Token导航 LogoToken导航TokenDH.com
前端设计敏感数据github未标认证来源可访问clear审计未展示

kafka-mcp-integrationKafka MCP 集成

Agent Skill

kafka-mcp-integration 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

424

周安装

17

GitHub Stars

127

下载量

137
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

3

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:kafka-mcp-integration(Kafka MCP 集成)
来源仓库:https://github.com/anton-abyzov/specweave
仓库路径:skills/kafka-mcp-integration
安装命令:
npx skills add https://github.com/anton-abyzov/specweave --skill kafka-mcp-integration
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/anton-abyzov/specweave --skill kafka-mcp-integration

简介

实现 Kafka 与 MCP 协议集成的中间件适配层。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中构建统一消息网关。
  • 通过 GitHub 仓库安装,使用 npx skills add 命令添加技能。
  • 协议转换需保证消息顺序和事务完整性。kafka-mcp-integration 属于前端设计类 Skill,可作为该场景下的辅助能力补充。
  • 部署时应隔离不同租户流量,防止相互干扰。

SKILL.md

Kafka MCP Server Integration

Expert knowledge for integrating SpecWeave with Kafka MCP (Model Context Protocol) servers. Supports 4 MCP server implementations with auto-detection and configuration guidance.


Code-First Recommendation: For most Kafka automation tasks, writing code is better than MCP (98% token reduction). Use kafkajs or kafka-node directly: ``typescript import { Kafka } from 'kafkajs'; const kafka = new Kafka({ brokers: ['localhost:9092'] }); const producer = kafka.producer(); await producer.send({ topic: 'events', messages: [{ value: 'Hello' }] }); `` When MCP IS useful: Quick interactive debugging, topic exploration, Claude Desktop integration. When to use code instead: CI/CD pipelines, test automation, production scripts, anything that should be committed and reusable.

Supported MCP Servers

1. kanapuli/mcp-kafka (Node.js)

Installation:

npm install -g mcp-kafka

Capabilities:

  • Authentication: SASL_PLAINTEXT, PLAINTEXT
  • Operations: produce, consume, list-topics, describe-topic, get-offsets
  • Best for: Basic Kafka operations, quick prototyping

Configuration Example:

{
  "mcpServers": {
    "kafka": {
      "command": "npx",
      "args": ["mcp-kafka"],
      "env": {
        "KAFKA_BROKERS": "localhost:9092",
        "KAFKA_SASL_MECHANISM": "plain",
        "KAFKA_SASL_USERNAME": "user",
        "KAFKA_SASL_PASSWORD": "password"
      }
    }
  }
}

2. tuannvm/kafka-mcp-server (Go)

Installation:

go install github.com/tuannvm/kafka-mcp-server@latest

Capabilities:

  • Authentication: SASL_SCRAM_SHA_256, SASL_SCRAM_SHA_512, SASL_SSL, PLAINTEXT
  • Operations: All CRUD operations, consumer group management, offset management
  • Best for: Production use, advanced SASL authentication

Configuration Example:

{
  "mcpServers": {
    "kafka": {
      "command": "kafka-mcp-server",
      "args": [
        "--brokers", "localhost:9092",
        "--sasl-mechanism", "SCRAM-SHA-256",
        "--sasl-username", "admin",
        "--sasl-password", "admin-secret"
      ]
    }
  }
}

3. Joel-hanson/kafka-mcp-server (Python)

Installation:

pip install kafka-mcp-server

Capabilities:

  • Authentication: SASL_PLAINTEXT, PLAINTEXT, SSL
  • Operations: produce, consume, list-topics, describe-topic
  • Best for: Claude Desktop integration, Python ecosystem

Configuration Example:

{
  "mcpServers": {
    "kafka": {
      "command": "python",
      "args": ["-m", "kafka_mcp_server"],
      "env": {
        "KAFKA_BOOTSTRAP_SERVERS": "localhost:9092"
      }
    }
  }
}

4. Confluent Official MCP (Enterprise)

Installation:

confluent plugin install mcp-server

Capabilities:

  • Authentication: OAuth, SASL_SCRAM, API Keys
  • Operations: All Kafka operations, Schema Registry, ksqlDB, Flink SQL
  • Advanced: Natural language interface, AI-powered query generation
  • Best for: Confluent Cloud, enterprise deployments

Configuration Example:

{
  "mcpServers": {
    "kafka": {
      "command": "confluent",
      "args": ["mcp", "start"],
      "env": {
        "CONFLUENT_CLOUD_API_KEY": "your-api-key",
        "CONFLUENT_CLOUD_API_SECRET": "your-api-secret"
      }
    }
  }
}

Auto-Detection

SpecWeave can auto-detect installed MCP servers:

/sw-kafka:mcp-configure

This command:

  1. Scans for installed MCP servers (npm, go, pip, confluent CLI)
  2. Checks which servers are currently running
  3. Ranks servers by capabilities (Confluent > tuannvm > kanapuli > Joel-hanson)
  4. Generates recommended configuration
  5. Tests connection

Quick Start

Option 1: Auto-Configure (Recommended)

/sw-kafka:mcp-configure

Interactive wizard guides you through:

  • MCP server selection (or auto-detect)
  • Broker URL configuration
  • Authentication setup
  • Connection testing

Option 2: Manual Configuration

  1. Install preferred MCP server (see installation commands above)
  2. Create .mcp.json configuration:
{
  "serverType": "tuannvm",
  "brokerUrls": ["localhost:9092"],
  "authentication": {
    "mechanism": "SASL/SCRAM-SHA-256",
    "username": "admin",
    "password": "admin-secret"
  }
}
  1. Test connection:
# Via MCP server CLI
kafka-mcp-server test-connection

# Or via SpecWeave
node -e "import('./dist/lib/mcp/detector.js').then(async ({ MCPServerDetector }) => {
  const detector = new MCPServerDetector();
  const result = await detector.detectAll();
  console.log(JSON.stringify(result, null, 2));
});"

MCP Server Comparison

FeaturekanapulituannvmJoel-hansonConfluent
LanguageNode.jsGoPythonOfficial CLI
SASL_PLAINTEXT
SCRAM-SHA-256
SCRAM-SHA-512
mTLS/SSL
OAuth
Consumer Groups
Offset Mgmt
Schema Registry
ksqlDB
Flink SQL
AI/NL Interface
Best ForPrototypingProductionDesktopEnterprise

Troubleshooting

MCP Server Not Detected

# Check if MCP server installed
npm list -g mcp-kafka         # kanapuli
which kafka-mcp-server        # tuannvm
pip show kafka-mcp-server     # Joel-hanson
confluent version             # Confluent

Connection Refused

  • Verify Kafka broker is running: kcat -L -b localhost:9092
  • Check firewall rules
  • Validate broker URL (correct host:port)

Authentication Failed

  • Double-check credentials (username, password, API keys)
  • Verify SASL mechanism matches broker configuration
  • Check broker logs for authentication errors

Operations Not Working

  • Ensure MCP server supports the operation (see comparison table)
  • Check broker ACLs (permissions for the authenticated user)
  • Verify topic exists: /sw-kafka:mcp-configure list-topics

Operations via MCP

Once configured, you can perform Kafka operations via MCP:

import { MCPServerDetector } from './lib/mcp/detector';

const detector = new MCPServerDetector();
const result = await detector.detectAll();

// Use recommended server
if (result.recommended) {
  console.log(`Using ${result.recommended} MCP server`);
  console.log(`Reason: ${result.rankingReason}`);
}

Security Best Practices

  1. Never commit credentials - Use environment variables or secrets manager
  2. Use strongest auth - Prefer SCRAM-SHA-512 > SCRAM-SHA-256 > PLAINTEXT
  3. Enable TLS/SSL - Encrypt communication with broker
  4. Rotate credentials - Regularly update passwords and API keys
  5. Least privilege - Grant only necessary ACLs to MCP server user

Related Commands

  • /sw-kafka:mcp-configure - Interactive MCP server setup
  • /sw-kafka:dev-env start - Start local Kafka for testing
  • /sw-kafka:deploy - Deploy production Kafka cluster

External Links

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Claude Code

29.09%
按下载量换算40

Antigravity

23.97%
按下载量换算33

Cursor

20.46%
按下载量换算28

Gemini CLI

13.74%
按下载量换算19

OpenCode

7.36%
按下载量换算10

Codex

3.9%
按下载量换算5

安全审计

暂无安全审计结果可展示。

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。

来源信息

继续浏览同类 Skills