Users can soon save AI requests in Chrome for quick reuse

Contents

Google is introducing a new browser feature for AI users in Chrome called “Skills.” Users can save frequently used AI requests directly in the browser to quickly and easily retrieve them again, instead of having to type them in again. This turns useful AI commands in Chrome into 1-click tools, as Google calls it. Initially, this feature will be introduced in English-language Chrome, with other languages likely to follow.

Last autumn, Google integrated its AI model Gemini into Chrome, making it an AI browser. This was also initially limited to the USA and English-language requests. Chrome now has a sidebar where Gemini can be instructed, for example, to summarize texts or compare data. This is even possible across multiple tabs, as Gemini knows which tabs are open and has access to their content.

Extension of Gemini in Chrome

The new AI function also utilizes this capability, as one of the examples of Skills mentioned by Google is comparing product data when shopping in multiple tabs. Additionally, testers of the feature have saved frequently requested AI prompts for calculating protein macronutrients in recipes and searching for important information in longer documents as Skills because they use these requests frequently.

“Until now, repeating an AI task — like asking for ingredient substitutions to make a recipe vegan — meant re-entering the same prompt as you visited different pages,” writes Chrome Product Manager Hafsah Ismail in the Google blog post. “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.”

Skill examples directly from Google

AI prompts are saved directly in the chat window, so past AI requests from the history can also be saved for quick reuse. To retrieve saved AI commands again, a slash (/) or a click on the plus symbol is sufficient, according to Google. Saved Skills can be changed at any time. If users use their Google account in Chrome, the saved AI prompts will also be accessible on other systems.

Google already provides a range of Skills that users can try out and adapt for their purposes. These include AI commands for modifying recipes by suggesting protein-rich alternatives, for help with gift selection, for adapting texts to the vocabulary and language patterns of a character, or for dramatizing a story, for example in the style of a movie preview.

Agentic commands only with confirmation

Google promises to use the same security and privacy safeguards for Skills as are used for Gemini in Chrome. Therefore, the browser requires explicit confirmation from the user for AI requests that lead to actions such as a new calendar entry or sending an email.

Skills for Gemini in Chrome is currently rolling out in the USA for users of Windows, Mac, and ChromeOS who have set their browser language to US English. Google has not yet provided any information on other language versions and countries.

(fds)