Google is rolling out a new feature called 'Skills' to its Chrome desktop browser, allowing users to save and execute repeatable Gemini AI prompts across multiple web pages.
The update, which began rolling out today for users with their language set to US English, enables any AI prompt to be saved as a reusable command.
“Until now, repeating an or AI task — like asking for ingredient substitutions to make a recipe vegan — meant re-entering the same prompt as you visited different pages,” Chrome product manager Hafsah Ismail said in an announcement.
“To make this easier, we’re launching Skills in Chrome, which lets you save and reuse your most helpful AI prompts and run them with a single click,” Ismail added.
Users can access their saved library by typing a forward slash ( / ) in Gemini or clicking the plus ( + ) button. The feature also includes a compass icon for managing these commands, according to the Google AI Blog.
Once a Skill is selected, it runs on the currently viewed web page. Users can also select additional tabs to run the same prompt across multiple sources simultaneously, according to reports from TechCrunch and Ars Technica.
Custom workflows and preset libraries
Google is providing a library of preset Skills to help users get started with common tasks in areas such as productivity, shopping, budgeting, and recipes. These presets can be customized by editing the underlying prompt to better suit individual needs.
Early testers have already utilized the feature for specific workflows. According to the Google AI Blog, these include calculating protein macros in recipes for health and wellness, generating side-by-side specification comparisons while shopping, and scanning lengthy documents for important information.
Users can save a prompt as a Skill directly from their Gemini chat history. Because the feature is tied to the Google account, saved Skills will be available on any Chrome desktop device where the user is signed in.
Security remains a primary focus for the integration. The Google AI Blog stated that Skills utilize the same safeguards as manual prompts, including automated red-teaming and auto-update capabilities.
“Skills are built on Chrome’s foundation of security and privacy, and they utilize the same safeguards we apply to prompts in Gemini in Chrome,” the company stated.
Specifically, the system will ask for explicit user confirmation before performing certain actions, such as sending an email or adding an event to a user's calendar.
This move positions Chrome more aggressively within a crowded AI-integrated browser market. TechCrunch noted that the feature arrives as Google faces competition from other AI-centric browser tools, including OpenAI's Atlas, Perplexity's Comet, and The Browser Company's Dia.
