Windows Screenshot MCP Server
Professional Windows screenshot capture server with Model Context Protocol (MCP) integration, real-time WebSocket streaming, Chrome tab capture, and advanced window targeting.
接入你的智能体
从发布者处安装。不是 PlugMCP 端点。
远程 URL 还是本机命令? 两种安装方式的区别
来自发布者
Professional Windows screenshot capture server with Model Context Protocol (MCP) integration, real-time WebSocket streaming, Chrome tab capture, and advanced window targeting.
🚀 Production-ready Windows Screenshot MCP Server with real-time WebSocket streaming, Chrome DevTools integration, and comprehensive APIs. Built with Go for high-performance native Windows screenshot automation.
A production-ready Go-based screenshot server that provides both REST API and MCP protocol support for capturing Windows application screenshots. Designed for automation, testing, and AI agent integration with advanced features like real-time streaming and hidden window capture.
# Download latest release
curl -L https://github.com/your-org/screenshot-mcp-server/releases/latest/download/screenshot-server.exe -o screenshot-server.exe
# Or build from source
git clone https://github.com/your-org/screenshot-mcp-server.git
cd screenshot-mcp-server
go build -o screenshot-server.exe ./cmd/server
# Start the server
./screenshot-server.exe --port 8080
# Health check
curl http://localhost:8080/health
# Basic window capture
curl "http://localhost:8080/api/screenshot?method=title&target=Notepad" -o notepad.png
# Full desktop capture
curl "http://localhost:8080/api/screenshot?method=desktop&monitor=0" -o desktop.png
GET /health
Returns server status and version information.
GET /api/screenshot
GET /v1/screenshot
Parameters:
method (required): title, pid, handle, classtarget (required): Window identifier (title, PID, handle, class name)format: png, jpeg, bmp, webp (default: png)quality: 1-100 for lossy formats (default: 95)cursor: true/false to include mouse cursorExamples:
# Window by title
curl "http://localhost:8080/api/screenshot?method=title&target=Calculator" -o calc.png
# Window by PID
curl "http://localhost:8080/api/screenshot?method=pid&target=1234&format=jpeg&quality=80" -o app.jpg
# Window by class name
curl "http://localhost:8080/api/screenshot?method=class&target=Notepad&cursor=true" -o notepad.png
GET /v1/chrome/instances # List Chrome instances
GET /v1/chrome/tabs # List all Chrome tabs
POST /v1/chrome/tabs/:id/screenshot # Capture specific tab
Connect to ws://localhost:8080/stream/{windowId} for real-time streaming.
Query Parameters:
fps: Frames per second (1-60, default: 10)quality: Compression quality (10-100, default: 75)format: jpeg or png (default: jpeg)Client Example:
<!DOCTYPE html>
<html>
<body>
<img id="stream" style="max-width: 100%;">
<script>
const ws = new WebSocket('ws://localhost:8080/stream/0?fps=15&quality=75&format=jpeg');
ws.onmessage = function(event) {
const data = JSON.parse(event.data);
if (data.type === 'frame') {
document.getElementById('stream').src = 'data:image/jpeg;base64,' + data.image;
}
};
</script>
</body>
</html>
The server supports MCP JSON-RPC 2.0 requests via POST /rpc.
Available Methods:
screenshot.capture - Capture screenshotswindow.list - List windows (placeholder)chrome.instances - List Chrome instanceschrome.tabs - List Chrome tabschrome.tabCapture - Capture Chrome tabstream.status - Get streaming statusExample MCP Request:
{
"jsonrpc": "2.0",
"method": "screenshot.capture",
"params": {
"method": "title",
"target": "Calculator",
"format": "png"
},
"id": 1
}
The server can be configured via environment variables or command-line flags:
# Start with custom port
./server.exe --port 9090
# Start with custom host
./server.exe --host 0.0.0.0 --port 8080
# Environment variables
export SCREENSHOT_PORT=8080
export SCREENSHOT_HOST=localhost
./server.exe
The server uses a default configuration that can be customized:
// Default settings
type Config struct {
Port int // Default: 8080
Host string // Default: "localhost"
DefaultFormat string // Default: "png"
Quality int // Default: 95
IncludeCursor bool // Default: false
LogLevel string // Default: "info"
ChromeTimeout string // Default: "30s"
StreamMaxSessions int // Default: 10
StreamDefaultFPS int // Default: 10
}
For Chrome tab capture, launch Chrome with debugging enabled:
# Windows
"C:\Program Files\Google\Chrome\Application\chrome.exe" --remote-debugging-port=9222
# Launch with temporary profile
chrome.exe --remote-debugging-port=9222 --user-data-dir=temp-profile
# Clone the repository
git clone https://github.com/your-org/screenshot-mcp-server.git
cd screenshot-mcp-server
# Install dependencies
go mod download
# Build the server
go build -o server.exe ./cmd/server
# Run tests
go test ./...
# Start the server
./server.exe
├── cmd/
│ ├── server/ # Main server application
│ └── mcpctl/ # MCP control utility
├── internal/
│ ├── screenshot/ # Screenshot capture engines
│ ├── chrome/ # Chrome DevTools integration
│ ├── window/ # Window management
│ └── ws/ # WebSocket streaming
├── pkg/
│ └── types/ # Shared data structures
└── examples/ # Usage examples and documentation
The server follows a modular architecture:
This project is licensed under the MIT License - see the LICENSE file for details.
/examples directory for usage examplesA powerful Windows screenshot server built for modern automation and AI integration.
🚀 Production-ready Windows Screenshot MCP Server with real-time WebSocket streaming, Chrome DevTools integration, and comprehensive APIs. Built with Go for high-performance native Windows screenshot
概览
🚀 Production-ready Windows Screenshot MCP Server with real-time WebSocket streaming, Chrome DevTools integration, and comprehensive APIs. Built with Go for high-performance native Windows screenshot automation.
工具
暂无工具列表。远程服务器在首次健康检查后会获得工具列表。
常见问题
在本页复制适用于 Cursor、Claude Desktop、Claude Code 或 VS Code 的 stdio 配置。它使用发布者的安装方式,而不是 PlugMCP 代理。
此记录未列出必需的环境变量。
Windows Screenshot 通过 stdio 在本地运行。
上次检查时间:2026-10-28。 🚀 Production-ready Windows Screenshot MCP Server with real-time WebSocket streaming, Chrome DevTools integration, and comprehensive APIs. Built with Go for high-performance native Windows screenshot automation.