Google has upgraded Opal, adding a new agent feature that enables users to create automated workflows with ease, even without coding knowledge.
Google has announced significant improvements to Opal, its no-code AI workflow tool that facilitates automated task management.
Opal, our no-code visual builder for AI workflows, just got a major upgrade. 🧠💎
We’ve added a new agent step that analyzes your goal, determines the best approach, and automatically calls the right tools — such as Veo for video or web search for research — to complete the task.
— Google Labs (@GoogleLabs) February 24, 2026
The new agent feature allows users to develop mini-apps or enhance existing applications by simply using text prompts. It utilizes the Gemini 3 Flash model to identify and activate the appropriate tools automatically.
Some key uses of this feature include:
- Linking with Google Sheets to retain data across sessions, such as tracking shopping lists;
- Leveraging web searches for information gathering or utilizing Veo to create video content.
This method empowers the assistant to create a detailed action plan and decide on subsequent steps independently.
Moreover, Google highlighted that the agents can ask users for further information as needed or offer suggestions on next actions, making the tool accessible to those without technical expertise.
First launched in July 2025, Opal aims to help users craft mini-apps and recently became accessible via the web version of Gemini, allowing for visual assembly of products without any coding required.
The rise of vibe coding has sparked competition among numerous companies and startups, with offerings from platforms like Lovable and Replit gaining traction. Meanwhile, major AI players, including OpenAI and Anthropic, are also making strides in this space, with notable developments like the introduction of a "team of agents" in Anthropic's Claude Opus 4.6 model.
