Skip to main content
Glama
Elgato MCP Server

Elgato MCP Server

Official
by elgatosf

Elgato MCP Server

npm version License: MIT

一个 Model Context Protocol (MCP) 服务器,用于在 AI 助手(如 Claude Desktop)与 Elgato 应用之间建立桥接。

概述

Elgato MCP Server 通过 IPC 充当 MCP 客户端与 Elgato 应用之间的协议桥接:

MCP Client <--MCP Transport--> Bridge <--Unix Socket/Named Pipe--> Elgato App

主要功能:

  • 🔌 动态工具发现 — 通过 MCP 自动发现并暴露来自已连接 Elgato 应用的工具

  • 🚀 双传输支持 — stdio(用于 Claude Desktop)和 HTTP(用于 Web 客户端)

  • 🌐 ngrok 集成 — 可选的公共隧道,用于远程访问

  • 🔄 热重连 — 应用可用时自动重新连接

  • 💻 跨平台 — 支持 Windows 和 macOS

  • 📢 通知转发 — 将应用通知转发给已连接的 MCP 客户端

Related MCP server: idh

安装

# Global installation (recommended)
npm install -g @elgato/mcp-server

# Or with pnpm
pnpm add -g @elgato/mcp-server

无需安装即可运行

您可以使用 npx 直接运行 Elgato MCP Server,无需全局安装:

# Run with stdio transport (default)
npx -y @elgato/mcp-server@latest

# Run with HTTP transport
npx -y @elgato/mcp-server@latest --http

# Run with HTTP transport on a custom port
npx -y @elgato/mcp-server@latest --http --port 3000

# Run with ngrok tunnel
NGROK_AUTHTOKEN=your_token npx -y @elgato/mcp-server@latest --http --ngrok

# Run with verbose logging
npx -y @elgato/mcp-server@latest --verbose

这对于在不进行永久安装的情况下试用服务器,或用于 CI/CD 环境非常有用。

用法

stdio 传输(默认)

用于与 Claude Desktop 或其他使用标准 I/O 的 MCP 客户端集成:

elgato-mcp-server

HTTP 传输

用于基于 Web 的客户端或远程访问:

# Start HTTP server on default port (9090)
elgato-mcp-server --http

# Custom port
elgato-mcp-server --http --port 3000

# With ngrok tunnel (requires NGROK_AUTHTOKEN env var)
NGROK_AUTHTOKEN=your_token elgato-mcp-server --http --ngrok

CLI 选项

Options:
  --transport <mode>  Transport mode: 'stdio' (default) or 'http'
  --http              Shorthand for --transport http
  --port <number>     HTTP server port (default: 9090)
  --ngrok             Enable ngrok tunnel (requires NGROK_AUTHTOKEN env var)
  --help, -h          Show help message
  --verbose, -v       Enable verbose logging (default: silent)

日志

日志写入 stderr,并带有严重性标签(ERROR、WARN、INFO、DEBUG)。错误和警告始终输出;info/debug 需要 --verbose。

Claude Desktop 配置

使用桌面扩展(推荐)

最简单的安装方式是使用打包好的桌面扩展(.mcpb),它完全自包含,并运行在 Claude Desktop 自带的 Node.js 运行时上——无需 Node.js、npm 或互联网访问:

  1. 从发布页面下载最新的 elgato_stream_deck-<version>.mcpb。

  2. 在 Claude Desktop 中,打开设置 → 扩展,将 .mcpb 文件拖入窗口(或双击该文件)。

  3. 确保 Stream Deck 应用正在运行——连接后其工具会自动出现。内置的 bridge_status 工具可随时报告连接状态。

或者,手动将服务器添加到您的 Claude Desktop 配置文件中:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

使用全局安装

{
    "mcpServers": {
        "elgato": {
            "command": "elgato-mcp-server"
        }
    }
}

使用 npx(无需安装)

或者,您可以使用 npx 运行服务器,无需全局安装:

{
    "mcpServers": {
        "elgato": {
            "command": "npx",
            "args": ["--yes", "@elgato/mcp-server@latest"]
        }
    }
}

注意: 基于 npx 的配置适用于 Claude Desktop 和其他支持 stdio 传输的 MCP 客户端。首次运行可能需要一些时间,因为 npx 会下载软件包。

HTTP 端点

在 HTTP 模式下运行时,以下端点可用:

Endpoint

Method

Description

/mcp

POST

MCP 请求端点

/mcp

GET

用于流式传输的 Server-Sent Events (SSE)

/mcp

DELETE

关闭会话

/health

GET

健康检查端点

要求

  • 支持 MCP 插件的 Elgato 应用(例如 Stream Deck)

  • Node.js 18 或更高版本

  • 支持的平台:Windows、macOS

隐私政策

此服务器完全在您的机器上运行,仅通过本地 IPC(macOS 上的 Unix 套接字、Windows 上的命名管道)与本地安装的 Elgato 应用通信。它不会收集、存储或向外部服务传输个人数据。工具输入和输出在您的 MCP 客户端(例如 Claude Desktop)与同一台机器上已连接的 Elgato 应用之间流动;可选的 ngrok 隧道仅在显式启用 --ngrok 时建立。

有关 Elgato/Corsair 的数据处理实践,请参阅 Elgato 隐私政策。

贡献

我们欢迎贡献!有关开发环境搭建、编码指南和贡献流程,请参阅 CONTRIBUTING.md。

有关详细的技术文档和架构信息,请参阅 TECHNICAL_SPECIFICATION.md。

许可证

MIT License - Copyright (c) Corsair Memory Inc.

详见 LICENSE。

链接

Available Tools

1 tool
bridge_statusElgato Bridge StatusA
Read-only

Reports which Elgato apps (e.g. Stream Deck) are currently connected to the bridge and how many of their tools are available. Use this to check whether the Stream Deck app is reachable when expected tools are missing.

Inputs: no named parameters.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A4.2/5.0

    Scored across 1 tool

    Disambiguation5/5

    With only one tool, there is no possibility of confusion or overlap. The single tool has a clear, distinct purpose.

    Naming Consistency5/5

    A single tool named 'bridge_status' follows a clear noun-based naming convention. While there is no pattern to compare, the name is descriptive and internally consistent.

    Tool Count3/5

    The server exposes only one tool, which feels thin for an Elgato MCP server. It's borderline acceptable for a minimal diagnostic utility, but lacks the breadth expected of a full-featured integration.

    Completeness1/5

    The tool only checks bridge status, implying the existence of other tools that are missing. For an Elgato server, this is severely incomplete—no device control, app management, or other operational tools are present, leaving the domain mostly uncovered.

    Maintenance

    ActivitySlowing
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      Bridges AI assistants to Linux D-Bus for system interaction, enabling clipboard, notifications, screenshots, and service discovery with configurable security levels.
      4
      MIT
    • F
      license
      Not graded
      quality
      C
      maintenance
      Enables AI agents to discover and interact with iOS apps through a local MCP gateway, converting remote Streamable HTTP MCP endpoints into stdio tools. Provides dynamic device discovery, tool schema introspection, and deterministic tool calling for app analysis.
      -
    • A
      license
      B
      quality
      D
      maintenance
      Enables AI assistants to control and automate OBS Studio via natural language, covering scenes, sources, audio, recording, streaming, transitions, filters, media playback, diagnostics, and multi-step workflows over the OBS WebSocket protocol.
      89
      MIT
    • A
      license
      Not graded
      quality
      D
      maintenance
      Enables AI assistants to list and read existing ChatGPT conversations and send messages into them, retrieving the assistant's reply through ChatGPT Desktop's native bridge. It works without cookies, UI automation, or an OpenAI API key.
      1
      Apache 2.0