Google adds AI features to Chrome for saving workflows

Google has announced new AI capabilities for its Chrome web browser. The new feature called Skills will allow users to save and reuse their favorite AI prompts that can run across different web pages without having to type them in again.

This feature ties into the integration of Gemini AI into Chrome, which has emerged alongside new competitors in the browser ecosystem, such as OpenAI (Atlas), Perplexity (Comet), and The Browser Company (Dia). Gemini already allows users to ask questions about a web page, summarize its information, or perform various tasks.

The Skills feature will enable users to create AI prompts that can be accessed time and again with just a click. For instance, if a user often asks Gemini to suggest vegan substitutions when looking at recipe websites, they can now save that prompt and use it across different web pages.

To access the feature, users need to save the AI prompt as a Skill directly from chat history. The Skill can then be reused in Gemini in Chrome by typing a forward slash (/) or clicking the plus sign (+) button. The Skill will run on the web page being viewed, along with any additional tabs that have been selected. These Skills can also be edited at any time.

In tests, the company found that early adopters used Skills in areas like health and wellness, for example, to calculate protein macros in recipes, or for shopping comparisons, or scanning and summarizing lengthy documents. To help users get started with Skills, the company is also launching a Skills library that will offer common tasks and workflows in areas like productivity, shopping, recipes, budgeting, and more.

To use one of the pre-programmed Skills, users just add it to their saved Skills in Chrome. The Skill can also be customized to fit a user’s needs by editing the prompt. Like other Gemini actions in Chrome, Skills will ask the user for confirmation before taking certain actions, such as sending an email or adding an event to your calendar.

Skills will begin rolling out today to Chrome desktop users who are signed into their Google account. The feature will initially work only if your Chrome browser’s language is set to English (US).