ssh-mcp

为什么要用

核心特性

Run commands on remote Linux/Windows servers over SSH from any MCP-compatible AI client.
Maintained by tufantunc (community)
Written in TypeScript
467+ GitHub stars at last check
Open source on GitHub

实时演示

实际使用效果

ssh-mcp.replay ▶ 就绪

0/0

安装

选择你的客户端

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

{
  "mcpServers": {
    "ssh-mcp": {
      "command": "TODO",
      "args": [
        "See README: https://github.com/tufantunc/ssh-mcp"
      ]
    }
  }
}

打开 Claude Desktop → Settings → Developer → Edit Config。保存后重启应用。

~/.cursor/mcp.json · .cursor/mcp.json

{
  "mcpServers": {
    "ssh-mcp": {
      "command": "TODO",
      "args": [
        "See README: https://github.com/tufantunc/ssh-mcp"
      ]
    }
  }
}

Cursor 使用与 Claude Desktop 相同的 mcpServers 格式。项目级配置优先于全局。

VS Code → Cline → MCP Servers → Edit

{
  "mcpServers": {
    "ssh-mcp": {
      "command": "TODO",
      "args": [
        "See README: https://github.com/tufantunc/ssh-mcp"
      ]
    }
  }
}

点击 Cline 侧栏中的 MCP Servers 图标，然后选 "Edit Configuration"。

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "ssh-mcp": {
      "command": "TODO",
      "args": [
        "See README: https://github.com/tufantunc/ssh-mcp"
      ]
    }
  }
}

格式与 Claude Desktop 相同。重启 Windsurf 生效。

~/.continue/config.json

{
  "mcpServers": [
    {
      "name": "ssh-mcp",
      "command": "TODO",
      "args": [
        "See README: https://github.com/tufantunc/ssh-mcp"
      ]
    }
  ]
}

Continue 使用服务器对象数组，而非映射。

~/.config/zed/settings.json

{
  "context_servers": {
    "ssh-mcp": {
      "command": {
        "path": "TODO",
        "args": [
          "See README: https://github.com/tufantunc/ssh-mcp"
        ]
      }
    }
  }
}

加入 context_servers。Zed 保存后热重载。

claude mcp add ssh-mcp -- TODO 'See README: https://github.com/tufantunc/ssh-mcp'

一行命令搞定。用 claude mcp list 验证，claude mcp remove 卸载。

使用场景

实战用法： ssh-mcp

Use ssh-mcp for its primary workflow

👤 Developers using AI coding agents ⏱ ~10 min beginner

何时使用： You need the capability: Run commands on remote Linux/Windows servers over SSH from any MCP-compatible AI client.

前置条件

MCP server installed — See https://github.com/tufantunc/ssh-mcp for install instructions

步骤

Install

Install ssh-mcp following the README at https://github.com/tufantunc/ssh-mcp✓ 已复制

→ MCP server appears in your client's available tools/skills
Invoke

Use ssh-mcp to complete the task described in its docs.✓ 已复制

→ Agent calls the relevant tool/skill and returns a result

结果： Working integration with the capabilities advertised by the project.

注意事项

README is the source of truth; details here may lag upstream — Check https://github.com/tufantunc/ssh-mcp for the latest setup steps and tool list