoring or analytics backend can track thousands of series, and an agent needs the exact name before it asks for a value or builds a chart. The tools return the metric names the backend knows: Prometheus the metrics it scrapes, Axiom the ones in a dataset, Mixpanel the metrics defined in a project. From the list the agent picks the series it means to query, which keeps it from asking for a metric that was renamed or never existed.

The set of metrics a backend exposes is fairly stable, so an agent that learns the relevant names once can hold them rather than re-list every session. Remembering which metric answers which question lets it go straight to the query instead of rediscovering the catalog each time.

Questions

How is listing metrics different from querying them?: Listing returns the metric names available; querying returns a metric's values over time. An agent lists first to discover the exact series name, then queries that series for the number. Asking for a value without the right name is how a query comes back empty.
Observability metrics or product metrics?: Both, depending on the server. Prometheus and Axiom list observability metrics (latency, error rate, throughput); Mixpanel lists product-analytics metrics defined in a project. The list call is the same idea on each: return the metric names so an agent can pick one to query.