Browser Automation

Full web automation: navigation, interaction, data extraction, screenshots, and performance monitoring.

38 modules

Module	Description
Click Element	Click an element on the page
Close Browser	Close the browser instance and release resources
Capture Console	Capture browser console logs (errors, warnings, info)
Manage Cookies	Get, set, or clear browser cookies
Handle Dialog	Handle alert, confirm, and prompt dialogs
Download File	Download file from browser
Drag and Drop	Drag and drop elements
Emulate Device	Emulate a device or set custom viewport
Ensure Browser	Ensure a browser session exists (reuse or launch)
Execute JavaScript	Execute JavaScript code in page context
Extract Data	Extract structured data from the page
Find Elements	Find elements in page and return element ID list
Fill Form	Smart form filling with automatic field detection
Switch Frame	Switch to iframe or frame context
Mock Geolocation	Mock browser geolocation
Go to URL	Navigate to a specific URL
Hover Element	Hover mouse over an element
Launch Browser	Launch a new browser instance with Playwright
Navigate History	Navigate browser history (back, forward, reload)
Network Monitor	Monitor and intercept network requests
List Pages	List all open browser pages/tabs
Paginate & Extract	Auto-paginate through pages and extract data
Generate PDF	Generate PDF from current page
Performance Metrics	Collect browser performance metrics
Press Key	Press a keyboard key
Record Actions	Record user actions as workflow
Release Browser	Release browser session (close only if owned)
Take Screenshot	Take a screenshot of the current page
Scroll Page	Scroll page to element, position, or direction
Select Option	Select option from dropdown element
DOM Snapshot	Capture DOM snapshot of the current page
Browser Storage	Access localStorage and sessionStorage
Manage Tabs	Create, switch, and close browser tabs
Browser Trace	Start, stop, or save browser performance traces
Type Text	Type text into an input field
Upload File	Upload file to file input element
Set Viewport	Get or set the browser viewport size
Wait	Wait for a duration or until an element appears

Modules

Click Element

browser.click

Click an element on the page

Parameters:

Name	Type	Required	Default	Description
`click_method`	select (`text`, `button`, `id`, `selector`)	No	`text`	Choose the easiest way to identify the element you want to click
`target`	string	No	-	e.g. "Submit", "Next Page", "Login"
`selector`	string	No	-	CSS selector, XPath, or text selector
`button`	select (`left`, `right`, `middle`)	No	`left`	Which mouse button to use for clicking
`click_count`	number	No	`1`	Number of clicks (2 for double-click, 3 for triple-click)
`force`	boolean	No	`False`	Force click even if element is not actionable (covered, invisible)
`modifiers`	array	No	-	Modifier keys to hold during click
`timeout`	number	No	`30000`	Maximum time to wait in milliseconds

Output:

Field	Type	Description
`browser`	object	Browser session (pass-through for chaining)
`status`	string	Operation status (success/error)
`selector`	string	Operation status (success/error)
`method`	string	Click method used

Example: Example

yaml

click_method: text
target: Submit

Example: Example

yaml

click_method: id
target: login-button

Example: Example

yaml

click_method: selector
selector: #submit-button

Close Browser

browser.close

Close the browser instance and release resources

Parameters:

Name	Type	Required	Default	Description
`_no_params`	boolean	No	`True`	This module requires no parameters

Output:

Field	Type	Description
`status`	string	Close the browser instance
`message`	string	Close the browser instance

Example: Example

yaml

Capture Console

browser.console

Capture browser console logs (errors, warnings, info)

Parameters:

Name	Type	Required	Default	Description
`level`	select (`all`, `error`, `warning`, `info`, `log`)	No	`all`	Filter console messages by level
`timeout`	number	No	`5000`	Maximum time to wait in milliseconds
`clear_existing`	boolean	No	`False`	Clear existing messages before capturing

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`messages`	array	Operation status (success/error)
`count`	number	Operation status (success/error)

Example: Example

yaml

timeout: 3000

Example: Example

yaml

level: error
timeout: 5000

Manage Cookies

browser.cookies

Get, set, or clear browser cookies

Parameters:

Name	Type	Required	Default	Description
`action`	select (`get`, `set`, `clear`, `delete`)	Yes	-	Action to perform on the storage
`name`	string	No	-	Name of the cookie
`value`	string	No	-	Value of the cookie
`domain`	string	No	-	Cookie domain
`path`	string	No	`/`	Cookie path
`secure`	boolean	No	`False`	Whether cookie is secure (HTTPS only)
`httpOnly`	boolean	No	`False`	Whether cookie is HTTP only
`expires`	number	No	-	Cookie expiration time (Unix timestamp)

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`cookies`	array	Operation status (success/error)
`count`	number	Operation status (success/error)

Example: Example

yaml

action: get

Example: Example

yaml

action: get
name: session_id

Example: Example

yaml

action: set
name: user_pref
value: dark_mode
domain: example.com

Example: Example

yaml

action: clear

Handle Dialog

browser.dialog

Handle alert, confirm, and prompt dialogs

Parameters:

Name	Type	Required	Default	Description
`action`	select (`accept`, `dismiss`, `listen`)	Yes	-	How to respond to the dialog
`prompt_text`	string	No	-	Text to enter in prompt dialog (for accept action)
`timeout`	number	No	`30000`	Maximum time to wait in milliseconds

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`message`	string	Operation status (success/error)
`type`	string	Operation status (success/error)
`default_value`	string	Result message describing the outcome

Example: Example

yaml

action: accept

Example: Example

yaml

action: dismiss

Example: Example

yaml

action: accept
prompt_text: Hello World

Example: Example

yaml

action: listen
timeout: 5000

Download File

browser.download

Download file from browser

Parameters:

Name	Type	Required	Default	Description
`selector`	string	No	-	CSS selector, XPath, or text selector to find the element
`save_path`	string	Yes	-	Path where to save the downloaded file
`timeout_ms`	number	No	`60000`	Maximum time to wait in milliseconds

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`path`	string	Operation status (success/error)
`filename`	string	Operation status (success/error)
`size`	number	File or resource path

Example: Example

yaml

selector: #download-btn
save_path: /downloads/report.pdf

Example: Example

yaml

selector: a.download
save_path: /downloads/large-file.zip
timeout_ms: 120000

Drag and Drop

browser.drag

Drag and drop elements

Parameters:

Name	Type	Required	Default	Description
`source`	string	Yes	-	CSS selector, XPath, or text selector to find the element
`target`	string	Yes	-	CSS selector, XPath, or text selector to find the element
`source_position`	object	No	-	Position within source element {x, y} as percentages
`target_position`	object	No	-	Position within target element {x, y} as percentages
`timeout`	number	No	`30000`	Maximum time to wait in milliseconds

Output:

Field	Type	Description
`status`	string	Position within target element {x, y} as percentages
`source`	string	Position within target element {x, y} as percentages
`target`	string	Operation status (success/error)

Example: Example

yaml

source: #item1
target: #dropzone

Example: Example

yaml

source: .draggable
target: .container
target_position: {"x": 0.5, "y": 0.5}

Emulate Device

browser.emulate

Emulate a device or set custom viewport

Parameters:

Name	Type	Required	Default	Description
`device`	select (`iphone_12`, `iphone_14`, `iphone_14_pro_max`, `iphone_se`, `pixel_7`, `pixel_5`, `galaxy_s21`, `galaxy_s23`, `ipad_pro`, `ipad_mini`, `galaxy_tab_s8`, `desktop_chrome`, `desktop_firefox`, `desktop_safari`, `desktop_edge`, `laptop`, `macbook_pro`, `custom`)	Yes	-	Device name to emulate (e.g. iPhone 13)
`width`	number	No	-	Viewport width in pixels
`height`	number	No	-	Viewport height in pixels
`user_agent`	string	No	-	Custom user agent string
`is_mobile`	boolean	No	-	Whether to emulate a mobile device
`has_touch`	boolean	No	-	Whether the device has touch support
`device_scale_factor`	number	No	-	Device pixel ratio

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`device`	string	Device name that was emulated
`viewport`	object	Current viewport dimensions
`is_mobile`	boolean	Whether mobile emulation is active

Example: Example

yaml

device: iphone_14

Example: Example

yaml

device: ipad_pro

Example: Example

yaml

device: custom
width: 400
height: 800
is_mobile: true
has_touch: true
device_scale_factor: 2

Example: Example

yaml

device: desktop_chrome
user_agent: CustomBot/1.0

Ensure Browser

browser.ensure

Ensure a browser session exists (reuse or launch)

Parameters:

Name	Type	Required	Default	Description
`headless`	boolean	No	`False`	Run browser without visible window
`width`	number	No	`1280`	Browser viewport width in pixels
`height`	number	No	`720`	Browser viewport height in pixels

Output:

Field	Type	Description
`status`	string	Whether browser was launched or reused
`message`	string	Whether browser was launched or reused
`is_owner`	boolean	Whether browser was launched or reused

Example: Example

yaml

headless: false

Example: Example

yaml

headless: true

Execute JavaScript

browser.evaluate

Execute JavaScript code in page context

Parameters:

Name	Type	Required	Default	Description
`script`	string	Yes	-	JavaScript code to execute (can use return statement)
`args`	array	No	-	Arguments to pass to the script function

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`result`	any	Operation status (success/error)

Example: Example

yaml

script: return document.title

Example: Example

yaml

script: return document.querySelectorAll("a").length

Example: Example

yaml

script: (selector) => document.querySelector(selector)?.textContent
args: ["#header"]

Example: Example

yaml

script: document.body.style.backgroundColor = "red"; return "done"

Extract Data

browser.extract

Extract structured data from the page

Parameters:

Name	Type	Required	Default	Description
`selector`	string	Yes	-	CSS selector, XPath, or text selector to find the element
`limit`	number	No	-	Maximum number of items to extract
`fields`	object	No	-	Define fields to extract from each item

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`data`	array	Operation status (success/error)
`count`	number	Operation status (success/error)

Example: Example

yaml

selector: .g
limit: 10
fields: {"title": {"selector": "h3", "type": "text"}, "url": {"selector": "a", "type": "attribute", "attribute": "href"}}

Find Elements

browser.find

Find elements in page and return element ID list

Parameters:

Name	Type	Required	Default	Description
`selector`	string	Yes	-	CSS selector, XPath, or text selector to find the element
`limit`	number	No	-	Maximum number of items to extract

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`count`	number	Operation status (success/error)
`element_ids`	array	Operation status (success/error)

Example: Find search results

yaml

selector: div.tF2Cxc
limit: 10

Fill Form

browser.form

Smart form filling with automatic field detection

Parameters:

Name	Type	Required	Default	Description
`form_selector`	string	No	-	CSS selector for the form element (optional)
`data`	object	Yes	-	Key-value pairs to fill (key = field name/id, value = content)
`field_mapping`	object	No	-	Custom selector mapping
`clear_before_fill`	boolean	No	`True`	Clear existing field values before filling
`submit`	boolean	No	`False`	Submit form after filling
`submit_selector`	string	No	-	CSS selector for submit button
`delay_between_fields_ms`	number	No	`100`	Delay between filling each field (for more human-like behavior)

Output:

Field	Type	Description
`filled_fields`	array	Delay between filling each field (for more human-like behavior)
`failed_fields`	array	List of fields that were filled
`submitted`	boolean	List of fields that were filled

Example: Example

yaml

data: {"email": "user@example.com", "password": "secret123"}
submit: true

Example: Example

yaml

data: {"username": "john_doe", "bio": "Hello world"}
field_mapping: {"username": "#user-name-input", "bio": "textarea.bio-field"}

Switch Frame

browser.frame

Switch to iframe or frame context

Parameters:

Name	Type	Required	Default	Description
`selector`	string	No	-	CSS selector, XPath, or text selector to find the element
`name`	string	No	-	Name attribute of the frame (alternative to selector)
`url`	string	No	-	URL pattern to match frame (alternative to selector)
`action`	string	No	`enter`	Frame action to perform
`timeout`	number	No	`30000`	Maximum time to wait in milliseconds

Output:

Field	Type	Description
`status`	string	Frame action to perform
`frame_url`	string	Operation status (success/error)
`frame_name`	string	Operation status (success/error)
`frames`	array	Frame URL

Example: Example

yaml

selector: iframe#content-frame

Example: Example

yaml

name: main-content

Example: Example

yaml

action: exit

Example: Example

yaml

action: list

Mock Geolocation

browser.geolocation

Mock browser geolocation

Parameters:

Name	Type	Required	Default	Description
`latitude`	number	Yes	-	Latitude coordinate (-90 to 90)
`longitude`	number	Yes	-	Longitude coordinate (-180 to 180)
`accuracy`	number	No	`100`	Position accuracy in meters

Output:

Field	Type	Description
`status`	string	Position accuracy in meters
`location`	object	Position accuracy in meters

Example: Example

yaml

latitude: 37.7749
longitude: -122.4194

Example: Example

yaml

latitude: 51.5074
longitude: -0.1278
accuracy: 10

Example: Example

yaml

latitude: 35.6762
longitude: 139.6503

Go to URL

browser.goto

Navigate to a specific URL

Parameters:

Name	Type	Required	Default	Description
`url`	string	Yes	-	URL to navigate to
`wait_until`	select (`load`, `domcontentloaded`, `networkidle`)	No	`domcontentloaded`	When to consider navigation complete
`timeout_ms`	number	No	`30000`	Maximum time to wait in milliseconds

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`url`	string	Navigate to a specific URL

Example: Example

yaml

url: https://www.google.com
wait_until: domcontentloaded

Hover Element

browser.hover

Hover mouse over an element

Parameters:

Name	Type	Required	Default	Description
`selector`	string	Yes	-	CSS selector, XPath, or text selector to find the element
`timeout_ms`	number	No	`30000`	Maximum time to wait in milliseconds
`position`	object	No	-	Click position relative to element (0-1 range)

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`selector`	string	Operation status (success/error)

Example: Example

yaml

selector: .menu-item

Example: Example

yaml

selector: #dropdown-trigger
timeout_ms: 5000

Launch Browser

browser.launch

Launch a new browser instance with Playwright

Parameters:

Name	Type	Required	Default	Description
`headless`	boolean	No	`False`	Run browser without visible window
`width`	number	No	`1280`	Browser viewport width in pixels
`height`	number	No	`720`	Browser viewport height in pixels
`browser_type`	select (`chromium`, `firefox`, `webkit`)	No	`chromium`	Browser engine to use (chromium, firefox, webkit)
`proxy`	string	No	-	Proxy server URL
`user_agent`	string	No	-	Custom user agent string
`locale`	string	No	`en-US`	Browser locale (e.g. en-US, zh-TW, ja-JP)
`slow_mo`	number	No	`0`	Slow down operations by specified milliseconds
`record_video_dir`	string	No	-	Directory to save recorded videos (enables Playwright video recording)

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`message`	string	Launch a new browser instance
`browser_type`	string	Type of browser launched
`headless`	boolean	Whether browser is running headless
`viewport`	object	Current viewport dimensions

Example: Example

yaml

headless: true

Example: Example

yaml

headless: false

Navigate History

browser.navigation

Navigate browser history (back, forward, reload)

Parameters:

Name	Type	Required	Default	Description
`action`	select (`back`, `forward`, `reload`)	Yes	`reload`	Which navigation action to perform
`wait_until`	select (`load`, `domcontentloaded`, `networkidle`)	No	`domcontentloaded`	When to consider navigation complete
`timeout_ms`	number	No	`30000`	Maximum time to wait in milliseconds

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`action`	string	Navigation action performed
`url`	string	Current URL after navigation

Example: Example

yaml

action: back

Example: Example

yaml

action: forward

Example: Example

yaml

action: reload
wait_until: networkidle

Network Monitor

browser.network

Monitor and intercept network requests

Parameters:

Name	Type	Required	Default	Description
`action`	select (`monitor`, `block`, `intercept`)	Yes	-	Network action to perform
`url_pattern`	string	No	-	Regex pattern to match request URLs
`resource_type`	string	No	-	Filter by resource type (document, script, image, etc)
`timeout`	number	No	`30000`	Maximum time to wait in milliseconds
`mock_response`	object	No	-	Response to return for intercepted requests

Output:

Field	Type	Description
`status`	string	Response to return for intercepted requests
`requests`	array	Response to return for intercepted requests
`blocked_count`	number	Operation status (success/error)

Example: Example

yaml

action: monitor
url_pattern: .*api.*
timeout: 10000

Example: Example

yaml

action: block
resource_type: image

Example: Example

yaml

action: intercept
url_pattern: .*users.*
mock_response: {"status": 200, "body": "{\"users\": []}"}

List Pages

browser.pages

List all open browser pages/tabs

Parameters:

Name	Type	Required	Default	Description
`include_details`	boolean	No	`True`	Include detailed information for each page
`include_content_info`	boolean	No	`False`	Include content type information for each page

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`pages`	array	List of open pages
`count`	number	Number of open pages
`current_index`	number	Index of the current active page

Example: Example

yaml

include_details: true

Example: Example

yaml

include_details: false

Example: Example

yaml

include_details: true
include_content_info: true

Paginate & Extract

browser.pagination

Auto-paginate through pages and extract data

Parameters:

Name	Type	Required	Default	Description
`mode`	select (`next_button`, `infinite_scroll`, `page_numbers`, `load_more`)	No	`next_button`	How to navigate between pages
`item_selector`	string	Yes	-	CSS selector for items to extract on each page
`fields`	object	No	-	Field definitions {name: {selector, attribute?}}
`next_selector`	string	No	-	CSS selector for next page button
`load_more_selector`	string	No	-	CSS selector for load more button
`max_pages`	number	No	`10`	Maximum number of pages to process (0 = unlimited)
`max_items`	number	No	`0`	Stop after collecting this many items (0 = unlimited)
`wait_between_pages_ms`	number	No	`1000`	Wait time between page navigations
`wait_for_selector`	string	No	-	Wait for this element after page change
`scroll_amount`	number	No	`1000`	Pixels to scroll for infinite scroll mode
`no_more_indicator`	string	No	-	Selector that appears when no more pages (stops pagination)

Output:

Field	Type	Description
`items`	array	Selector that appears when no more pages (stops pagination)
`total_items`	integer	All extracted items from all pages
`pages_processed`	integer	All extracted items from all pages
`stopped_reason`	string	Number of pages processed

Example: Example

yaml

mode: next_button
item_selector: .product-card
fields: {"title": {"selector": ".title"}, "price": {"selector": ".price"}, "link": {"selector": "a", "attribute": "href"}}
next_selector: .pagination .next
max_pages: 5

Example: Example

yaml

mode: infinite_scroll
item_selector: .feed-item
fields: {"content": {"selector": ".content"}, "author": {"selector": ".author"}}
max_items: 100
no_more_indicator: .end-of-feed

Example: Example

yaml

mode: load_more
item_selector: .list-item
load_more_selector: button.load-more
max_pages: 10

Generate PDF

browser.pdf

Generate PDF from current page

Parameters:

Name	Type	Required	Default	Description
`path`	string	No	-	Path where the output file will be saved
`page_size`	select (`A4`, `Letter`, `Legal`, `A3`, `A5`)	No	`A4`	Page size format
`orientation`	select (`portrait`, `landscape`)	No	`portrait`	Page orientation
`print_background`	boolean	No	`True`	Include background graphics
`scale`	number	No	`1`	Scale of the webpage rendering (0.1-2)
`margin`	number	No	`20`	Page margin in millimeters
`header`	string	No	-	Header text for each page
`footer`	string	No	-	Footer text for each page

Output:

Field	Type	Description
`status`	string	Scale of the webpage rendering (0.1-2)
`path`	string	Operation status (success/error)
`size`	number	Operation status (success/error)

Example: Example

yaml

path: /output/page.pdf

Example: Example

yaml

path: /output/landscape.pdf
landscape: true

Example: Example

yaml

path: /output/custom.pdf
margin: {"top": "1cm", "bottom": "1cm", "left": "2cm", "right": "2cm"}

Performance Metrics

browser.performance

Collect browser performance metrics

Parameters:

Name	Type	Required	Default	Description
`metrics`	array	No	`['all']`	Performance metrics to collect
`timeout_ms`	number	No	`3000`	Timeout in milliseconds
`setup_observers`	boolean	No	`True`	Set up performance observers before collecting

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`metrics`	object	Collected performance metrics

Example: Example

yaml

metrics: ["all"]
timeout_ms: 5000

Example: Example

yaml

metrics: ["lcp", "fcp", "cls"]

Example: Example

yaml

metrics: ["ttfb", "domContentLoaded", "load"]
timeout_ms: 0

Press Key

browser.press

Press a keyboard key

Parameters:

Name	Type	Required	Default	Description
`key`	string	Yes	-	The key to press (e.g., Enter, Escape, Tab)

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`key`	string	Press a keyboard key

Example: Example

yaml

key: Enter

Example: Example

yaml

key: Escape

Record Actions

browser.record

Record user actions as workflow

Parameters:

Name	Type	Required	Default	Description
`action`	string	Yes	-	Recording action to perform
`output_format`	string	No	`yaml`	Format for recorded workflow
`output_path`	string	No	-	Path where the output file will be saved

Output:

Field	Type	Description
`status`	string	Format for recorded workflow
`recording`	array	Operation status (success/error)
`workflow`	string	Operation status (success/error)

Example: Example

yaml

action: start

Example: Example

yaml

action: stop
output_format: yaml

Example: Example

yaml

action: get

Release Browser

browser.release

Release browser session (close only if owned)

Parameters:

Name	Type	Required	Default	Description
`force`	boolean	No	`False`	Close browser even if not owned by this template

Output:

Field	Type	Description
`status`	string	Close browser even if not owned by this template
`message`	string	What action was taken
`was_owner`	boolean	What action was taken

Example: Example

yaml

Example: Example

yaml

force: true

Take Screenshot

browser.screenshot

Take a screenshot of the current page

Parameters:

Name	Type	Required	Default	Description
`path`	string	No	`screenshot.png`	Path where the output file will be saved
`full_page`	boolean	No	`False`	Capture the entire scrollable page
`format`	select (`png`, `jpeg`, `webp`)	No	`png`	Image format for the screenshot
`quality`	number	No	-	Quality for JPEG/WebP format (0-100)

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`filepath`	string	Take a screenshot of the current page

Example: Example

yaml

path: output/page.png

Scroll Page

browser.scroll

Scroll page to element, position, or direction

Parameters:

Name	Type	Required	Default	Description
`selector`	string	No	-	CSS selector, XPath, or text selector to find the element
`direction`	select (`down`, `up`, `left`, `right`)	No	`down`	Direction to scroll the page
`amount`	number	No	`500`	Pixels to scroll (ignored if selector is provided)
`behavior`	select (`smooth`, `instant`)	No	`smooth`	How the scroll animation behaves

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`scrolled_to`	object	Operation status (success/error)

Example: Example

yaml

selector: #footer

Example: Example

yaml

direction: down
amount: 500

Example: Example

yaml

direction: up
amount: 10000
behavior: smooth

Select Option

browser.select

Select option from dropdown element

Parameters:

Name	Type	Required	Default	Description
`selector`	string	Yes	-	CSS selector, XPath, or text selector to find the element
`value`	string	No	-	Option value attribute to select
`label`	string	No	-	Option text content to select (alternative to value)
`index`	number	No	-	Option index to select (0-based)
`timeout`	number	No	`30000`	Maximum time to wait in milliseconds

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`selected`	array	Operation status (success/error)
`selector`	string	Operation status (success/error)

Example: Example

yaml

selector: select#country
value: us

Example: Example

yaml

selector: select#country
label: United States

Example: Example

yaml

selector: select#country
index: 2

DOM Snapshot

browser.snapshot

Capture DOM snapshot of the current page

Parameters:

Name	Type	Required	Default	Description
`format`	select (`html`, `mhtml`, `text`)	No	`html`	Output format (html or text)
`selector`	string	No	-	CSS selector to snapshot a specific element
`path`	string	No	-	Path to save the snapshot

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`format`	string	Format of the snapshot
`content`	string	Snapshot content
`path`	string	Path where snapshot was saved
`size_bytes`	number	Size of the snapshot in bytes

Example: Example

yaml

format: html

Example: Example

yaml

format: mhtml
path: /tmp/page.mhtml

Example: Example

yaml

format: text
selector: article.main-content

Example: Example

yaml

format: html
selector: #main
path: /tmp/section.html

Browser Storage

browser.storage

Access localStorage and sessionStorage

Parameters:

Name	Type	Required	Default	Description
`action`	select (`get`, `set`, `remove`, `clear`, `keys`, `length`)	Yes	-	Action to perform on the storage
`type`	select (`local`, `session`)	No	`local`	Browser storage type to access
`key`	string	No	-	Storage key
`value`	string	No	-	Value to store

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`value`	any	Operation status (success/error)
`keys`	array	Operation status (success/error)
`length`	number	The returned value

Example: Example

yaml

action: get
type: local
key: user_token

Example: Example

yaml

action: set
type: session
key: temp_data
value: {"id": 123}

Example: Example

yaml

action: clear
type: local

Example: Example

yaml

action: keys
type: local

Manage Tabs

browser.tab

Create, switch, and close browser tabs

Parameters:

Name	Type	Required	Default	Description
`action`	string	Yes	-	Tab action to perform
`url`	string	No	-	URL to navigate to
`index`	number	No	-	Tab index to switch to or close (0-based)

Output:

Field	Type	Description
`status`	string	Tab index to switch to or close (0-based)
`tab_count`	number	Tab index to switch to or close (0-based)
`current_index`	number	Operation status (success/error)
`tabs`	array	The tab count

Example: Example

yaml

action: new
url: https://example.com

Example: Example

yaml

action: switch
index: 0

Example: Example

yaml

action: close

Example: Example

yaml

action: list

Browser Trace

browser.trace

Start, stop, or save browser performance traces

Parameters:

Name	Type	Required	Default	Description
`action`	string	Yes	-	Trace action (start, stop, save)
`categories`	array	No	`['devtools.timeline']`	Trace categories to capture
`screenshots`	boolean	No	`True`	Include screenshots in the trace
`path`	string	No	-	Path to save the trace file

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`tracing`	boolean	Whether tracing is currently active
`path`	string	Path where trace was saved
`size_bytes`	number	Size of the trace file in bytes

Example: Example

yaml

action: start
screenshots: true

Example: Example

yaml

action: start
categories: ["devtools.timeline", "v8.execute"]
screenshots: false

Example: Example

yaml

action: stop
path: /tmp/performance-trace.json

Type Text

browser.type

Type text into an input field

Parameters:

Name	Type	Required	Default	Description
`type_method`	select (`placeholder`, `label`, `name`, `id`, `selector`)	No	`placeholder`	Choose the easiest way to identify the input field
`target`	string	No	-	e.g. "Enter your email", "Email", "username"
`selector`	string	No	-	CSS selector, XPath, or text selector
`input_type`	select (`text`, `password`, `email`)	No	`text`	Type of input field — use Password to mask the value in the builder
`text`	string	Yes	-
`sensitive_text`	string	Yes	-
`delay`	number	No	`0`	Delay between keystrokes in milliseconds
`clear`	boolean	No	`False`	Clear the input field before typing
`timeout`	number	No	`30000`	Maximum time to wait in milliseconds

Output:

Field	Type	Description
`browser`	object	Browser session (pass-through for chaining)
`status`	string	Operation status (success/error)
`selector`	string	Type text into an input field
`method`	string	Type method used

Example: Example

yaml

type_method: placeholder
target: Enter your email
text: user@example.com

Example: Example

yaml

type_method: label
target: Email
text: user@example.com

Example: Example

yaml

type_method: placeholder
target: Password
input_type: password
sensitive_text: ${env.LOGIN_PASSWORD}

Example: Example

yaml

type_method: selector
selector: #email
text: user@example.com

Upload File

browser.upload

Upload file to file input element

Parameters:

Name	Type	Required	Default	Description
`selector`	string	Yes	-	CSS selector, XPath, or text selector to find the element
`file_path`	string	Yes	-	Local path to the file to upload
`timeout_ms`	number	No	`30000`	Maximum time to wait in milliseconds

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`filename`	string	Operation status (success/error)
`size`	number	Operation status (success/error)
`selector`	string	Name of the file

Example: Example

yaml

selector: input[type="file"]
file_path: /path/to/image.png

Example: Example

yaml

selector: #file-upload
file_path: /path/to/document.pdf

Set Viewport

browser.viewport

Get or set the browser viewport size

Parameters:

Name	Type	Required	Default	Description
`width`	number	Yes	`1280`	Viewport width in pixels
`height`	number	Yes	`720`	Viewport height in pixels

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`viewport`	object	Current viewport dimensions
`previous_viewport`	object	Previous viewport dimensions

Example: Example

yaml

width: 375
height: 667

Example: Example

yaml

width: 768
height: 1024

Example: Example

yaml

width: 1920
height: 1080

Example: Example

yaml

width: 1366
height: 768

Wait

browser.wait

Wait for a duration or until an element appears

Parameters:

Name	Type	Required	Default	Description
`duration_ms`	number	No	`1000`	Duration of the operation in milliseconds
`selector`	string	No	-	CSS selector, XPath, or text selector to find the element
`state`	select (`visible`, `hidden`, `attached`, `detached`)	No	`visible`	State to wait for (visible, hidden, attached, detached)
`timeout_ms`	number	No	`30000`	Maximum time to wait in milliseconds

Output:

Field	Type	Description
`status`	string	Operation status (success/error)
`selector`	string	Operation status (success/error)
`duration_ms`	number	Wait for a duration or element to appear

Example: Example

yaml

duration_ms: 2000

Example: Example

yaml

selector: #loading-complete
timeout_ms: 5000

Browser Automation ​

Modules ​

Click Element ​

Close Browser ​

Capture Console ​

Manage Cookies ​

Handle Dialog ​

Download File ​

Drag and Drop ​

Emulate Device ​

Ensure Browser ​

Execute JavaScript ​

Extract Data ​

Find Elements ​

Fill Form ​

Switch Frame ​

Mock Geolocation ​

Go to URL ​

Hover Element ​

Launch Browser ​

Navigate History ​

Network Monitor ​

List Pages ​

Paginate & Extract ​

Generate PDF ​

Performance Metrics ​

Press Key ​

Record Actions ​

Release Browser ​

Take Screenshot ​

Scroll Page ​

Select Option ​

DOM Snapshot ​

Browser Storage ​

Manage Tabs ​

Browser Trace ​

Type Text ​

Upload File ​

Set Viewport ​

Wait ​

Browser Automation

Modules

Click Element

Close Browser

Capture Console

Manage Cookies

Handle Dialog

Download File

Drag and Drop

Emulate Device

Ensure Browser

Execute JavaScript

Extract Data

Find Elements

Fill Form

Switch Frame

Mock Geolocation

Go to URL

Hover Element

Launch Browser

Navigate History

Network Monitor

List Pages

Paginate & Extract

Generate PDF

Performance Metrics

Press Key

Record Actions

Release Browser

Take Screenshot

Scroll Page

Select Option

DOM Snapshot

Browser Storage

Manage Tabs

Browser Trace

Type Text

Upload File

Set Viewport

Wait