Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/vercel-labs/agent-browser/llms.txt

Use this file to discover all available pages before exploring further.

Overview

Interaction commands allow you to simulate user interactions with page elements.

Mouse Interactions

click

Click an element.
ParameterDescription
selectorCSS selector, ref (@e1), or text locator
Options:
OptionDescription
—new-tabOpen link in new tab
Examples:
# Click by ref
agent-browser click @e2

# Click by CSS selector
agent-browser click "#submit-button"

# Click by text
agent-browser click "text=Sign In"

# Click and open in new tab
agent-browser click @e5 --new-tab

dblclick

Double-click an element.
ParameterDescription
selectorElement selector
Examples:
agent-browser dblclick @e3
agent-browser dblclick ".file-name"

hover

Hover over an element.
ParameterDescription
selectorElement selector
Examples:
agent-browser hover @e4
agent-browser hover ".dropdown-trigger"

focus

Focus an element.
ParameterDescription
selectorElement selector
Examples:
agent-browser focus @e1
agent-browser focus "#email-input"

Text Input

fill

Clear and fill an input field.
ParameterDescription
selectorInput element selector
textText to fill (multiple words joined with spaces)
Examples:
# Fill by ref
agent-browser fill @e3 "test@example.com"

# Fill by selector
agent-browser fill "#email" "user@domain.com"

# Fill with multiple words
agent-browser fill @e2 "John Doe"

type

Type text into an element (without clearing).
ParameterDescription
selectorInput element selector
textText to type
Examples:
agent-browser type @e1 "additional text"
agent-browser type "#search" "query"

press

Press a keyboard key.
ParameterDescription
keyKey name (Enter, Tab, Escape) or key combination (Control+a)
Alias: key Examples:
# Press Enter
agent-browser press Enter

# Press Tab
agent-browser press Tab

# Keyboard shortcuts
agent-browser press Control+a
agent-browser press Meta+v

keyboard type

Type text with real keystrokes (current focus).
ParameterDescription
textText to type
Examples:
agent-browser keyboard type "Hello World"

keyboard inserttext

Insert text without key events (current focus).
ParameterDescription
textText to insert
Examples:
agent-browser keyboard inserttext "pasted content"

Form Controls

check

Check a checkbox.
ParameterDescription
selectorCheckbox selector
Examples:
agent-browser check @e5
agent-browser check "#agree-terms"

uncheck

Uncheck a checkbox.
ParameterDescription
selectorCheckbox selector
Examples:
agent-browser uncheck @e6
agent-browser uncheck "#newsletter"

select

Select dropdown option(s).
ParameterDescription
selectorSelect element selector
valueOption value(s) to select
Examples:
# Select single option
agent-browser select @e4 "option1"

# Select multiple options
agent-browser select "#country" "US" "CA"

Scrolling

scroll

Scroll the page or an element.
ParameterDescription
directionDirection: up, down, left, right (default: down)
amountPixels to scroll (default: 300)
Options:
OptionDescription
-s, —selectorScroll specific element instead of page
Examples:
# Scroll down 300px (default)
agent-browser scroll

# Scroll down 500px
agent-browser scroll down 500

# Scroll up
agent-browser scroll up

# Scroll specific element
agent-browser scroll down 200 --selector "#content"

scrollintoview

Scroll element into view.
ParameterDescription
selectorElement selector
Alias: scrollinto Examples:
agent-browser scrollintoview @e10
agent-browser scrollinto "#footer"

Drag and Drop

drag

Drag an element to another element.
ParameterDescription
sourceSource element selector
targetTarget element selector
Examples:
agent-browser drag @e1 @e2
agent-browser drag ".draggable" ".drop-zone"

File Upload

upload

Upload file(s) to an input element.
ParameterDescription
selectorFile input selector
filesFile path(s) to upload
Examples:
# Upload single file
agent-browser upload @e7 "/path/to/file.pdf"

# Upload multiple files
agent-browser upload "#file-input" "/path/to/file1.jpg" "/path/to/file2.jpg"

Workflow Examples

Login Form

agent-browser open example.com/login
agent-browser fill @e1 "user@example.com"
agent-browser fill @e2 "password123"
agent-browser check @e3  # Remember me
agent-browser click @e4  # Submit button

Search and Navigate

agent-browser open example.com
agent-browser fill @e1 "search query"
agent-browser press Enter
agent-browser wait --load networkidle
agent-browser click @e5  # First result

Form with Dropdowns

agent-browser fill @e1 "John Doe"
agent-browser select @e2 "United States"
agent-browser check @e3
agent-browser upload @e4 "/path/to/resume.pdf"
agent-browser click @e5

Build docs developers (and LLMs) love