CursorTouch/Windows-MCP

MCP Server for Computer Use in Windows

View on GitHub ↗Jump to charts ↓Open shareable report

Summary Information

Updated 17 minutes ago
Added to GitGenius on September 9th, 2026
Created on May 13th, 2025
Open Issues & Pull Requests: 21 (+0)
GitHub issues: Enabled
Number of forks: 837
Total Stargazers: 6,983 (+0)
Total Subscribers: 42 (+0)

Repository Insights (GitGenius)

Median issue/PR response: 3.6 hours
Mean response time: 2.3 days
90th percentile: 7.5 days
Tracked items: 145

How this project is maintained

Around half of the issues opened in the past year never receive a reply. Only 17% of issues opened in the past year have been closed. Three people close 87% of everything that gets resolved.

Charts & Analytics

Fetching additional details & charts...

Issue Activity (beta)

Open issues: 16
New in 7 days: 4
Closed in 7 days: 0
Avg open age: 53 days
Stale 30+ days: 5
Stale 90+ days: 1

Recent activity

Opened in 7 days: 4
Closed in 7 days: 0
Comments in 7 days: 1
Events in 7 days: 1

Top labels

  • enhancement (2)

Detailed Description

Windows-MCP is an MCP server that enables AI agents to control and interact with Windows operating systems.

The tool solves the problem of connecting large language models to Windows desktop environments by acting as a bridge that translates AI instructions into native Windows operations. Rather than relying on computer vision or specialized models, it works with any LLM by providing a rich toolset for UI automation, including keyboard and mouse control, window management, application launching, and UI state capture. This approach reduces setup complexity and allows agents to perform file navigation, application control, UI interaction, and quality assurance testing directly on Windows systems.

Windows-MCP suits developers building AI automation tools, QA testing frameworks, or desktop agents that need to interact with Windows applications and UI elements. It works across Windows 7 through Windows 11 and requires minimal dependencies, making it accessible for projects of varying scale. The tool distinguishes itself by not requiring vision-based techniques or fine-tuned models, instead leveraging the reasoning capabilities of any LLM to understand and execute desktop tasks. Real-time interaction latency typically ranges from 0.2 to 0.5 seconds between actions.

The project maintains active engagement with its user community through Discord and social channels. Development activity shows consistent updates and improvements to the codebase. The tool has been integrated into the official MCP Registry and is distributed through PyPI, indicating ongoing maintenance and support for standard Python packaging workflows.