strimzi-kafka-cli — Claude Code MCP server: Kafka on Kubernetes
Gives your AI agent 42 tools to manage Kafka clusters, topics, users and connectors through the Strimzi operator.
claude mcp add strimzi-kafka-cli -- uvx strimzi-kafka-cliWhat strimzi-kafka-cli does
Strimzi Kafka CLI wraps the kfk command for running Apache Kafka on Kubernetes with the Strimzi operator. The bundled MCP server exposes that to an agent: clusters, topics, users, Kafka Connect, MirrorMaker2, ACLs, node pools and backups. 42 tools in total, from list_kafkas and create_topic to install_operator and backup_kafka.
Start it with kfk mcp, or register it for Claude with claude mcp add strimzi-kafka-cli -- kfk mcp. Strimzi resources download on the first kfk call, and the CLI talks to your cluster through the Python Kubernetes client. Python 3.11 or newer is required, so older interpreters won't install it.
Install via pip or pipx, adding the [mcp] extra for server support. The Homebrew route ships MCP by default and keeps your main Python environment untouched. If you already run Kafka on Kubernetes, driving it from the agent instead of the terminal saves a lot of kubectl typing.
What it helps with
- create a topic with a custom config without hand-writing YAML
- list Kafka clusters and check their status
- add a user with the right access rules
- deploy a Kafka Connect connector from the agent
- back up a cluster before making changes
Example prompts for the agent
- List all Kafka clusters in the kafka namespace
- Create a topic named orders with 6 partitions and replication factor 3
- Back up my-cluster and save it to a file
How to install strimzi-kafka-cli
Platform CLI
claude mcp add strimzi-kafka-cli -- uvx strimzi-kafka-cliuvx runs the command named after the package; if the README names another one, use uvx --from strimzi-kafka-cli <command>
Platform CLI
codex mcp add strimzi-kafka-cli -- uvx strimzi-kafka-cliuvx runs the command named after the package; if the README names another one, use uvx --from strimzi-kafka-cli <command>
Config file
{
"mcpServers": {
"strimzi-kafka-cli": {
"command": "uvx",
"args": [
"strimzi-kafka-cli"
]
}
}
}uvx runs the command named after the package; if the README names another one, use uvx --from strimzi-kafka-cli <command>
Config file
{
"servers": {
"strimzi-kafka-cli": {
"type": "stdio",
"command": "uvx",
"args": [
"strimzi-kafka-cli"
]
}
}
}uvx runs the command named after the package; if the README names another one, use uvx --from strimzi-kafka-cli <command>
Platform CLI
gemini mcp add strimzi-kafka-cli uvx strimzi-kafka-cliuvx runs the command named after the package; if the README names another one, use uvx --from strimzi-kafka-cli <command>
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about strimzi-kafka-cli
How do I install strimzi-kafka-cli in Claude Code?
Run claude mcp add strimzi-kafka-cli -- uvx strimzi-kafka-cli. You need Python 3.11 or newer first.
What does strimzi-kafka-cli do?
It's a CLI and MCP server for managing Apache Kafka on Kubernetes through the Strimzi operator. Your agent gets 42 tools covering clusters, topics, users, connectors, MirrorMaker2, ACLs and backups.
How is the MCP server different from the plain CLI?
The CLI runs in your terminal; the MCP server (kfk mcp) exposes the same operations to an AI assistant. Installing with MCP support needs the [mcp] extra on pip or pipx.
