cua-patterns

Community

AI agents that control computers.

AuthorLuisSambrano
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill enables AI agents to interact with computer interfaces like humans, automating tasks that require screen observation, mouse movements, and keyboard input.

Core Features & Use Cases

  • Perception-Reasoning-Action Loop: The fundamental pattern for vision-based agents.
  • Sandboxed Environments: Ensures secure execution of automation tasks.
  • Anthropic Computer Use Implementation: Leverages advanced LLM capabilities for sophisticated control.
  • Use Case: Automate the process of filling out a web form by having an AI agent "see" the form on the screen, "decide" what to type where, and then "execute" the typing and clicking actions.

Quick Start

Use the cua-patterns skill to build an AI agent that can take screenshots and click on specific coordinates.

Dependency Matrix

Required Modules

None required

Components

scriptsreferences

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: cua-patterns
Download link: https://github.com/LuisSambrano/antigravity-config/archive/main.zip#cua-patterns

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 223,000+ vetted skills library on demand.