kafka-streaming

Inspect Kafka brokers, topics, consumer groups, and consumer lag for incident investigation.

Updated Mar 26, 2026
One-click install
npx skills add https://github.com/erwinv2k-TKG/AgentesVSC --skill kafka-streaming-erwinv2k-tkg
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kafka-streaming
Source: https://github.com/erwinv2k-TKG/AgentesVSC/tree/main/packs/incidentfox/sre-agent/.claude/skills/streaming-kafka
Command: npx skills add https://github.com/erwinv2k-TKG/AgentesVSC --skill kafka-streaming-erwinv2k-tkg

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires confluent-kafka, and includes scripts (resource) components.

What problem does it solve? Investigating Kafka issues during an incident usually requires juggling CLI tools and remembering admin commands. This Skill provides ready-to-run Python scripts that report broker health, topic configuration, consumer group membership, and consumer lag in structured JSON, so you can diagnose streaming pipeline problems quickly. ## Core Features & Use Cases - Broker and Cluster Health Checks: Retrieve cluster ID, controller, and broker inventory as the mandatory first step of any investigation. - Topic Inspection: List topics and describe partitions, replication, configs, and per-partition offsets, including under-replicated partition detection. - Consumer Lag Analysis: Compute per-partition lag with automatic health classification (healthy, minor_lag, lagging, severely_lagging). - Use Case: During an outage, run the broker info script, list consumer groups, then check lag for the affected group to determine whether consumers are stuck or the cluster itself is unhealthy. ## Quick Start Ask the agent to check the consumer lag for a specific Kafka consumer group and report whether it is healthy.

Frequently Asked Questions about kafka-streaming

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

FAQPage Schema
How do I check Kafka consumer lag from Python?

Use the get_consumer_lag.py script with a consumer group ID to compare committed offsets against partition high watermarks. It returns per-partition lag, total lag, and a health classification from healthy to severely_lagging.

How to list Kafka topics and consumer groups with confluent-kafka?

The list_topics.py and list_consumer_groups.py scripts use the confluent-kafka AdminClient to enumerate topics with partition details and all consumer groups with their states, outputting results as JSON.

Does this Kafka tooling support SASL and SSL authentication?

Yes, the shared kafka_client module supports PLAINTEXT, SSL, SASL_SSL, and SASL_PLAINTEXT protocols configured via environment variables. Credentials are injected transparently by a proxy layer, so scripts run without manual secret handling.

Why does describe_topic report under-replicated partitions?

A partition is under-replicated when its in-sync replica count is lower than its configured replica count, usually indicating a broker is down or lagging. The script flags these partitions so you can identify availability risks.

What Kafka consumer lag is considered a problem?

Total lag of zero is healthy, under 1,000 is minor_lag, under 100,000 is lagging, and 100,000 or more is severely_lagging. These thresholds help prioritize whether a consumer group needs immediate intervention.