This project provides a web interface for visually following browser automation tasks and building a "receipt" of actions to be executed by a browser extension.
- Task Management: Add and remove tasks or questions.
- Visualizer: A log panel shows a history of actions taken within the interface. This is a placeholder for future visual feedback from the browser.
- Receipt Generation: Compile the task list into a structured JSON format.
- Download Receipt: Download the generated JSON receipt file. This file is intended to be used by a companion browser extension.
- Open
index.htmlin a web browser. - Type a task (e.g., "Go to wikipedia.org") into the input field and click "Add Task" or press Enter.
- Continue adding tasks to build your automation sequence.
- You can remove tasks by clicking the trash can icon next to them.
- Once your task list is complete, click "Download Receipt".
- A JSON file named
receipt-....jsonwill be downloaded. This file contains your structured list of tasks.
The downloaded receipt is a JSON file with the following structure:
{
"version": "1.0",
"createdAt": "2023-10-27T10:00:00.000Z",
"tasks": [
{
"id": 1672531200000,
"type": "user-question",
"params": {
"query": "Go to wikipedia.org"
}
},
{
"id": 1672531205000,
"type": "user-question",
"params": {
"query": "Search for 'Artificial Intelligence'"
}
}
]
}This receipt can then be loaded into a browser extension designed to interpret and execute these tasks.
- Connect to a live browser extension via WebSockets for real-time execution and visual feedback (screenshots, DOM highlights).
- Implement task reordering (e.g., drag-and-drop).
- Add more sophisticated task types beyond simple "user-question".
- Allow editing of existing tasks.