Anthropic Gives Claude Hands and Eyes on Your Desktop
Claude 3.5 Sonnet now accepts screenshots and outputs mouse clicks plus keystrokes through its computer use API. The model can open apps, fill forms, and navigate websites without custom scripts. Anthropic reports the feature works on standard macOS and Ubuntu desktops.
Teams stop treating AI as a text box and start treating it as an extra employee who already knows every interface. The workflow shift moves from prompt engineering to task delegation across existing software.
The Anthropic research team released the computer use preview to select API customers. Early internal tests showed the model completing multi step web tasks at roughly the speed of a careful human operator.
Step 1: Sign up for the Anthropic API at https://console.anthropic.com and request computer use access. Step 2: Send a screenshot plus a natural language instruction such as 'click the login button and enter [email protected]'. Step 3: Receive coordinate outputs that your code translates into actual mouse events on the target machine.