AutoHub scene demonstration: Unveiling the black technology of browser intelligent automation
Explore the future of browser automation and experience revolutionary tools that improve efficiency.
Core content:
1. AutoHub - a browser automation tool based on a large language model
2. Intelligent dialogue interaction for web page operation automation and information extraction
3. Practical application of cross-site task one-click execution and general question-and-answer function
With the integration of browser plug-in technology and large model capabilities, traditional web page operation modes are gradually being replaced by intelligent automation. AutoHub , a browser automation tool based on cutting-edge large language models (LLM) launched by OpenCSG , aims to help users browse the web and complete tasks more efficiently through intelligent dialogue interaction and automation technology. It can not only automate tedious web page operations, but also provide users with accurate information retrieval and page content summaries. With AutoHub , users can save a lot of time and energy, simplify complex web page operations, and thus improve work efficiency.
Task requirements : Please help me extract the text in the column corresponding to Architecture in the Text Generation table and output it in a new line.
Feature highlights : AutoHub achieves contextual linkage understanding with the current web page content through the "page dialogue mode". Without the need for users to search or copy and paste, the plug-in can extract the specified content through natural language instructions, demonstrating LLM's powerful understanding and information integration capabilities.
Scenario steps :
Switch mode : the dialog box switches to [Page Dialog] mode;
Input requirement : Input "Please help me extract the text in the column corresponding to Architecture in the Text Generation table and output it in a new line.";
Start answering : AutoHub understands the task context and automatically extracts the corresponding column content based on web page DOM analysis and LLM reasoning;
Get results : Users can view the output structured information directly in the sidebar, saving time and effort.
Task requirements : Help me find an article related to Jack Ma on Zhihu, and then open it for me to read.
Feature highlights : In "automatic workflow mode", AutoHub can understand natural language input, automatically generate task execution plans, and automatically complete complex operation processes including searching, clicking, and jumping in the browser.
Scenario steps :
Switch mode : The dialog box switches to [Automatic Workflow] mode;
Input requirement : Input "Help me find an article related to Jack Ma on Zhihu, and then open it for me to read.";
Generate plan : The plug-in analyzes keywords and operation intentions to generate a structured web page operation sequence;
Automatic execution : The plug-in automatically opens Zhihu, searches for "Jack Ma" and clicks to enter the relevant page;
Complete the action : Users do not need to intervene and eventually view the desired article in a new tab.
Task requirements : How does Sglang support new models?
Feature highlights : The AutoHub sidebar supports a general Q&A function, allowing users to ask questions about technology, models, programming, etc. to the plug-in at any time. The plug-in will combine the context and LLM capabilities to give accurate answers instantly, truly achieving "asking questions while browsing."
Usage scenario : The user wants to record a common web page operation process (such as logging into a website, opening links in batches, etc.) so that it can be repeated with one click.
Scenario steps :
Click to record : The user clicks the [Start Recording] button in the plug-in;
Name the workflow : Enter a name for the current process (e.g. "Automatic submission of daily work reports");
Operation record : When users perform normal operations in the browser, such as clicking, filling in, jumping, etc., the plug-in background automatically records the operation sequence;
Stop recording : Click [Stop Recording] and the plug-in will generate an executable workflow;
Execution playback : Click [Execute Workflow] and the plug-in will simulate the execution operation and quickly reuse the original process.
Usage scenarios : For technical users or advanced users, AutoHub provides a "drag-and-drop" visual configuration workflow function and supports customized web task processes.
Scenario steps :
Enter configuration : Click [Configure Workflow] to enter the workflow editing interface in the plug-in;
Drag and drop : Users can drag operation components such as "open webpage", "enter keywords", "click buttons" from the operation library to customize the arrangement process;
Execute workflow : After completing the configuration, click [Start Execution] and the plug-in will execute each step in sequence to complete the entire process task.
AutoHub is redefining the way we interact with the web. From extracting web information through natural language , to intelligent automatic browsing operations , to visual recording and reuse of workflows , AutoHub is no longer just a plug-in, but also an intelligent operation assistant for users in the digital world.
Whether you are a daily office user, a technical developer, or a social content creator, AutoHub can free your hands and improve your efficiency in your usage scenarios.
