Discover the toolingbehind modern intelligence
The operating system for AI infrastructure discovery. Search a live map of MCP servers, AI agents, LLM tools, automation systems, and developer infrastructure.
Windows-MCP
MCP Server for Computer Use in Windows
5.9kstars
741forks
Python
aidesktopmcptoolswindowswindows-automation
Quality signals
ActiveNo testsNo CIThin docsNo licensePopular
AI summary
Auto-generated overview of what this skill does and when to reach for it.
MCP server that lets AI agents control Windows like a human—open apps, click buttons, type, and automate UI tasks.
Use this when
- Need an AI agent to open Notepad and type text
- Want to automate repetitive Windows GUI tasks without scripting
- Testing a Windows app by simulating real user interactions
- Running browser automation inside Chrome/Edge/Firefox
- Integrating Windows tools into an LLM workflow
Don't use when
- Running on non-Windows systems (Linux/macOS)
- Need low-level system calls or kernel access
- Requiring real-time performance-critical automation
Capabilities
- Opens and controls Windows applications via UI automation
- Simulates keyboard and mouse input with sub-second latency
- Captures window and UI element states for inspection
- Supports DOM-based browser automation in Chrome/Edge/Firefox
- Works with any LLM (no vision or fine-tuning required)
- Runs as a lightweight MCP server with minimal dependencies
InstallMCPRun via uvx (Python)
Universal installANY TOOL
$ npx skillcity@latest CursorTouch/Windows-MCPAuto-detects the skill or MCP server in this repo and installs it for Claude Code, Codex, Cursor or Cline.
Requires Node.js 18+ (ships with npx).
Or install manually
uvx windows-mcp
MCP Configuration
Add this to your claude_desktop_config.json
{
"mcpServers": {
"windows-mcp": {
"command": "uvx",
"args": [
"windows-mcp"
]
}
}
}Loading README…
Ask a simulated agent to run this skill. It loads the real SKILL.md and reasons over your task with NVIDIA.
windows-mcp — sandboxAI sandbox
Pick a quick prompt below or type your own — watch the agent execute this skill step by step.
guest@skillcity:~$Try a prompt