grafana

Diagnoses incidents by querying Prometheus, Loki, and Tempo for metrics, logs, traces.

1|Updated Apr 8, 2026

Search Tags:#troubleshooting #kubernetes #observability #grafana #prometheus #loki #tempo

Ready to run

npx skills add https://github.com/roderik/fold --skill grafana-roderik

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: grafana
Source: https://github.com/roderik/fold/tree/main/skills/grafana
Command: npx skills add https://github.com/roderik/fold --skill grafana-roderik

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Grafana observability data across Loki logs, Prometheus metrics, and Tempo traces is hard to correlate manually; this skill streamlines digging into alerts, dashboards, and incidents by providing unified queries and workflow guidance.

Core Features & Use Cases

Query Loki for logs to diagnose errors, latency, and restarts.
Query Prometheus for metrics and uptime to identify bottlenecks.
Query Tempo traces to trace requests to pinpoint slow services; integrate with alert investigations.
Use case: When a production issue arises, investigate with consolidated logs, metrics, and traces across k8s pods.

Quick Start

Query Loki for recent logs in the target namespace and correlate with Prometheus metrics to identify root causes.

Frequently Asked Questions about grafana

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I correlate Loki logs and Prometheus metrics to troubleshoot Kubernetes incidents?▼

To troubleshoot Kubernetes incidents, you query Loki for logs and Prometheus for metrics to correlate errors and bottlenecks. This unifies observability data, letting you identify root causes across pods by analyzing restarts and latency alongside uptime metrics.

What's the best way to trace slow requests in a Kubernetes cluster using Tempo?▼

The best way to trace slow requests is to query Tempo traces to pinpoint the specific services causing latency. By investigating these traces alongside your infrastructure metrics, you can isolate the exact bottleneck within your distributed system architecture.

Do I need a Grafana CLI API key to investigate alerts with this approach?▼

Yes, investigating alerts requires access to a Grafana data source and the Grafana CLI API. The workflow uses 1Password for credential retrieval if configured, supporting both local and remote Grafana instances for comprehensive alert investigations.

Can I use dashboard-driven troubleshooting to diagnose production issues across data sources?▼

Yes, dashboard-driven troubleshooting diagnoses production issues by applying unified queries across Loki, Prometheus, and Tempo. This streamlines digging into consolidated logs, metrics, and traces to identify root causes during infrastructure incidents.

Why does manually correlating observability data fail during alert investigations?▼

Manually correlating observability data fails because Grafana data across Loki logs, Prometheus metrics, and Tempo traces is difficult to align during alert investigations. Unified queries and workflow guidance are needed to streamline digging into incidents effectively.