Computer use tool

The computer use tool is a beta API capability that lets Claude autonomously interact with desktop computer environments. Claude uses its vision capabilities to capture screenshots of a display, then issues actions — mouse clicks, mouse movement, keyboard input, and scrolling — to control the computer as a human would. This allows it to operate standard desktop applications, web browsers, and other GUI-based software without requiring custom API integrations for each tool. Instead of building task-specific integrations, developers teach Claude general computer skills, enabling it to complete multi-step workflows across arbitrary applications. For example, Claude can open a browser, navigate to a website, fill out a form, switch to a spreadsheet application, and save results — all autonomously within a single session. The feature is implemented through an 'agent loop': the client application captures a screenshot and sends it to Claude via the API; Claude analyzes it and returns one or more tool-use actions (e.g., click at coordinate X,Y or type text); the client executes those actions on the actual OS; a new screenshot is captured; and the loop repeats until the task is complete or a maximum iteration limit is reached.