computer-use-agents

Mano-P: Open-source GUI-VLA agent for edge devices. #1 on OSWorld (specialized, 58.2%). Runs locally on Apple M4 Mac mini/MacBook — no data leaves your device.Mano-P 是一个开源 GUI-VLA 项目，支持在 Mac mini/MacBook 上或通过算力棒本地运行推理，实现纯视觉驱动的跨平台 GUI 自动化操作。数据完全本地处理，支持复杂多步骤任务规划与执行。

Compile a demonstrated GUI workflow into a deterministic, locally executable program. Zero model calls on healthy runs; governed repair; halts instead of guessing. Launcher for openadapt-flow: pip install openadapt

[ICLR 2026 Oral] ScaleCUA is the open-sourced computer use agents that can operate on cross-platform environments (Windows, macOS, Ubuntu, Android).

"AnyTool: Universal Tool-Use Layer for AI Agents"

Open-source alternative to Claude Co-Work built by Claude. with BYOK

A Swift library for macOS automation — mouse, keyboard, screenshots, image recognition, and AI-powered agents.

Open-source framework for computer use agents: VeriGen verifiable task synthesis, online RL training (AgentRL), and OSWorld/ScienceBoard evaluation.

let LLMs control your desktop for tasks

AOI: a model-agnostic perception layer giving computer-use agents adaptive video + audio perception with zero retraining, plus DynaCU-Bench (100 dynamic browser tasks). Paper + benchmark + code.

Repository for the implementation of MacArena

A Lightweight Desktop GUI Agent via Dynamic Focus Vision and Hierarchical Memory. Your AI-powered hands and eyes for desktop automation.

Computer Use Examples with API Integration Full Stack

Lightweight recorder that logs everyday computer use as structured traces for agent development

Demo: Herculis-CUA-GUI-Actioner-4B is a Computer Use Agent (CUA) multimodal model designed for GUI understanding, UI localization, and action execution across web, desktop, and mobile environments

AI-powered desktop automation agent. Takes screenshots, sends to any LLM (OpenAI-compatible), receives a ctions, executes mouse/keyboard operations, and loops until task is complete. Supports SOM (element indexing via Windows UIA) and pure vision modes AI 驱动的桌面自动化 Agent。截图 → 发送给 LLM → 执行鼠标键盘操作 → 循环直到任务完成。支持 SOM 模式（Windows UIA 元素索引）和纯视觉模式。

Minicor (operated by Laminar Run, Inc.) is a platform for building and running desktop automations at scale using computer-use AI agents. Legacy desktop software such as EHRs, ERPs, DMS, WMS and PMS systems often has no API, so the only way to read or write data is to drive the desktop like a human.

An enterprise-grade AI agent management platform. Multi-tenant, multi-model, enterprise-ready.

📊 Convert agent framework logs into a standard trajectory format for easy analysis and integration within data pipelines.

Measuring reliability in agentic computer-use RL: the cua_reliability toolkit + reference code for the paper 'Teach it to stop, not just to click' (arXiv:2607.17136). Single-run RL numbers mislead - report K seeds

Context-first benchmark for evaluating suspicious popup adoption in computer-use agents.