# Does Apache Kafka have an MCP server?

**Sort of — Community only.** Apache Kafka does not maintain an official MCP server. Community-built servers exist.

> A first-party Apache Kafka MCP server is proposed in KIP-1318, but it is still under discussion and nothing has shipped.

## Facts

- Verdict: Community only (Sort of)
- Access: Local server — Runs on your machine over stdio, launched by your MCP client.
- Repository: [confluentinc/mcp-confluent](https://github.com/confluentinc/mcp-confluent) — ★ 169, last push 2026-09-29, MIT
- Maintainer: Confluent Inc.
- Language: TypeScript
- Category: [Data Platforms](https://mcpyet.com/category/data-platforms/)
- Verified against vendor docs: 2026-09-29
- GitHub data refreshed: 2026-09-29 (re-checked every 6 hours)

## The receipts

- **Top community server:** [confluentinc/mcp-confluent](https://github.com/confluentinc/mcp-confluent) — ★ 169, pushed 2026-09-29, MIT (active)

## How to connect it

Runs on your machine over stdio, launched by your MCP client.

### Claude Code

```sh
claude mcp add kafka -- npx -y <package-from-readme>
```

Swap in the exact package and flags from the repo README — most servers need an API key passed as an environment variable. Add `--scope user` to make it available in every project.

### Claude Desktop

Open **Settings → Developer → Edit Config** and add the server to `claude_desktop_config.json`, then restart Claude Desktop:

```json
{
  "mcpServers": {
    "kafka": {
      "command": "npx",
      "args": [
        "-y",
        "<package-from-readme>"
      ],
      "env": {
        "API_KEY": "…"
      }
    }
  }
}
```

### Cursor

Add to `.cursor/mcp.json` in your project, or `~/.cursor/mcp.json` for every project:

```json
{
  "mcpServers": {
    "kafka": {
      "command": "npx",
      "args": [
        "-y",
        "<package-from-readme>"
      ],
      "env": {
        "API_KEY": "…"
      }
    }
  }
}
```

### VS Code

Add to `.vscode/mcp.json`, then pick the server from Copilot's agent-mode tools menu:

```json
{
  "servers": {
    "kafka": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "<package-from-readme>"
      ]
    }
  }
}
```

### ChatGPT

ChatGPT connects to **remote servers only**, and this one runs on your machine. Either expose it through an MCP gateway that gives it a public URL, or use a client with native local support (Claude or Cursor).

## Before you connect this

A community server is unaudited third-party code that will hold your Apache Kafka API credentials and see whatever data they unlock. Independent research puts SSRF vulnerabilities in roughly a third of public MCP servers, and 41% ship with no authentication at all.

- Read the source before running it — especially any network calls it makes.
- Check who maintains it and whether commits are recent (this one: last push 2026-09-29).
- Issue a scoped, revocable API key rather than an account-wide token.
- Prefer read-only credentials until you trust it with writes.

## How we know

The Apache Kafka project ships no MCP server: KIP-1318 is 'Under Discussion' and its KAFKA-20436 ticket is unresolved, and no MCP Registry entry uses an apache.org namespace. Confluent's confluentinc/mcp-confluent (~170 stars, active) is the strongest dedicated server and supports standalone Apache Kafka clusters as well as Confluent Cloud and Confluent Platform.

- GitHub data refreshed 2026-09-29, then every 6 hours.
- Vendor docs verified on 2026-09-29.
- Corrections: hello@mcpyet.com.

## Also in Data Platforms

- [Tableau](https://mcpyet.com/mcp/tableau.md) — Official (Yes)
- [Databricks](https://mcpyet.com/mcp/databricks.md) — Official (Yes)
- [Power BI](https://mcpyet.com/mcp/power-bi.md) — Official (Yes)
- [Looker](https://mcpyet.com/mcp/looker.md) — Official (Yes)
- [Metabase](https://mcpyet.com/mcp/metabase.md) — Official (Yes)
- [Amplitude](https://mcpyet.com/mcp/amplitude.md) — Official (Yes)

---

Source: https://mcpyet.com/mcp/kafka/ — data refreshed 2026-09-29