codebolt.browser

Browser automation and web interaction tools for controlling web browsers programmatically.

Available Tools

• browser_navigate - Navigate to a specific URL
• browser_click - Click on an element using a selector
• browser_type - Type text into an input element
• browser_screenshot - Take screenshot of current page
• browser_get_content - Get content of current page
• browser_scroll - Scroll the page in a specific direction
• browser_get_url - Get the current page URL
• browser_get_html - Get the HTML content of the current page
• browser_get_markdown - Get the page content in markdown format
• browser_search - Search for text in an input element
• browser_enter - Simulate pressing the Enter key
• browser_close - Close the current browser page

Tool Parameters

browser_navigate

Navigates the browser to a specified URL.

Parameter	Type	Required	Description
url	string	Yes	The URL to navigate to.
instance_id	string	No	Optional browser instance ID for multi-instance support.

browser_click

Clicks on an element in the browser page.

Parameter	Type	Required	Description
element_id	string	Yes	The ID of the element to click.
instance_id	string	No	Optional browser instance ID for multi-instance support.

browser_type

Types text into an element in the browser page.

Parameter	Type	Required	Description
element_id	string	Yes	The ID of the element to type into.
text	string	Yes	The text to type.
instance_id	string	No	Optional browser instance ID for multi-instance support.

browser_screenshot

Takes a screenshot of the current browser page.

Parameter	Type	Required	Description
instance_id	string	No	Optional browser instance ID for multi-instance support.
full_page	boolean	No	Whether to take a full page screenshot (default: false).
quality	number	No	Image quality from 0-100 (for JPEG format).
format	string	No	Image format: 'png' or 'jpeg'.

browser_get_content

Retrieves the text content of the current browser page.

Parameter	Type	Required	Description
instance_id	string	No	Optional browser instance ID for multi-instance support.

browser_scroll

Scrolls the browser page in a specified direction.

Parameter	Type	Required	Description
direction	string	Yes	Scroll direction: 'up', 'down', 'left', 'right'.
pixels	string	Yes	Number of pixels to scroll.
instance_id	string	No	Optional browser instance ID for multi-instance support.

browser_get_url

Retrieves the current URL of the browser page.

Parameter	Type	Required	Description
instance_id	string	No	Optional browser instance ID for multi-instance support.

browser_get_html

Retrieves the HTML content of the current browser page.

Parameter	Type	Required	Description
instance_id	string	No	Optional browser instance ID for multi-instance support.

browser_get_markdown

Retrieves the current browser page content as Markdown.

Parameter	Type	Required	Description
instance_id	string	No	Optional browser instance ID for multi-instance support.

browser_search

Performs a search within an element on the browser page.

Parameter	Type	Required	Description
element_id	string	Yes	The ID of the element to perform the search in.
query	string	Yes	The search query.
instance_id	string	No	Optional browser instance ID for multi-instance support.

browser_enter

Simulates pressing the Enter key in the browser.

Parameter	Type	Required	Description
instance_id	string	No	Optional browser instance ID for multi-instance support.

browser_close

Closes the browser page.

Parameter	Type	Required	Description
instance_id	string	No	Optional browser instance ID for multi-instance support.

Sample Usage

// Navigate to a specific URL
const navigateResult = await codebolt.tools.executeTool(
  "codebolt.browser",
  "browser_navigate",
  { url: "https://example.com" }
);

// Click on an element using element ID
const clickResult = await codebolt.tools.executeTool(
  "codebolt.browser",
  "browser_click",
  { element_id: "submit-button" }
);

// Type text into an input element
const typeResult = await codebolt.tools.executeTool(
  "codebolt.browser",
  "browser_type",
  { element_id: "search-input", text: "test" }
);

// Take a screenshot
const screenshotResult = await codebolt.tools.executeTool(
  "codebolt.browser",
  "browser_screenshot",
  { full_page: true, format: "png" }
);

// Get page content
const contentResult = await codebolt.tools.executeTool(
  "codebolt.browser",
  "browser_get_content",
  {}
);

// Scroll the page
const scrollResult = await codebolt.tools.executeTool(
  "codebolt.browser",
  "browser_scroll",
  { direction: "down", pixels: "100" }
);

// Get current URL
const urlResult = await codebolt.tools.executeTool(
  "codebolt.browser",
  "browser_get_url",
  {}
);

// Get HTML content
const htmlResult = await codebolt.tools.executeTool(
  "codebolt.browser",
  "browser_get_html",
  {}
);

// Get markdown content
const markdownResult = await codebolt.tools.executeTool(
  "codebolt.browser",
  "browser_get_markdown",
  {}
);

// Search in an input element
const searchResult = await codebolt.tools.executeTool(
  "codebolt.browser",
  "browser_search",
  { element_id: "search-input", query: "test query" }
);

// Simulate Enter key
const enterResult = await codebolt.tools.executeTool(
  "codebolt.browser",
  "browser_enter",
  {}
);

// Close the browser
const closeResult = await codebolt.tools.executeTool(
  "codebolt.browser",
  "browser_close",
  {}
);

info

This functionality provides browser automation capabilities through the MCP interface.