Shell Server
A lightweight MCP server that provides AI assistants with access to a system's terminal through a secure terminal tool. It enables users to execute shell commands and receive stdout, stderr, and exit codes directly within an MCP-compatible client.
README
<div align="center">
Shell Server
A lightweight MCP server that gives AI assistants access to your terminal.
Run shell commands through the Model Context Protocol — connect any MCP-compatible AI client to your system's terminal.
</div>
Features
- Single tool, full power — exposes a
terminaltool that runs any shell command - Stdout + stderr — returns combined output with clear labeling
- Timeout protection — commands are capped at 30 seconds
- Error reporting — non-zero exit codes are surfaced automatically
- Stdio transport — works with any MCP client out of the box
Quickstart
With uv (recommended)
# Clone the repo
git clone https://github.com/joandiazcapell/shellserver.git
cd shellserver
# Install dependencies and run
uv sync
uv run server.py
With Docker
# Build the image
docker build -t shellserver .
# Run the container
docker run --rm -i shellserver
Or pull directly from Docker Hub:
docker run --rm -i yourusername/shellserver:latest
MCP Client Configuration
Add this to your MCP client config to connect:
Local (uv)
{
"mcpServers": {
"shell": {
"command": "uv",
"args": ["run", "server.py"],
"cwd": "/path/to/shellserver"
}
}
}
Docker
{
"mcpServers": {
"shell": {
"command": "docker",
"args": ["run", "--rm", "-i", "shellserver"]
}
}
}
Tool Reference
terminal
Run a shell command and return its output.
| Parameter | Type | Description |
|---|---|---|
command |
string |
The shell command to execute |
Returns — stdout, stderr (if any), and exit code (if non-zero).
> terminal("echo hello world")
hello world
> terminal("ls nonexistent")
STDERR:
ls: nonexistent: No such file or directory
Exit code: 1
Project Structure
shellserver/
├── server.py # MCP server implementation
├── pyproject.toml # Project metadata & dependencies
├── uv.lock # Locked dependencies
├── Dockerfile # Container build file
└── README.md
Requirements
- Python 3.13+
- uv (for local development)
- Docker (for containerized usage)
<div align="center">
</div>
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。