Empower large language model (LLM) agents like Claude and GPT to securely and efficiently manage Apache Kafka clusters on OCI Streaming. This server, built on the Model Context Protocol (MCP), offers a robust suite of 42 structured tools for comprehensive cluster, topic, consumer, and observability operations. It incorporates essential safety guardrails, including read-only defaults, risk classification for operations, and a policy guard requiring confirmation for destructive actions. With built-in AI diagnostic tools for scaling recommendations and lag analysis, enterprise-grade security features like SASL/SCRAM-SHA-512 + TLS, and structured audit logging, it delivers a secure, intelligent, and auditable solution for Kafka administration.
Key Features
01Policy guard with risk classification (LOW/MEDIUM/HIGH) and confirmation for destructive actions
0242 structured tools for comprehensive Kafka management, including cluster, topic, and consumer operations
03Enterprise-grade security with SASL/SCRAM-SHA-512 + TLS and private networking
04Read-only by default with explicit `--allow-writes` flag for write operations
05AI diagnostic tools to produce scaling recommendations and lag root cause analyses
060 GitHub stars
Use Cases
01Diagnosing Kafka performance issues and generating intelligent scaling recommendations using AI-powered tools
02Automating Kafka cluster and topic management via LLM agents for increased operational efficiency
03Securing Kafka administration with built-in policy guardrails, audit logging, and risk-classified operations
