Google is transforming Gemini into a more active tool within the Chrome browser. Initially designed as a sidebar assistant for tasks like summarizing text and comparing prices, Gemini is evolving into an AI agent capable of executing complex tasks autonomously. This shift is rooted in a new feature called “Skills,” found in testing on an internal page (chrome://skills). Users will soon be able to customize Gemini’s abilities by naming skills and providing specific instructions, allowing the AI to manage repetitive tasks tailored to individual workflows. The move from a passive helper to an active agent marks a significant change in how users will interact with Gemini. Until now, tasks required users to provide context while Gemini generated outputs. With Skills, Gemini will engage more directly with the browser's functionalities, handling tasks without users needing to manage each step. This transition aims to bridge the gap between asking Gemini for information and having it perform actions on behalf of the user. Integration with the broader Google ecosystem is also a key part of this upgrade. Gemini is set to work seamlessly with apps like Google Calendar, YouTube, and Maps directly from the browser. For instance, it could automate scheduling by identifying details from a confirmation email and adding events without manual input. While these features are still in internal testing, their development indicates a shift toward a more capable AI presence in the browser.