cloud-command
Enables LLMs to create, destroy, and manage AWS EC2 instances, including running commands via interactive shells.
README
Cloud Command
Cloud Command is a web-based GUI for creating and managing multiple AWS EC2 instances. It takes roughly a minute and 30 seconds to create an instance and start the first interactive shell.
The built-in MCP server allows LLMs to connect to Cloud Command to create/destroy EC2s and run commands on them.


Features
- Create and destroy EC2 instances
- Interactive terminal via the Web UI
- File upload/download via SFTP
- MCP server for AI agent integration
Setup
Cloud Command requires that the aws cli be installed and connected to an AWS account.
Args
usage: cloud_command [-h] [--version] [--host HOST] [--port PORT]
[--private-key-path PRIVATE_KEY_PATH]
[--public-key-path PUBLIC_KEY_PATH]
[--list-all] [--delete-all]
GUI for commanding multiple AWS assets
options:
-h, --help show this help message and exit
--version show program's version number and exit
uvicorn options:
--host HOST Uvicorn server IP, default: 127.0.0.1
--port PORT Uvicorn server port, default: 8080
--private-key-path PRIVATE_KEY_PATH
Path to SSL private key file
--public-key-path PUBLIC_KEY_PATH
Path to SSL certificate file
management options:
--list-all Lists all EC2s with a 'cloud_command' tag
--delete-all Destroys all EC2s with a 'cloud_command' tag
MCP integration
The MCP server is hosted at /api/mcp, these are the configs for claude:
{
"mcpServers": {
"cloud-command": {
"type": "http",
"url": "https://127.0.0.1:8080/api/mcp/"
}
}
}
NOTE: you may need to disable SSL verification when using self-signed certificates
Deployment
Linux: Download the .bin file, make it executable, then move it to /usr/bin/cloud_command to add it to your PATH:
chmod +x cloud_command.bin
sudo mv cloud_command.bin /usr/bin/cloud_command
Windows: Download the .exe file, move it to a permanent location (e.g. %APPDATA%\CloudCommand\), then add that directory to your system PATH:
- Open Settings → System → About → Advanced system settings
- Click Environment Variables
- Under System variables, select Path and click Edit
- Click New and add the path to the directory containing
cloud_command.exe - Click OK to save
Build from source
# Create virtual environment
python3 -m venv .venv
# Activate (Linux)
source .venv/bin/activate
# Activate (Windows)
.venv/Scripts/activate.ps1
# Install dependencies
pip install .[dev]
# Run normally
python3 cloud_command/runner.py
# Build standalone executable
python3 build.py --help
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。