ADevillers/Pupil

Open-source Computer Use Agent that works for Any App (GUI or CLI) by Any Model (Vision or Text).

Local MCP server that gives any AI agent safe desktop control on Windows, macOS and Linux. Reads the screen as text and structure (accessibility + OCR) and acts by element name, with screenshots as the last resort. Multi-monitor aware, verifies every action, one safety gate.

MCP Server for Computer Use in Windows

Open-source desktop AI agent that sees what you see — one hotkey over any window, it answers, acts, or guides you step-by-step.

A lightweight, Windows-native AI desktop assistant. Focus any app, press Ctrl+Win, and let it click, type, and automate with human-in-the-loop safety.

Advancing Foundation GUI Agents with In-Context Demonstrations

An open-sourced end-to-end VLM-based GUI Agent

Windows desktop app automation and reproducible evidence Agent Skill

Agent Desktop gives any agent reliable computer use on the desktop. Built with Rust, it sees any app's real UI structure through OS accessibility trees and operates it — refs stay stable and actions stay safe to retry, instead of guessing from pixels.