What it does
The server provides four tools: kafka_consumer_lag, which reports consumer group lag with incident-pattern diagnosis; kafka_topic_info, which shows topic details with ISR and replication health checks; kafka_broker_status, which reports brokers, the controller and under-replicated partitions; and kafka_list_topics, which lists topics with built-in descriptions.
The diagnosis logic is based on operational experience with a CDC pipeline rather than generic rules. A consumer group with no active members and growing lag is flagged as a likely Flink job failure, drawn from a 46-hour outage caused by a MySQL DELETE, a Debezium tombstone and a Flink NPE. It warns that resetting a consumer group alone is insufficient for Flink because checkpoints must be deleted first. It detects missing brokers under ClusterIdMismatch and suggests a Docker volume conflict as the root cause, and it treats an ISR below min.insync.replicas as a critical write-failure condition.
It is intended for teams running Kafka-based data pipelines who want lag and broker diagnostics within Claude Code. It is registered in a .mcp.json file. The environment variable KAFKA_BOOTSTRAP_SERVERS sets the bootstrap servers, with a default of localhost:9092, and KAFKA_COMMAND_TIMEOUT sets the command timeout in seconds, with a default of 10.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
io.github.Aguantar/kafka-dataops-mcp
DataOps Kafka MCP server with consumer lag diagnosis and broker monitoring
v0.1.0As shown on registry.modelcontextprotocol.io when we last read it