acu

New

This repository is a curated, living index of resources about AI agents for computer use. It aggregates articles, survey and research papers, frameworks and models, UI grounding methods, datasets, benchmarks, safety analyses, open source projects, commercial offerings, environment and sandbox tooling, and automation libraries. The README organizes content into topical sections such as Articles, Papers (Surveys, Frameworks & Models, UI Grounding, Dataset, Benchmark, Safety), Projects (Open Source, Environment & Sandbox, Automation) and Commercial offerings. Entries often include links to code, papers, datasets and websites where available. The collection is intended to help researchers, developers, and practitioners discover recent work and tooling in areas like vision-language grounding, web and mobile agent design, trajectory synthesis datasets, and agent evaluation benchmarks. The repository also documents contribution instructions and highlights representative projects and benchmarks.

Stars

1370

App URL

https://github.com/trycua/acu

Github Repository

https://github.com/trycua/acu/blob/main/README.md

Features

Comprehensive topical organization with a detailed table of contents that groups resources into Articles, Surveys, Frameworks & Models, UI Grounding, Datasets, Benchmarks, Safety and Projects. Curated lists of open source projects include frameworks and implementations such as AutoGen, Auto-GPT, Browser Use, Notte and OpenInterpreter and other tools for computer control. Dataset and benchmark entries point to resources like OS-Genesis, AgentTrek, AssistantBench, AppWorld and AndroidAgentArena and link to code where available. UI grounding and vision-language entries highlight projects such as OmniParser and AskUI/PTA-1. Environment and sandbox sections catalogue testing platforms and containerized systems for macOS/Linux and Windows. Automation libraries such as nut.js and PyAutoGUI are listed. The README includes contribution guidance and frequent updates to recent papers and code examples.

Use Cases

This curated collection helps users quickly find and compare literature, datasets, frameworks, benchmarks and implementations relevant to building and evaluating agents for computer and mobile device control. Researchers can use it to locate recent surveys and papers across UI grounding, reinforcement learning, model architectures and safety analyses. Developers gain pointers to open source frameworks, sandboxing tools and automation libraries to prototype or test agents. Practitioners can discover benchmarks and datasets to evaluate agent capabilities on web, mobile and desktop tasks. The README also links to commercial products and demos to understand industry offerings and tradeoffs. Finally, the repository provides contribution instructions so the community can keep the index current and add new papers, code and tools.

acu

Basic Information

Links

Categorization

App Details

Categories

Similar Listings

awesome rag

ai agent papers

Deeper Seeker

LLM Tool Survey

ii researcher

Featured Listings

Terry Bison Ranch

The Singapore Flyer

Tags

More Filters

acu