Figure 2: The CUAWright architecture. Terminal serves as the sole interface between the agent
and the task environment. At each step of task execution process, the agent model receives com-
mand line output as observation or call a CLI tool to load image into context on demand, and output
commands that gets executed. The underlying execution environment contains a runtime, an ex-
pandable set of installed CLI tools and applications, and a file system as the agent’s workspace to
programmatically manage system state and craft new tools.