/grafana-billing

Grafana Billing Metrics Skill

Query key billing metrics from Prometheus and Loki through Grafana's data source proxy API.

What This Skill Does

• Queries active time series count from Prometheus (primary billing metric)
• Calculates data points per minute (DPM) ingestion rate
• Analyzes cardinality to identify top metrics by series count
• Queries Loki ingestion rate (GB/day)
• Compares metrics across staging and production environments

Core Capabilities

1. Prometheus Metrics: Active series, samples/sec, storage size, cardinality analysis
2. Loki Metrics: Ingestion rate, total bytes, active streams, memory chunks
3. Multi-environment: Query staging, prod, or both simultaneously
4. Flexible output: Human-readable tables or JSON for automation

Auto-Trigger Keywords

Primary Keywords

• billing metrics
• cost analysis
• active series
• observability costs

Secondary Keywords

• prometheus cardinality
• loki ingestion
• storage usage
• grafana usage
• metrics invoice
• time series count
• DPM (data points per minute)

Error Pattern Keywords

• "high cardinality"
• "ingestion rate too high"
• "storage growing"

When to Use

• Analyzing observability infrastructure costs
• Investigating high cardinality metrics
• Comparing staging vs production usage
• Preparing for capacity planning
• Debugging ingestion rate issues

Quick Usage

# Query all environments
uv run .claude/skills/grafana-billing/scripts/billing_metrics.py

# Query specific environment
uv run .claude/skills/grafana-billing/scripts/billing_metrics.py --env staging

# JSON output for scripting
uv run .claude/skills/grafana-billing/scripts/billing_metrics.py --json

# Only Prometheus metrics
uv run .claude/skills/grafana-billing/scripts/billing_metrics.py --service prometheus

Environment Variables Required

File Structure

grafana-billing/
├── SKILL.md              # Skill instructions for Claude
├── README.md             # This file
├── scripts/
│   ├── billing_metrics.py      # Main CLI
│   ├── grafana_client.py       # Grafana API client
│   ├── prometheus_metrics.py   # Prometheus queries
│   ├── loki_metrics.py         # Loki queries
│   └── formatters.py           # Output formatting
└── references/
    └── billing-metrics.md      # Metric definitions

Dependencies

Managed via uv inline script metadata:

• httpx - HTTP client
• python-dotenv - Environment variable loading
• rich - Table formatting

Official Documentation

• Grafana Cloud Metrics Billing
• Active Series and DPM
• Prometheus TSDB Status
• Loki Metrics

Related Skills

• grafana-plugin-scaffolding - Grafana plugin development
• kubernetes-operations - K8s cluster management