inlined-refs

Generate Kafka consumer group lag dashboards with Prometheus exporters and Grafana.

Updated Jul 7, 2026
One-click install
npx skills add https://github.com/ricardolui/gcp-custom-agent-skills --skill inlined-refs-ricardolui
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: inlined-refs
Source: https://github.com/ricardolui/gcp-custom-agent-skills/tree/main/confluent-skill-reviewer/evals/mock-skills/inlined-refs
Command: npx skills add https://github.com/ricardolui/gcp-custom-agent-skills --skill inlined-refs-ricardolui

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Setting up Kafka consumer lag monitoring requires choosing between exporters, configuring Prometheus metrics, and wiring Grafana dashboards, which involves scattered documentation and trial-and-error decisions. ## Core Features & Use Cases - Exporter Selection Guidance: Compares kafka-exporter (Danielqsj) and Confluent Control Center with pros, cons, and cluster-size recommendations. - Grafana Dashboard Setup: Recommends Grafana dashboard 7589 and the kafka_consumergroup_lag metric plotted over a 5-minute window. - Alert Configuration: Provides a default alert threshold of lag greater than 10,000 messages for 5 minutes, adjustable per topic SLA. - Use Case: A platform engineer needs to monitor consumer lag on a 50-broker cluster; the Skill recommends kafka-exporter, supplies its launch flags, and specifies the Grafana dashboard and alert rule. ## Quick Start Set up a consumer lag dashboard for my Kafka cluster with Prometheus and Grafana.

Frequently Asked Questions about inlined-refs

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

FAQPage Schema
How do I set up a Kafka consumer lag dashboard?

Choose an exporter such as kafka-exporter, point it at your Kafka bootstrap server, then import Grafana dashboard 7589. Plot the kafka_consumergroup_lag metric filtered by consumer group over a 5-minute window.

kafka-exporter vs Confluent Control Center for lag monitoring?

kafka-exporter is a lightweight single binary exposing per-group lag directly, best for clusters under 100 brokers and 1000 consumer groups. Confluent Control Center is official and covers full Confluent Platform metrics but requires paid licensing and a heavier footprint.

What is a good Kafka consumer lag alert threshold?

A reasonable default is lag greater than 10,000 messages sustained for 5 minutes. Adjust the threshold per topic based on its SLA and normal consumption throughput.

When should I not use kafka-exporter?

Avoid kafka-exporter on very large clusters because it is stateful, consuming __consumer_offsets internally, and can fall behind on clusters with many brokers or consumer groups. Enterprise Confluent Platform users may prefer Control Center instead.

Does this cover Kafka producer or broker JMX metrics?

No, the scope is limited to consumer group lag monitoring. Producer metrics, broker JMX monitoring, and Kafka Streams-specific monitoring are explicitly out of scope.