Google announced Tuesday that it is adding Skills to Chrome, a feature that allows users to save and reuse AI prompts across different websites. The capability extends the company's Gemini AI integration in the browser, which launched as Google moved to compete with other browser-based AI assistants from OpenAI, Perplexity, and The Browser Company.
Skills enables users to create custom AI prompts that can be accessed repeatedly with minimal effort. Rather than typing the same instruction each time, a user can save a prompt once and trigger it across multiple web pages. Google provided the example of a user who frequently asks Gemini to suggest vegan recipe substitutions. By saving that prompt as a Skill, the user can apply it whenever browsing recipe websites without re-entering the instruction.
Accessing Skills is straightforward. Users save an AI prompt as a Skill directly from their chat history within Gemini in Chrome. Once saved, Skills can be reused by typing a forward slash (/) or clicking a plus sign (+) button. When invoked, the Skill runs on the currently viewed web page and can also be applied to additional selected tabs. Google notes that Skills can be edited at any time to refine or adjust their behavior.
During internal testing, Google observed that early adopters applied Skills across several categories. Health and wellness emerged as a primary use case, with users employing Skills to calculate protein macros in recipes. Shopping comparisons and document summarization were also common applications, indicating that users found value in automating routine research and analysis tasks.
To help new users get started, Google is launching a Skills library featuring pre-built workflows organized by category. The library includes common tasks in productivity, shopping, recipes, budgeting, and additional domains. Users can add any library Skill to their saved Skills with a single action. The pre-built Skills remain customizable—users can edit the underlying prompts to align with their specific needs.
Safety and control remain central to the implementation. As with other Gemini actions in Chrome, Skills will request user confirmation before executing certain sensitive actions, such as sending emails or creating calendar events. This design ensures users maintain oversight over automated tasks.
The rollout begins today for Chrome desktop users who are signed into their Google accounts. The initial availability is limited to users whose Chrome browser language is set to English (US), suggesting a phased expansion may follow as the feature stabilizes and receives localization work.