feat: add 9 advanced automation tools with DPI coordinate system support

README.md

@@ -399,6 +399,27 @@ MCP Client can access the following tools to interact with Windows:
 - `Process`: List running processes or terminate them by PID or name.
 - `Notification`: Send a Windows toast notification with a title and message.
 - `Registry`: Read, write, delete, or list Windows Registry values and keys.
+- `CursorPosition`: Get the current mouse cursor (x, y) coordinates.
+- `PixelColor`: Get the RGB color value at screen coordinates with hex code and color name.
+- `KeyHold`: Press or release keyboard keys independently for hold operations (e.g., hold Shift while clicking).
+- `ScreenInfo`: Get information about all connected monitors (resolution, position, primary).
+- `ScreenHighlight`: Highlight a screen region with a colored border for visual debugging.
+- `MousePath`: Move the mouse cursor smoothly through a series of waypoints.
+- `ScreenReader`: Read text from a screen region using OCR (Windows built-in or pytesseract fallback).
+- `WaitForChange`: Wait until a screen region visually changes beyond a threshold.
+- `FindImage`: Find a template image on screen using visual template matching (requires `pip install 'windows-mcp[vision]'`).
+- `VolumeControl`: Get/set system volume (0-100), mute/unmute/toggle via Windows Core Audio COM API.
+- `BrightnessControl`: Get/set display brightness (0-100) via WMI.
+- `AppList`: List all running applications, check if a specific app is running, or force-quit by name.
+- `Dialog`: Display message boxes, input prompts, or file/folder picker dialogs.
+- `SystemInfoExtended`: Detailed system info — OS version, CPU, RAM, disk, battery, network, uptime.
+- `DarkMode`: Get or toggle Windows dark/light app mode via registry.
+- `SayText`: Text-to-speech using Windows SAPI with optional voice and rate settings.
+- `PortCheck`: Check if a TCP/UDP port is in use, find the owning process, or list all listening ports.
+- `FileWatcher`: Watch a file or directory for changes (create, modify, delete) with timeout.
+- `SearchFiles`: Search for files by name pattern or content text within a directory tree.
+- `NetworkDiagnostics`: Ping hosts, DNS lookup, trace routes, check connectivity.
+- `AccessibilityInspector`: Inspect UI element accessibility properties at coordinates or for the focused element.
 
 ## 🤝 Connect with Us
 Stay updated and join our community:

pyproject.toml

@@ -43,6 +43,18 @@ dev = [
     "pytest>=8.0.0",
     "pytest-asyncio>=0.24.0",
 ]
+vision = [
+    "opencv-python-headless>=4.8.0",
+    "numpy>=1.26.0",
+]
+ocr = [
+    "pytesseract>=0.3.10",
+]
+all = [
+    "opencv-python-headless>=4.8.0",
+    "numpy>=1.26.0",
+    "pytesseract>=0.3.10",
+]
 
 [project.urls]
 Homepage = "https://github.com/CursorTouch"