Cloud BOT Operator is a feature that enables an AI agent to perform browser operations based on natural language instructions. By leveraging cutting-edge AI models such as those from OpenAI, it can handle tasks that require complex decision-making. This allows for flexible and highly accurate business process automation, going beyond traditional routine task automation.
We are currently offering a preview version free of charge.
The preview version is provided as an experimental feature and is available free of charge. We are continuously improving it based on extensive user feedback.
Please note that it may take some time after registration before the service becomes available.
With Cloud BOT Operator, you can create “Operator Tasks” that allow you to give instructions to the AI in natural language. The AI agent then autonomously operates a virtual browser. It can flexibly adapt to changes in page structure and complex UIs, and once the task is complete, it can seamlessly hand off the process to a traditional RPA task.
How to Use the Operator Feature HereCloud BOT Operator allows you to choose from the following AI models depending on your needs and cost considerations. These models are broadly categorized into two types.
These models analyze the HTML structure and perform browser operations based on user prompts. They enable fast and efficient automation.
Model Name | Features |
---|---|
Structure Recognition - ECO | Low-cost and high-speed model suitable for simple operations (uses gemini-2.0-flash) |
Structure Recognition - Smart | High-performance and fast model with flexibility and stability (uses gpt-4.1) |
These models visually interpret web pages as images and execute operations based on user instructions. They offer higher accuracy than structure recognition models but operate at a slower speed.
Model Name | Features |
---|---|
Visual Recognition | High-performance model capable of operations via visual recognition (uses computer-use-preview) |
It detects and closes pop-ups that appear irregularly, such as news or ads, ensuring smooth operation of RPA.
In scenarios such as clicking on items with changing order every month or selecting a specific image, AI-driven decision-making enables automatic operations.
Easily instruct common search and data extraction operations for different websites using natural language, enabling centralized automation.
By combining with the Cloud BOT Agent, AI-driven automation can be securely and flexibly implemented for external services with IP restrictions or web systems within a company network.
More Information About Cloud BOT Agent Here
Cloud BOT supports secure web systems that require client certificates, and the automation of Operator tasks can also be protected using client certificates.
For Information on Setting Up Client Certificates, Click Here
By using various triggers such as schedule triggers and email receipt triggers available in Cloud BOT, Cloud BOT Operator can be invoked to flexibly handle a wide range of business scenarios.
For Information on Triggers, Click Here
The BOT automatically operates the virtual browser, faithfully following the recorded task steps.
When an Operator task is initiated, the Operator (AI agent) takes over the virtual browser operations in real-time and autonomously performs tasks based on the prompt instructions. Once the operation is complete, control is handed back to the BOT, and the subsequent tasks are automatically executed.
We are currently offering a preview version free of charge.
The preview version is provided as an experimental feature and is available free of charge. We are continuously improving it based on extensive user feedback.
Please note that it may take some time after registration before the service becomes available.