Google Rolls Out New Gemini-Powered 'Skills' AI Feature for Chrome

Google announced this Tuesday that it is expanding the artificial intelligence capabilities built into its Chrome web browser with a brand-new customizable tool called Skills. The feature lets users save and reuse their most-used custom AI prompts across any web page, eliminating the repetitive hassle of retyping the same prompts over and over again.

Skills builds on Google’s existing integration of its Gemini large language model into Chrome, a development that arrives as the global browser ecosystem sees a wave of new AI-native competitors from companies including OpenAI (with its upcoming Atlas browser), Perplexity (Comet), and The Browser Company (Dia), among many others.

Gemini already enables Chrome users to ask questions about active web page content, summarize on-page information, and complete a wide range of small tasks directly in the browser. Skills takes this functionality a step further by letting users store custom AI prompts that they can pull up and reuse any time with just a single click.

As a sample use case, Google highlights that if a user regularly asks Gemini to suggest vegan ingredient substitutions when browsing cooking recipe sites, they can now save that pre-written prompt and deploy it instantly across any new recipe page they open.

To create a custom Skill, users simply save their existing AI prompt as a Skill directly from their Gemini chat history. Once saved, the tool can be accessed in Chrome’s Gemini interface either by typing a forward slash (/) or clicking the plus (+) button. The saved prompt will then automatically run against the currently active web page, as well as any other selected open tabs. Google also confirms that saved Skills can be edited or adjusted at any point to fit changing user needs.

During internal testing, the company found that early adopters used Skills for a diverse range of common workflows, spanning health and wellness (such as automatically calculating protein macros for recipe ingredients), side-by-side shopping comparisons, and scanning and summarizing lengthy documents.

To help new users get started with the feature immediately, Google is also launching a public Skills library that hosts pre-built prompts for popular tasks across categories including productivity, shopping, cooking, personal budgeting, and more. To use any of these pre-programmed prompts, users just add it to their personal collection of saved Skills in Chrome, and can edit the prompt text at any time to customize it for their specific needs.

Consistent with other Gemini-powered actions in Chrome, Skills will always request explicit user confirmation before completing any sensitive actions, such as sending an email or adding a new event to a user’s Google Calendar.

The Skills feature begins rolling out globally starting today, and is available initially to all desktop Chrome users who are signed into a Google account. At launch, the feature will only work for users who have set their Chrome browser’s default language to English (United States).