Grafana Dashboards

Creates Grafana dashboard JSON models with well-organized panels, template variables for dynamic filtering, annotation queries for deployment markers, and optimized data source queries. Supports Prometheus, InfluxDB, Elasticsearch, CloudWatch, and other data sources with proper visualization types for different metric categories.

Usage

Describe what you want to monitor and the key metrics you care about. Specify your data source (Prometheus, InfluxDB, etc.), the services or infrastructure components, and any specific visualization preferences. The skill generates complete dashboard JSON that can be imported directly into Grafana or provisioned via the API.

Examples

• "Create a RED method dashboard for a web service showing request rate, error rate, and duration percentiles"
• "Build a Node Exporter dashboard with CPU, memory, disk I/O, and network panels grouped by host"
• "Design a Kubernetes cluster overview dashboard with namespace-level resource usage and pod status"
• "Create a business metrics dashboard showing signups, conversions, and revenue with time comparisons"

Guidelines

• Use template variables for environment, service, and instance filtering to make dashboards reusable
• Group related panels into rows with descriptive titles and collapse less-important sections by default
• Choose appropriate visualization types: time series for trends, stat for current values, heatmap for distributions
• Set meaningful Y-axis units (bytes, seconds, percent) and auto-scale ranges with sensible min/max
• Add panel descriptions explaining what the metric means and when to be concerned about its value
• Use $__rate_interval instead of hardcoded intervals in Prometheus rate() and increase() queries
• Configure appropriate refresh intervals: 10s for real-time dashboards, 1m+ for historical analysis
• Add annotations for deployments and incidents to correlate changes with metric movements