arXiv:2610.04116 · Agent 智能体
CUAWright: A Minimal Unified Interface for Digital Agents
CUAWright:面向数字 Agent 的极简统一接口
CUAWright: A Minimal Unified Interface for Digital Agents
- 类型:arxiv
- 标识:2610.04116
- 链接:https://arxiv.org/abs/2610.04116
- 主分类:agent
- 形态:method
- TLDR:The prevailing approach to computer-use agents couples a model with a domain-specific harness: a browser or desktop environment equipped with human engineered tools that are fixed before task execution. As models' coding capabilities improve, the GUI native and static harness prevents them from direct programmatic operation on system state, as well as flexible construction of tools. To this end, we introduce CUAWright, a minimal terminal harness of roughly 3K lines of code that uses bash commands as its sole action interface, and a file system as its evolvable space for dynamically creating to
- 副分类:evaluation
- 待LLM分类:否
- 标题中文:CUAWright:面向数字 Agent 的极简统一接口
- TLDR中文:主流的计算机使用 Agent 方法将模型与特定领域 harness 耦合:浏览器或桌面环境配备任务执行前固定的、由人工设计的工具。随着模型编码能力提升,原生 GUI 与静态 harness 阻碍了它们对系统状态进行直接编程化操作以及灵活构建工具。为此,我们提出 CUAWright,一个约 3K 行代码的极简终端 harness,以 bash 命令作为唯一动作接口,并以文件系统作为可动态创建工具的可演化空间
- 来源文件:
- /inbox/tom/_candidates/2026-10-06-rag-retrieval-reranking-candidates.json