Gemini in Chrome adds ‘Select from screen’ tool

Key Points:

• Gemini in Chrome introduces a new “Select from screen” tool that allows users to quickly add images to prompts by selecting any text or image from the current tab, available in Chrome 149 after a browser restart.
• The “Select from screen” feature is accessible at the bottom of the ‘plus’ menu and enhances prompt creation by integrating visual content directly from the browsing experience.
• Google has upgraded Gemini to version 3.5 Flash, which now includes a built-in computer use tool, replacing the previous standalone Gemini 2.5 Computer Use model and enabling more advanced automation across browser, mobile, and desktop environments.
• Developers can leverage Gemini 3.5 Flash to build custom agents capable of seeing, reasoning, and taking actions, improving long-term task performance such as continuous software testing and enterprise knowledge work.
• Enhanced safety features for enterprise users include mandatory user confirmation for sensitive actions and automatic task termination if indirect prompt injection is detected; Gemini 3.5 Flash is currently available via the Gemini API.