Google adds a way to create automated workflows to Opal

On Tuesday, Google unveiled an exciting update to its vibe-coding application, Opal, introducing a new method for users to create automated workflows. This update features a new agent capable of generating mini applications that allow users to plan and execute tasks simply by using text prompts. The innovative feature leverages the Gemini 3 Flash model, which intelligently selects the appropriate tools needed to carry out various tasks. For example, it can utilize Google Sheets to keep track of information across different sessions, such as maintaining a shopping list for e-commerce applications. Impressively, the new agent is designed to autonomously create and outline the next steps for tasks. Google emphasized that these agents are inherently interactive, meaning they can request additional information from users or present choices to help determine subsequent actions as required. This functionality aims to empower users, even those without technical expertise, to construct intricate workflows within their applications. Initially launched for users in the U.S. in July 2025, Opal allows individuals to create mini web apps or modify existing ones. By October 2025, Google expanded the availability of Opal to 15 additional countries, including Canada, India, Japan, South Korea, Vietnam, Indonesia, Brazil, and Singapore. In December, the tool was further integrated into the Gemini web app, enabling users to develop custom applications through a visual editor, eliminating the need for coding. The trend of utilizing natural language prompts to build applications is gaining traction beyond Google, with several startups, including Lovable and Replit, also offering similar tools. Other emerging players like Wabi, founded by the former creator of Replika, along with Softbank and Lightspeed-backed Emergent, and Accel-backed Rocket.new, are also making significant strides in this space.