A PowerShell-based utility for AI agents to programmatically control mouse movements, clicks, and keyboard inputs on Windows systems.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install win-control
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install win-control using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Windows Mouse and Keyboard Control?
The Windows Mouse and Keyboard Control skill allows developers and AI agents to interact directly with the Windows operating system graphical user interface. By utilizing a set of specialized PowerShell scripts, this Openclaw Skills entry enables the simulation of physical user inputs such as clicking specific screen coordinates or typing text and shortcuts. This is particularly useful for interacting with applications that lack a native API or CLI.
Built for reliability and speed, the skill bypasses standard execution policies to ensure smooth integration into automated workflows. It serves as a foundational component for building desktop-aware agents within the Openclaw Skills framework, allowing for seamless bridge between LLM reasoning and OS-level execution.
Windows Mouse and Keyboard Control Use Cases
- Automating data entry in legacy desktop software where no API is available.
- Executing complex keyboard shortcuts like ALT+TAB or WIN for window management.
- Programmatic GUI testing and quality assurance automation for Windows applications.
- Orchestrating multi-step desktop workflows that require precise mouse interaction.
How Windows Mouse and Keyboard Control Works
- The AI agent or user identifies the required GUI action, such as a coordinate-based click or a string of text.
- The skill invokes a dedicated PowerShell script (e.g., mouse-click.ps1 or send-keys.ps1) with the necessary parameters.
- The script executes using the Bypass execution policy to ensure it can interact with the Windows input stream without interruption.
- The Windows OS translates the script logic into actual mouse movements or virtual keypresses as if performed by a human user.
Windows Mouse and Keyboard Control Setup
Ensure you are running on a Windows environment with PowerShell available. The scripts are located within the skill's subdirectory. To perform a mouse click, execute the following command:
powershell -ExecutionPolicy Bypass -File skills/win-control/scripts/mouse-click.ps1 -X 500 -Y 300 -ClickType Left
To send text or trigger specific keyboard shortcuts, use the send-keys script:
powershell -ExecutionPolicy Bypass -File skills/win-control/scripts/send-keys.ps1 -Text "Hello World"
Windows Mouse and Keyboard Control Data Schema & Taxonomy
| Component | Parameter | Type | Description |
|---|---|---|---|
| Mouse Control | -X, -Y | Integer | Target screen coordinates in pixels |
| Mouse Control | -ClickType | String | Supports Left or Right button clicks |
| Keyboard Input | -Text | String | Plain text string to be typed into the active window |
| Keyboard Input | -Keys | String | Special keys and combinations (e.g., CTRL+C, ALT+TAB, F1-F12) |
Windows Mouse and Keyboard Control Advanced Features
- Support for complex macro-style shortcuts including system-level keys like WIN, ALT, and function keys.
- Headless-friendly execution utilizing PowerShell policy bypassing for automated environments.
- Full compatibility with the Openclaw Skills automation ecosystem for multi-agent workflows.
- Precise coordinate-based mouse control compatible with various display resolutions.
SKILL.md
Loading
Related Openclaw Skills
Agentforce Persona Designer
A comprehensive framework for designing, scoring, and encoding high-fidelity AI agent personas for Salesforce Agentforce.
pilot-discover
A high-performance discovery tool for locating and filtering peers within the Pilot Protocol network based on capabilities and reliability.
OpenDexter
An automated marketplace interface for finding and consuming pay-per-call x402 APIs across multiple blockchain networks.
sf-ai-agentforce
A professional development skill for managing Agentforce Builder metadata, declarative topics, and AI-driven actions within the Salesforce ecosystem.
Pilot Discord Bridge
A bidirectional integration bridge that enables Pilot Protocol agents to send notifications to Discord and process incoming messages as event streams.
pilot-dataset
A specialized skill for exchanging structured datasets with automated schema negotiation and metadata tracking over the Pilot Protocol.
Featured*
