What problem does it solve? Observability costs from Grafana Cloud are driven by active time series, data points per minute, and log ingestion volume, but these metrics are scattered across Prometheus and Loki internals. This Skill consolidates them into a single report so you can analyze billing drivers, spot high-cardinality metrics, and compare staging versus production usage without writing PromQL by hand. ## Core Features & Use Cases - Prometheus Billing Metrics: Retrieves active series count, samples per second (converted to DPM), TSDB storage size, and top-10 cardinality breakdown via the TSDB status endpoint, with fallbacks for Cortex/Mimir and AWS Managed Prometheus. - Loki Ingestion Metrics: Reports ingestion rate in GB/day, cumulative bytes received, active streams, memory chunks, and rejected bytes from rate limiting. - Multi-Environment Comparison: Queries staging and production Grafana workspaces in one run, with human-readable Rich tables or JSON output for automation. - Use Case: Before a capacity planning review, run the CLI against both environments to get active series, DPM, and Loki GB/day side by side, then drill into the top cardinality metrics to identify cost reduction targets. ## Quick Start Ask the AI to run the billing metrics script to compare Prometheus and Loki usage across staging and production, making sure GRAFANA_STAGING_API_KEY and GRAFANA_PROD_API_KEY are set first.
