PrompteryxBeta
    ⌘K
    PlatformSupport
    Menu
    IntroductionInstall Desktop & extensionQuick StartPlatform Overview
    OverviewWhat you can makeAI Browser Agent in the chatModels and AutoConnectorsSchedulesProjectsExplore and publishingShare linksMemory and SkillsPersonal agent (beta)Welcome, settings and skinsPlans, credits and top-upsPrompteryx DesktopGroups (pilot)Phone app (early access)
    OverviewCreating WorkflowsTesting & DebuggingScheduling
    OverviewGetting StartedCommandsRecordingSettingsStructured OutputParallel AgentsTips & Best Practices
    OverviewInstallationSidebar AgentUsing Attached FilesExecute on Current TabExporting to Visual Studio
    Overview
    OverviewOpen PageClickClick CoordinatesFillExtract TextScreenshotWaitPress KeyFocus ElementBlur ElementSelect TextHighlight ElementSet ViewportGet Bounding BoxScrollHoverSelect OptionAssertionsCookiesEvaluate JavaScriptExample Workflows
    OverviewPrompteryx Browser AgentOpenAI
    OverviewSchedule TriggerWebhook Trigger
    OverviewGoogle SheetsGoogle CalendarConnect Hub (3,000+ apps)Example Workflows
    AI Vision SettingsChrome Automation ProfilesCloud BrowserBrowser Profiles & Persistent ContextVideo RecordingExecution LogsSharing & CollaborationEdit History
    Selectors GuideVariables
    Start hereGet an API keyRun the AI browser agentDrive a cloud browserRun workflows from anywhereUse connected apps (Connect Hub)SDK (TypeScript)CLIMCP for AI assistants
    REST APIWebhooksMCP toolsCloud Browser APICloud Browser quickstartMigrate from Browserbasen8nZapierMake.com

    Need help? Contact Support

    1. Documentation
    2. Getting Started
    3. Quick Start

    Quick Start Guide

    Your first task in the app in one minute, then your first workflow in five

    Part 1: your first task in the app

    Everything starts in the app. When you sign in you land on its home, with one message box.

    The Prompteryx app home: the sidebar, the message box and quick-start buttons such as Do a task, Build workflow and AI Browser Agent
    The app home. Type a request, or start from one of the buttons.
    1. Type Go to example.com and tell me the main heading and press Enter.
    2. The app sees that this needs a website and starts the AI Browser Agent. A browser opens beside the chat, and you can watch it work. Nothing needs installing: it uses a cloud browser unless you set up your own Chrome.
    3. The run ends with an outcome card: the answer and the steps it took, or why it didn't finish, with Try again and Try with … another model. Ask a follow-up in the same chat if you want more. A browser run can also be kept as a workflow you can edit and run again (see Recording), and Schedule in the message box repeats a task.

    The same box builds websites and games, makes images, video and music, writes documents, researches with citations and uses your connected apps. See What you can make and The Prompteryx app for the full tour.

    Part 2: build a workflow in Visual Studio

    This part walks you through building a simple browser automation by hand: it opens a website, extracts some text, and saves it. By the end, you'll understand the basics of building workflows in Visual Studio.

    Where it runs: pick Cloud Browser in the Visual Studio profile dropdown to run with nothing installed, or keep a local Chrome profile if you have Prompteryx Desktop and Prompteryx in Chrome set up.Installation guide →

    Building your first automation

    1
    Open Visual Studio

    In the app's sidebar, under Create, click Visual Studio, or go directly to:

    /visual-studio

    You'll see an empty canvas, with the node library on the left containing all available nodes.

    2
    Add a Manual Trigger

    Every workflow needs a trigger to start execution. For this example, we'll use a Manual Trigger:

    1. Open the "Triggers" tab in the node library on the left
    2. Drag the Manual trigger node onto the canvas
    3. This node starts the workflow when you click the "Run Workflow" button

    💡 Tip: Manual Trigger is perfect for testing. Later, you can use Schedule Trigger to run automations automatically.

    3
    Add an Open Page Node

    Now add a browser node to navigate to a website:

    1. Open the "Browser" tab in the node library
    2. Drag the Open Page node onto the canvas
    3. Connect the Manual Trigger to the Open Page node (drag from the trigger's output handle to the Open Page input handle)
    4. Click on the Open Page node to open its configuration panel
    5. Enter a URL, for example: https://example.com
    4
    Add an Extract Text Node

    Let's extract the page title:

    1. Drag the Extract Text node onto the canvas
    2. Connect the Open Page node to the Extract Text node
    3. Click on Extract Text to configure it
    4. In the Element Selector field, enter: h1
    5. In the Variable Name field, enter: pageTitle

    This will extract the text from the first <h1> element on the page and store it in a variable called "pageTitle".

    5
    Save Your Workflow

    Before running, save your workflow:

    1. Type a name, like "My First Automation", in the workflow name field at the top left of the toolbar
    2. Open the "Save" menu next to it
    3. Choose "Save" (or "Save As..." to save a copy under a new name)
    6
    Execute Your Automation

    Time to run your automation!

    1. Click the Run Workflow button in the centre of the top toolbar
    2. Watch as each node executes in sequence
    3. The browser (your local Chrome or the Cloud Browser session) navigates to the URL and extracts the text
    4. Green checkmarks will appear on successfully completed nodes

    Congratulations! You've created and run your first automation!

    Understanding What Happened

    Workflow Execution Flow
    1

    Manual Trigger started the workflow execution

    2

    Open Page opened a browser page and navigated to your URL

    3

    Extract Text found the h1 element and saved its text to the "pageTitle" variable

    Where Did the Data Go?

    The extracted text was stored in a variable called "pageTitle". You can use this variable in subsequent nodes by referencing it as {{pageTitle}}.

    To see past runs and their extracted data, switch the Editor / History toggle at the top right of Visual Studio to "History".

    Next Steps: Expand Your Automation

    Now that you've created a basic automation, try adding more functionality:

    Add a Click Node

    Click on buttons or links on the page

    Learn about Click nodes →
    Add a Fill Node

    Type text into forms and input fields

    Learn about Fill nodes →
    Add API Nodes

    Send data to external services like Gmail, Slack, etc.

    Explore integration nodes →
    Add Conditional Logic

    Use IF nodes to make decisions based on data

    Learn about conditional logic →

    Learn More

    Creating Workflows →

    Deep dive into workflow creation

    Browser Nodes Reference →

    Explore all available browser nodes

    Try the AI Browser →

    Record automations using natural language

    Testing & Debugging →

    Learn how to test your workflows