cua-driver

Name: cua-driver
Availability: InStock
Author: ruifengfu

Automates macOS app interactions via the Accessibility tree using cua-driver.

Updated May 27, 2026

Search Tags:#accessibility #macos #ui interaction #app automation #cua-driver

Ready to run

npx skills add https://github.com/RuifengFu/agent-skills --skill cua-driver-ruifengfu

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: cua-driver
Source: https://github.com/RuifengFu/agent-skills/tree/main/hermes/openclaw-imports/cua-driver
Command: npx skills add https://github.com/RuifengFu/agent-skills --skill cua-driver-ruifengfu

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires cua-driver, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates interactions with native macOS apps by driving them via the cua-driver CLI or MCP server, enabling precise control of applications for automation tasks without foregrounding or activating the app.

Core Features & Use Cases

App Automation: Automate operations such as clicking, typing, scrolling, and more in any macOS application.
AX Tree Interaction: Interact with the Accessibility tree of an app to perform actions on specific UI elements.
Background Operation: Perform actions in the background, ensuring the user's frontmost app remains active.
Use Case: Use this Skill to automate tasks like opening a specific app, clicking on a button within an app, typing text, and more without user intervention.

Quick Start

Use the cua-driver skill to open the Calculator app in the background and calculate '17 + 23'.

Frequently Asked Questions about cua-driver

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

FAQPage Schema

How does macOS app automation work using the Accessibility tree?▼

macOS app automation via the Accessibility tree manipulates specific UI elements directly, enabling background actions like clicking, typing, and scrolling without foregrounding or activating the target application.

How do I automate clicking and typing in a macOS background app?▼

To automate clicking and typing in a background macOS app, you use cua-driver CLI or MCP server to drive the app's UI elements via the Accessibility tree, ensuring your frontmost app remains active throughout the process.

Do I need special macOS permissions for app automation with cua-driver?▼

Yes, macOS app automation with cua-driver requires system permissions for Accessibility and Screen Recording to read the Accessibility tree and interact with UI elements across native applications.

What is the best way to automate native macOS applications without user intervention?▼

The best way to automate native macOS applications without user intervention is using cua-driver CLI or MCP server, which performs background operations by manipulating the Accessibility tree of target apps.

Can I use cua-driver MCP server for macOS UI interaction tasks?▼

Yes, you can use the cua-driver MCP server for macOS UI interaction tasks, as it supports automating operations like opening apps, clicking buttons, and typing text by driving the Accessibility tree.

Why does my macOS UI automation task require Screen Recording permissions?▼

macOS UI automation tasks require Screen Recording permissions alongside Accessibility access to fully read and interpret the visual layout and Accessibility tree of the target applications for precise manipulation.