Prevent screen savers, system sleep, and keep your macOS awake during AI processing tasks. jigglypuff is a lightweight mouse jiggler and activity manager that uses imperceptible cursor movements to simulate user activity. Perfect for developers using AI agents like Cursor, Qoder and Claude Desktop.

Keywords: mouse jiggler, cursor mover, screen saver preventer, system wakefulness, AI automation, macOS automation, MCP tools, Model Context Protocol, keep screen awake, prevent sleep, automation tools, developer utilities

Version: 1.0.0

• Prevents screen savers and system sleep during AI processing tasks
• Allows system to sleep when waiting for user input to conserve energy
• Controlled via MCP tools for seamless integration with AI agents
• Lightweight and efficient implementation with minimal system impact
• Small, imperceptible mouse movements that won't interfere with your work
• Automatic jiggling rules - Automatically enables/disables jiggling based on task status
• Cross-AI agent compatibility - Works with Cursor, Qoder, Claude Desktop, and other MCP-compatible tools

• macOS system (10.15+ recommended)
• Homebrew package manager
• Python 3.11+
• cliclick CLI tool for mouse automation
• MCP-compatible AI agent (Qoder, Claude Desktop, Cursor, etc.)

1. Clone the repository:

   git clone https://github.com/trose/jigglypuff.git
   cd jigglypuff

2. Install dependencies:

   # Install cliclick using Homebrew
   brew install cliclick
   
   # Create a virtual environment
   python3 -m venv venv
   source venv/bin/activate
   
   # Install Python MCP SDK
   pip install mcp

3. Configure accessibility permissions:

   • Open System Preferences > Security & Privacy > Privacy
   • Select "Accessibility" from the left sidebar
   • Click the lock icon and authenticate
   • Add your terminal application (Terminal.app, iTerm2, etc.) to the list

4. Make scripts executable:

   chmod +x jiggly_puff.sh mcp_server.py

To start the MCP server that exposes the jigglypuff tools:

source venv/bin/activate
python mcp_server.py

The server will listen on stdio transport for MCP-compatible clients.

The server exposes several tools for AI agent integration:

1. wake_up_jiggly - Initiates the cursor jiggling process

   • interval: Time between jiggles in seconds (5-300, default: 30)
   • offset: Mouse movement offset in pixels (1-10, default: 1)
   • Returns: Status message with PID and settings

2. put_jiggly_to_sleep - Terminates the running cursor jiggling process

   • Returns: Confirmation message with PID

3. check_jiggly_status - Checks current cursor jiggling state

   • Returns: Current status (jiggling, sleeping, or stopped)

4. enable_jiggling_before_tasks - Implements the rule: ALWAYS use jigglypuff MCP to enable jiggling before beginning tasks

   • Automatically starts jiggling with default settings

5. disable_jiggling_after_tasks - Implements the rule: ALWAYS disable jiggling when task complete

   • Automatically stops jiggling

The server provides helpful prompts for user interaction:

1. jigglypuff_help - Comprehensive usage guide and best practices
2. jigglypuff_troubleshooting - Solutions for common issues and problems

The server exposes resources for context data management:

1. jigglypuff-config - Current configuration and system status (JSON format)
2. jigglypuff-rules - Official usage rules and best practices (JSON format)

Example configuration:

{
  "mcpServers": {
    "jigglypuff": {
      "command": "/Users/trose/src/mouse-jiggler/venv/bin/python",
      "args": ["mcp_server.py"],
      "cwd": "/Users/trose/src/mouse-jiggler"
    }
  }
}

Run the provided test suite to verify functionality:

source venv/bin/activate

# Run individual test files
python tests/test_functions.py
python tests/test_comprehensive.py

# Or run all tests (if you have pytest installed)
pytest tests/

This MCP server is registered with LobeHub and includes:

• ✅ MCP Tools: 5 tools for mouse jiggling control
• ✅ MCP Prompts: 2 prompts for help and troubleshooting
• ✅ MCP Resources: 2 resources for configuration and rules
• ✅ Installation Methods: Multiple deployment options
• ✅ README Documentation: Comprehensive usage guide
• ✅ MIT License: Open source licensing

Contributions are welcome! Please feel free to submit a Pull Request.

This project is licensed under the MIT License - see the LICENSE file for details.