Skip to main content

Model Context Protocol Server via Cross-Platform (Mac & Windows) Split Screen Functions

Project description

Computer Split Screen MCP

Cross-platform MCP server that exposes split-screen tools (halves, quadrants, thirds) plus maximize/minimize for both Windows and macOS.
Works with MCP clients via uvx.

Features

  • Cross-platform support: Windows and macOS
  • Split-screen layouts: Halves, quadrants, thirds, and two-thirds variations
  • Window controls: Maximize and minimize
  • MCP integration: Full Model Context Protocol server support
  • Native performance: Uses PyObjC on macOS and pywin32 on Windows for optimal performance
  • No external scripts: Pure Python implementation with native OS APIs

Install / Run via MCP client

Configure your MCP client:

{
  "mcpServers": {
    "computer-split-screen": {
      "command": "uvx",
      "args": ["computer-split-screen-mcp"],
      "env": {}
    }
  }
}

Available Tools

  • left-half - Snap current window to left half
  • right-half - Snap current window to right half
  • top-half - Snap current window to top half
  • bottom-half - Snap current window to bottom half
  • top-left - Top-left quadrant
  • top-right - Top-right quadrant
  • bottom-left - Bottom-left quadrant
  • bottom-right - Bottom-right quadrant
  • left-third - Left third (1/3)
  • middle-third - Middle third (1/3)
  • right-third - Right third (1/3)
  • left-two-thirds - Left two-thirds (2/3)
  • right-two-thirds - Right two-thirds (2/3)
  • maximize - OS maximize (bordered)
  • fullscreen - Fullscreen (no borders via macOS)(bordered via Windows)
  • minimize - Minimize window

Platform Dependencies

  • Windows: Requires pywin32>=306
  • macOS: Requires pyobjc-framework-Cocoa>=10.0, pyobjc-framework-Quartz>=10.0, and pyobjc-framework-ApplicationServices>=10.0

What's New in v1.3.0

  • Complete macOS rewrite: Replaced AppleScript with native PyObjC implementation
  • Improved performance: Faster window operations on macOS using Accessibility APIs
  • Better reliability: More robust window detection and positioning
  • Enhanced error handling: Graceful fallbacks when platform-specific features are unavailable
  • Multi-monitor support: Improved handling of complex display configurations

Technical Details

macOS Implementation

  • Uses PyObjC frameworks for native macOS integration
  • Leverages Accessibility APIs (AX) for precise window manipulation
  • Handles coordinate system conversion between macOS and AX coordinate systems
  • Respects macOS-specific features like Dock and menu bar positioning
  • Supports both standard windows and document windows

Windows Implementation

  • Uses pywin32 for Windows API access
  • Implements DPI awareness for high-DPI displays
  • Uses DWM (Desktop Window Manager) for accurate frame bounds
  • Handles window insets and shadows properly
  • Calculates positions as fractions of monitor work area

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

computer_split_screen_mcp-1.3.2.tar.gz (9.3 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

computer_split_screen_mcp-1.3.2-py3-none-any.whl (11.2 kB view details)

Uploaded Python 3

File details

Details for the file computer_split_screen_mcp-1.3.2.tar.gz.

File metadata

File hashes

Hashes for computer_split_screen_mcp-1.3.2.tar.gz
Algorithm Hash digest
SHA256 f5fd56d921fd2db0dc1e25bd5f1572115d4fc2eb8e9f6373397c868a35d88448
MD5 c41914d514b770247b18d986b5b2e75d
BLAKE2b-256 b006a7d856c7c0f245c5d3e26d292b41227f0680d9c579c9a907563d7e46b6fd

See more details on using hashes here.

File details

Details for the file computer_split_screen_mcp-1.3.2-py3-none-any.whl.

File metadata

File hashes

Hashes for computer_split_screen_mcp-1.3.2-py3-none-any.whl
Algorithm Hash digest
SHA256 a9b2b6245ad709d099194f6202b88e8a88bb0e378ea9e6ef100809dad53ce72c
MD5 53088fcb60872ce5f8c49d110629cd1e
BLAKE2b-256 a33c795954807fece08ba938db1e7a4ea2cf80d8a9f4ca070ede0944e6a14d1f

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page