Skip to content
MCP server

Kafka MCP

Debug Kafka from an agent: find messages, measure lag, unblock consumers, compare clusters

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
CLI, self-hosted
Runs
Locally
Version
0.2.8
Listing updated
3 days ago

About Kafka MCP

Kafka MCP is an open-source MCP server for developer tools. It runs locally on your machine and is started by your AI client.

Kafka incidents usually start with a question like where is the message for order 1042, or why is this consumer stuck. This server turns those checks into tools an agent can call, and every write shows a preview before anything changes. Each answer states what it covered, so a partial scan is not mistaken for a full one.

One process can connect to several clusters. In stdio mode an endpoint is chosen for the session; in HTTP mode each endpoint gets its own path, so a session is bound to one cluster by how it connects rather than by a parameter a caller could forget. Endpoint policy lives in a kafka-mcp.yaml file: you can mark an endpoint read_only or switch off single tools such as copy_message, and keep passwords out of the file with {env:VAR} references. TLS, mTLS and SCRAM authentication are supported, and the client library is franz-go.

It ships as a binary, an MCP bundle for Claude Desktop, a container image and a registry entry, and a Kafka debugging skill is embedded in the binary.

Install

Run in a terminal

claude mcp add kafka-mcp -- docker run -i --rm ghcr.io/denizgursoy/kafka-mcp:v0.2.8

Add to claude_desktop_config.json

{
  "mcpServers": {
    "kafka-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/denizgursoy/kafka-mcp:v0.2.8"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "kafka-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/denizgursoy/kafka-mcp:v0.2.8"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "kafka-mcp": {
      "type": "stdio",
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/denizgursoy/kafka-mcp:v0.2.8"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "kafka-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/denizgursoy/kafka-mcp:v0.2.8"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "kafka-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/denizgursoy/kafka-mcp:v0.2.8"
      ]
    }
  }
}

Configuration

  • CONFIG_FILEWhere the mounted configuration is read from inside the container
  • CONFIG_FILEYour kafka-mcp.yaml: the clusters to connect to and the endpoints to expose.Required
  • CONFIG_FILEYour kafka-mcp.yaml: the clusters to connect to and the endpoints to expose.Required
  • CONFIG_FILEYour kafka-mcp.yaml: the clusters to connect to and the endpoints to expose.Required
  • CONFIG_FILEYour kafka-mcp.yaml: the clusters to connect to and the endpoints to expose.Required

Tools this server provides (3)

Taken from the project documentation; the running server may differ.

server_config
Reports the endpoint's description so a caller knows its intended use.
skill_reference
Reads the built-in Kafka debugging skill.
copy_message
Copies a message; can be switched off per endpoint in the configuration.

What it does

Find messages
Locates a message by a value such as an order id across a topic.
Consumer lag
Measures lag per consumer group.
Preview before write
Every change shows a preview before it is applied.
Multi-cluster endpoints
One process serves several clusters, each endpoint with its own policy and path.
Read-only mode
Endpoints can be read_only or disable individual tools.
Several install paths
Binary, MCP bundle, container image or registry install.

Good for

  • Trace a message by order id through a topic
  • Check how far behind a consumer group is
  • Unblock a stuck consumer with a previewed change
  • Compare two clusters before and after a migration
  • Give agents read-only access to production while keeping a separate writable endpoint

Pricing

Open source

Prices change. Check denizgursoy.github.io before you buy.

Kafka MCP FAQ

Can an agent change data by accident?

Writes show a preview first, and endpoints can be set read_only or have tools such as copy_message turned off.

Does it support TLS and SASL?

Yes. The sample config shows TLS with a CA file, optional mTLS, and SCRAM-SHA-256 or SCRAM-SHA-512.

How do I keep passwords out of the config?

Use {env:VAR} references or password_file instead of writing the password in the YAML.

What does it need to run from source?

The site says Go 1.27 or later, and the tests need Docker.

Is Kafka MCP free?

Yes. Kafka MCP is open source, so you can use and self-host it for free.

How do I add Kafka MCP to Claude Code?

Run this command in your terminal: claude mcp add kafka-mcp -- docker run -i --rm ghcr.io/denizgursoy/kafka-mcp:v0.2.8

What are the best alternatives to Kafka MCP?

The closest alternatives to Kafka MCP are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to Kafka MCP

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools