Programming etc

Browser Use UI

user1272047's photo
user1272047
·

3 min read

Table of contents

Here is the extracted and formatted text from the six tabs of the Browser Use UI, with the fields and labels presented clearly and separated by ----------------------------------.

Tab 1: Agent Settings

  • [Agent Type]:
    Label: Select the type of agent to use
    Options:

    • [org]

    • [custom]

  • [Max Run Steps]:
    Label: Maximum number of steps the agent will take
    Min: 1
    Max: 200
    Default: 100

  • [Max Actions per Step]:
    Label: Maximum number of actions the agent will take per step
    Min: 1
    Max: 20
    Default: 10

  • [Use Vision]:
    Label: Enable visual processing capabilities
    Type: Checkbox

  • [Use Tool Calls in Content]:
    Label: Enable Tool Calls in content
    Type: Checkbox

Tab 2: LLM Configuration

  • [LLM Provider]:
    Label: Select your preferred language model provider
    Options:

    • [openai]

  • [Model Name]:
    Label: Select a model from the dropdown or type a custom model name
    Default: gpt-4o

  • [Temperature]:
    Label: Controls randomness in model outputs
    Min: 0
    Max: 2
    Default: 1

  • [Base URL]:
    Label: API endpoint URL (if required)
    Type: Text Field

  • [API Key]:
    Label: Your API key (leave blank to use .env)
    Type: Text Field

Tab 3: Browser Settings

  • [Use Own Browser]:
    Label: Use your existing browser instance
    Type: Checkbox

  • [Keep Browser Open]:
    Label: Keep Browser Open between Tasks
    Type: Checkbox

  • [Headless Mode]:
    Label: Run browser without GUI
    Type: Checkbox

  • [Disable Security]:
    Label: Disable browser security features
    Type: Checkbox

  • [Enable Recording]:
    Label: Enable saving browser recordings
    Type: Checkbox

  • [Window Width]:
    Label: Browser window width
    Default: 1280
    Type: Number Field

  • [Window Height]:
    Label: Browser window height
    Default: 1100
    Type: Number Field

  • [Recording Path]:
    Label: Path to save browser recordings
    Default: ./tmp/record_videos
    Type: Text Field

  • [Trace Path]:
    Label: Path to save Agent traces
    Default: ./tmp/traces
    Type: Text Field

Tab 4: Run Agent

  • [Task Description]:
    Label: Describe what you want the agent to do
    Type: Text Field

  • [Additional Information]:
    Label: Optional hints to help the LLM complete the task
    Type: Text Field

  • [Run Agent Button]:
    Label: Execute the task
    Type: Button

  • [Stop Button]:
    Label: Stop the agent execution
    Type: Button

Tab 5: Results

  • [Latest Recording]:
    Label: Video or log of the latest browser session
    Type: Media Player

  • [Final Result]:
    Label: Output of the agent task
    Type: Text Field

  • [Errors]:
    Label: Errors encountered during task execution
    Type: Text Field

  • [Model Actions]:
    Label: Actions performed by the model
    Type: Text Field

  • [Model Thoughts]:
    Label: Model's internal reasoning or steps
    Type: Text Field

  • [Trace File]:
    Label: Link to the trace file
    Type: Link

Tab 6: Recordings

  • [Recordings Section]:
    Label: List of saved browser recordings
    Type: File Viewer

  • [Refresh Recordings Button]:
    Label: Update the list of recordings
    Type: Button

browser-usebrowser-use-ui