kitfactory/pymcpautogui

Developer tools

このページは AI により自動翻訳されています。技術的な詳細は原文をご確認ください。原文を見る

23 stars 4 forks 品質 62 トレンド 10MIT

Repository Install

GUI manipulation MCP server

概要

✨ PyMCPAutoGUI provides a bridge between your AI agents (like those in Cursor or other MCP-compatible environments) and your computer's graphical user interface (GUI). It allows your agent to see the screen 👁️, control the mouse 🖱️ and keyboard ⌨️, and interact with windows 🪟, just like a human user! Stop tedious manual GUI tasks and let your AI do the heavy lifting 💪. Perfect for automating repetitive actions, testing GUIs, or building powerful AI assistants 🤖. * Give your AI agents the power to interact directly with desktop applications. * Works seamlessly with MCP-compatible clients like the Cursor editor. It's plug and play! * Get started with a simple server command. Seriously, it's that easy. * Offers a wide range of GUI automation functions from the battle-tested PyAutoGUI and PyGetWindow. * Includes tools for taking screenshots and locating images on the screen – let your agent see! * Control window position, size, state (minimize, maximize), and more.

README

以下は原文 README の抜粋です。原文を見る。

PyMCPAutoGUI 🖱️⌨️🖼️ - GUI Automation via MCP

Supercharge your AI Agent’s capabilities! ✨ PyMCPAutoGUI provides a bridge between your AI agents (like those in Cursor or other MCP-compatible environments) and your computer’s graphical user interface (GUI). It allows your agent to see the screen 👁️, control the mouse 🖱️ and keyboard ⌨️, and interact with windows 🪟, just like a human user!

Stop tedious manual GUI tasks and let your AI do the heavy lifting 💪. Perfect for automating repetitive actions, testing GUIs, or building powerful AI assistants 🤖.

🤔 Why Choose PyMCPAutoGUI?

🤖 Empower Your Agents: Give your AI agents the power to interact directly with desktop applications.
✅ Simple Integration: Works seamlessly with MCP-compatible clients like the Cursor editor. It’s plug and play!
🚀 Easy to Use: Get started with a simple server command. Seriously, it’s that easy.
🖱️⌨️ Comprehensive Control: Offers a wide range of GUI automation functions from the battle-tested PyAutoGUI and PyGetWindow.
🖼️ Screen Perception: Includes tools for taking screenshots and locating images on the screen – let your agent see!
🪟 Window Management: Control window position, size, state (minimize, maximize), and more. Tidy up that desktop!
💬 User Interaction: Display alert, confirmation, and prompt boxes to communicate with the user.

🛠️ Supported Environments

Operating Systems: Windows, macOS, Linux (Requires appropriate dependencies for pyautogui on each OS)
Python: 3.11+ 🐍
MCP Clients: Cursor Editor, any client supporting the Model Context Protocol (MCP)

🚀 Getting Started - It’s Super Easy!

1. Installation (Recommended: Use a Virtual Environment!)

Using a virtual environment keeps your project dependencies tidy.

# Create and activate a virtual environment (example using venv)
python -m venv .venv
# Windows PowerShell
.venv\Scripts\Activate.ps1
# macOS / Linux bash
source .venv/bin/activate

# Install using pip (from PyPI or local source)
# Make sure your virtual environment is active!
pip install pymcpautogui # Or pip install . if installing from local source

(Note: pyautogui might have system dependencies like scrot on Linux for screenshots. Please check the pyautogui documentation for OS-specific installation requirements.)

2. Running the MCP Server

Once installed, simply run the server from your terminal:

# Make sure your virtual environment is activated!
python -m pymcpautogui.server

The server will start and listen for connections (defaulting to port 6789). Look for this output:

GitHub で README 全文を見る

インストール

This server does not publish a one-line install command.

Open the repository installation guide MCP 初心者？mcpskill.org のガイドから始めましょう →

設定

{
        "mcpServers": {
            // ... other MCP server configs if any ...
            "PyMCPAutoGUI": {
                // Sets the working directory. ${workspaceFolder} is usually correct.
                "cwd": "${workspaceFolder}",

                // Command to run Python. 'python' works if the venv is active in the terminal
                // where Cursor was launched, or specify the full path.
                "command": "python", // Or ".venv/Scripts/python.exe" (Win) or ".venv/bin/python" (Mac/Linux)

                // Arguments to start the server module.
                "args": ["-m", "pymcpautogui.server"]
            }
            // ... other MCP server configs if any ...
        }
    }