Google has launched Chrome Skills, a feature enabling its Gemini AI assistant to automate multi-step workflows directly within the Chrome browser. The capability, announced this week, allows users to create persistent automation tasks that execute across multiple web pages without requiring repeated manual intervention.
The feature represents Google’s most significant integration of AI capabilities into Chrome since the browser’s launch in 2008. Unlike traditional browser extensions or one-off AI queries, Chrome Skills enables users to define repeatable tasks—such as monitoring price changes, aggregating information from multiple sites, or executing regular web-based workflows—that Gemini can perform autonomously over time.
According to reports from The Verge, the feature operates through natural language commands, with users describing desired workflows in conversational terms. Gemini then translates these instructions into automated browser actions, navigating pages, extracting data, and executing tasks without further user input. The AI maintains context across sessions, enabling workflows that span days or weeks.
The technical implementation relies on Chrome’s existing extension architecture combined with Gemini’s large language model capabilities. Google has built safeguards requiring user permission for actions involving financial transactions or personal data, though the company has not disclosed specific details about security protocols or data handling practices.
Chrome currently commands approximately 65% of the global browser market share, providing Google with a substantial distribution advantage for AI features. The Skills launch positions Chrome not merely as a web access tool but as a platform for AI agents—software that acts semi-autonomously on behalf of users.
The business implications extend across several sectors. E-commerce platforms may face increased price monitoring and comparison shopping automation, potentially intensifying price competition. Web publishers could see changes in traffic patterns as AI agents aggregate content on users’ behalf. Enterprise software vendors offering workflow automation tools face a new competitor with direct access to billions of users.
For Google, Chrome Skills strengthens the strategic rationale for maintaining Chrome’s market dominance whilst providing a consumer-facing showcase for Gemini capabilities. The feature creates potential lock-in effects, as users who invest time creating custom workflows become less likely to switch browsers. It also generates additional usage data that can improve Gemini’s training.
Competitors are pursuing similar capabilities. Microsoft has integrated AI features into Edge through Copilot, whilst OpenAI has discussed browser-based agents. Apple’s Safari remains notably absent from this space, though the company’s focus on privacy may complicate AI agent implementation.
Privacy advocates have raised concerns about the data access required for effective workflow automation. Chrome Skills necessarily observes user behaviour across websites to execute tasks, creating a detailed activity log. Google states that users control which sites Skills can access, but the feature’s utility depends on broad permissions.
The initial release supports a limited set of workflow types, focusing on information gathering and monitoring tasks. Google has indicated plans to expand capabilities based on user feedback and safety considerations. The company has not disclosed whether Chrome Skills will remain free or eventually become part of a premium Gemini subscription tier.
Industry observers will be monitoring several metrics: adoption rates among Chrome’s user base, the complexity of workflows users actually create, and whether automated agent behaviour creates problems for websites dealing with non-human traffic at scale. The feature’s success may determine whether browsers evolve into AI agent platforms or whether such capabilities remain niche tools for power users.
Chrome Skills launches as technology companies compete to demonstrate practical AI applications beyond chatbots. Google’s integration into the browser—software people use daily for hours—provides a testing ground for whether consumers genuinely want persistent AI agents or prefer more bounded AI interactions.
