Computer-use agent
An AI agent that operates software through its graphical interface, viewing the screen and controlling the mouse and keyboard as a person would.
In more detail
A computer-use agent interacts with software the way a person does: it perceives the screen, moves a cursor, clicks, and types. This lets it operate applications that expose no API or integration, which is a large share of the systems real businesses depend on. It is the fallback for reaching software that cannot be reached any other way.
The trade-off is fragility relative to an API. Interfaces change, screens load slowly, and visual interpretation can misfire, so screen-driven automation is generally less robust than a direct integration. The practical rule is to use an API when one exists and reserve computer use for the systems that leave no better option.
Where this shows up at Ceven
Ceven prefers direct integrations, connecting across 1,000+ tools through their APIs, which is the more reliable path. Screen-driven computer use is the approach for the long tail of systems that offer no integration at all, and even there the consequential actions can sit behind an approval gate and are recorded in the audit trail.