windows-control

Automate Windows UI tasks via CLI commands with JSON outputs.

1|1|Updated Mar 18, 2026
One-click install
npx skills add https://github.com/xianmingyao/openclaw-CaySon --skill windows-control
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: windows-control
Source: https://github.com/xianmingyao/openclaw-CaySon/tree/main/skills/windows-control
Command: npx skills add https://github.com/xianmingyao/openclaw-CaySon --skill windows-control

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Windows UI automation often requires brittle scripts or manual steps. This skill provides a stable, CLI-driven interface to control mouse, keyboard, windows, UI capture, and system/file operations by wrapping jingmai-agent actions.

Core Features & Use Cases

  • Command-line control of Windows UI elements: mouse, keyboard, window management, and UI capture.
  • JSON-friendly outputs for AI agents and automation pipelines, plus an interactive REPL mode.
  • Architecture compatibility with CLI-Anything for deterministic task execution and extensibility.

Quick Start

Open Notepad and type a message using cli-windows-control.

Frequently Asked Questions about windows-control

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate Windows UI tasks from the command line?

You can automate Windows UI tasks from the command line by using a CLI harness that wraps jingmai-agent actions. This approach provides stable commands for mouse, keyboard, and window operations without brittle scripts.

Can I use command-line automation to control mouse and keyboard actions in production Windows environments?

Yes, command-line automation can control mouse and keyboard actions in production Windows environments. The CLI harness wraps jingmai-agent actions to ensure stable and deterministic execution of UI automation tasks.

Does Windows UI automation output JSON for AI agent integration?

Windows UI automation outputs JSON-friendly data for AI agent integration. This allows automation pipelines to consume the results of mouse, keyboard, window, and system operations programmatically.

What is the best way to run interactive Windows automation commands?

The best way to run interactive Windows automation commands is using REPL mode. This allows you to execute CLI commands for UI capture and system operations interactively within the CLI-Anything architecture.

How do I capture Windows UI elements for automation scripts?

You capture Windows UI elements for automation scripts by using the UI capture commands within the CLI harness. These commands wrap jingmai-agent actions to identify and interact with UI elements.

Why use a CLI harness for Windows automation instead of traditional scripts?

Use a CLI harness for Windows automation instead of traditional scripts to avoid brittleness. It provides a stable, JSON-friendly interface for AI agents, REPL support, and deterministic task execution via the CLI-Anything architecture.