GetScreen MCP Server
Enables capturing screenshots of all connected monitors with automatic JPEG compression and multi-platform support including WSL environments. Supports quality adjustment and image resizing through simple natural language commands.
README
GetScreen MCP Server
A lightweight Model Context Protocol (MCP) Server for rapid capture of screenshots from all available monitors with automatic compression.
Features
- Multi-Monitor Support: Automatically captures all connected displays
- WSL Support: Works seamlessly in Windows WSL through PowerShell integration
- Fast Compression: Utilizes Sharp for optimal JPEG compression
- Simple Integration: Single tool with optional parameters
- Lightweight: Minimal dependencies, maximum performance
- Cross-Platform: Linux, macOS, and Windows (including WSL)
New Features
Enhanced WSL Support
WSL support has been improved with automatic detection and PowerShell integration. The tool now works seamlessly whether you use the wsl command approach or the simpler Node-based direct execution method.
Key Features:
- Automatic WSL detection via
/proc/versionenvironment inspection - Seamless PowerShell integration for capturing Windows host displays
- Support for both direct Node execution and WSL command wrapper
- No additional configuration needed for WSL environment detection
How It Works: When running in WSL, the tool automatically detects the environment and leverages PowerShell on the Windows host to capture screenshots. This approach ensures access to all display information from the Windows side while maintaining full compatibility with WSL-based tools.
Installation
git clone https://github.com/Kirchlive/getscreen_mcp.git
cd getscreen_mcp
npm install
npm run build
Configuration
Claude Code
Add the following to your Claude configuration:
macOS:
~/Users/USERNAME/.claude.json
Windows:
C:\Users\USERNAME\.claude.json
Linux/WSL:
~/home/USERNAME/.claude.json
{
"mcpServers": {
"getscreen": {
"command": "node",
"args": ["/ABSOLUT/PATH/TO/getscreen_mcp/build/index.js"]
}
}
}
Usage
Screenshot Tool
The screenshot tool captures all available monitors and returns the compressed images.
Parameters (all optional):
quality(number, 1-100): JPEG quality. Default: 80maxWidth(number): Maximum width in pixels. Images are resized proportionally.
Examples:
Take a screenshot
Take a screenshot with high quality (quality: 95)
Take a screenshot and resize it to max. 1920px width (maxWidth: 1920)
Technical Details
Platform Detection
The tool automatically detects the environment and selects the optimal method:
- WSL: Uses PowerShell on the Windows host via
powershell.exe - Linux/macOS: Uses
screenshot-desktopfor native capture
Components
- WSL Capture: PowerShell System.Windows.Forms and System.Drawing
- Native Capture:
screenshot-desktopfor Linux/macOS - Compression:
sharpfor fast image processing and JPEG compression - Format: Images are returned as base64-encoded JPEGs
- Performance: Parallel capture of all monitors for maximum speed
WSL Implementation
In WSL environments:
- Detection via
/proc/version(checks for "microsoft" or "WSL") - PowerShell script is executed on Windows host
- All monitors are captured via
System.Windows.Forms.Screen::AllScreens - Screenshots captured as PNG, then compressed to JPEG with desired quality
- Base64-encoded transmission back to WSL
License
MIT
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。