Download for MacApple silicon, macOS 14+
Skill

dpm-finder

@grafana Updated 2026-10-17

Find the Prometheus metrics that drive your Grafana Cloud bill. `dpm-finder` is a Grafana Professional Services CLI that ranks metrics by Data Points per Minute (DPM) with per-label-set breakdown, optional `--cost-per-1000-series` pricing, and a Prometheus-exporter mode. Use when investigating high Grafana Cloud spend, hunting noisy / high-cardinality metrics, comparing pre/post recording-rule cardinality, or feeding cost data into dashboards — even when the user says "why is my Mimir bill so high?", "find the biggest metrics", "cardinality offenders", or "optimize Prometheus cost" without naming dpm-finder.

Install

With your agent

Install OMGSkills skill "grafana/skills:dpm-finder" into this agent's normal skills directory: call the OMGSkills MCP get_install_plan tool for your agent and run the script it returns (it installs only the pinned commit and verifies SKILL.md), and stop if MCP or the pin is unavailable. Do not overwrite or change existing skills without asking me.

Manual command

git clone https://github.com/grafana/skills /tmp/skills && ln -s /tmp/skills/skills/grafana-cloud/dpm-finder ~/.claude/skills/dpm-finder

From README

dpm-finder Grafana PS tool ranking Prometheus metrics by DPM with per-series breakdown. Source: https://github.com/grafana-ps/dpm-finder Prerequisites Python 3.9+ A Grafana Cloud Prometheus endpoint URL + numeric stack ID + API key (glc…, metrics:read scope) Common Workflows One-shot analysis (most common) If step 5 is empty, lower -m or confirm the endpoint URL has no trailing path after .net. Discover stack details with gcx If gcx is installed it can derive the endpoint + username: The Prometheus endpoint pattern is https://prometheus-{clusterslug}.grafana.net. Username is the numeric stack ID. Without gcx: look up in the Grafana Cloud portal, or query grafanacloudinstanceinfo{name="STACKNAME."} on the datasource. Multi-stack runs Limit to max 3 concurrent runs to avoid GCloud rate limits. Batch the stacks and wait for each batch before the next.