forepaw

Automate macOS GUI apps via accessibility tree and input simulation.

1|Updated Mar 31, 2026
One-click install
npx skills add https://github.com/aviraccoon/forepaw --skill forepaw
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: forepaw
Source: https://github.com/aviraccoon/forepaw/tree/main/.agents/skills/forepaw
Command: npx skills add https://github.com/aviraccoon/forepaw --skill forepaw

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Forepaw automates macOS GUI interactions by combining UI inspection, input simulation, and multi-step workflows, eliminating repetitive manual tasks.

Core Features & Use Cases

  • Snapshot and OCR for locating UI elements, then click, type, ocr-click, drag, scroll, or hover to automate tasks across Finder, Spotify, Discord, and other apps.
  • Batch sequences to perform multi-step interactions while preserving app focus and state, reducing context switches and errors.
  • Window targeting with --app and --window, region-based actions, and cross-app automation, including permissions prompts and accessibility constraints.

Quick Start

Install forepaw, grant Accessibility and Screen Recording permissions, then run a sample command to observe an app and begin automating tasks.

Frequently Asked Questions about forepaw

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

FAQPage Schema
How do I perform multi-step batch interactions without losing application focus on macOS?

You can perform multi-step batch interactions by chaining actions like click, type, and scroll in a sequence, which preserves application focus and state, reducing context switches and errors during complex cross-app automation workflows.

How do I automate macOS GUI tasks like clicking and typing across multiple applications?

To automate macOS GUI tasks, you can use accessibility tree inspection combined with input simulation to interact with windows, buttons, and text fields across apps like Finder, Spotify, and Discord, executing multi-step workflows without manual intervention.

Does macOS GUI automation require special permissions to control windows and take screenshots?

Yes, macOS GUI automation requires granting Accessibility and Screen Recording permissions to allow the tool to read the accessibility tree, capture screenshots, and simulate keyboard and mouse inputs across different applications.

What is the best way to locate and interact with UI elements that are not exposed in the accessibility tree?

For UI elements not in the accessibility tree, you can use snapshot and OCR workflows to visually locate elements on screen, then perform click, type, or drag actions based on the recognized text and screen coordinates.

Can I target a specific application window when running cross-app automation workflows?

Yes, you can target specific application windows during cross-app automation by using --app and --window flags, ensuring actions are directed to the correct UI region while preserving app focus and state throughout batch sequences.

How do I perform multi-step batch interactions without losing application focus on macOS?

You can perform multi-step batch interactions by chaining actions like click, type, and scroll in a sequence, which preserves application focus and state, reducing context switches and errors during complex cross-app automation workflows.