Browser Testing

Provides a controllable browser testing environment via an MCP server to test web applications end-to-end.

python

0

GitHub Stars

python

Language

6 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "romangod6-browserbot": {
      "command": "python",
      "args": [
        "/full/path/to/browser_testing_mcp.py"
      ]
    }
  }
}

You can run a Browser Testing MCP Server locally to control a real browser for testing web applications. This server lets you launch, navigate, interact, and verify web pages programmatically, making it easier to reproduce issues, verify flows, and capture visual evidence during development.

How to use

You connect to the MCP server from your client, then use the provided tools to manage a browser session, perform actions on your app, and verify results. Start by launching the server as a local process, then issue commands to control the browser, inspect console output, monitor network activity, and capture screenshots for visual verification.

How to install

{
  "mcpServers": {
    "browser-testing": {
      "command": "python",
      "args": ["/full/path/to/browser_testing_mcp.py"]
    }
  }
}

Prerequisites you need before running the MCP server are:
- Python installed on your system
- pip available to install Python dependencies
- Access to the path where you will run the MCP script
Next, install dependencies and prepare the script as shown above, then start the MCP server using the command in the snippet.

Additional sections

Configuration notes: the MCP server is run as a local process. The entry you will use in your MCP client config points to the Python script that implements the server behavior. Ensure the path to the script is correct and that Python can execute it.

Security and usage tips: keep the script and dependencies up to date, run in a controlled environment, and avoid exposing the local browser controller to untrusted networks. When testing, consider running the browser in visible mode for debugging by setting headless to false in your test flow.

Troubleshooting: if the server fails to start, verify that Python is installed, the path to the script is correct, and that the script has execute permissions. Check for any error output from the Python process and ensure the required dependencies are installed with the correct versions.

Available tools

launch_browser

Start a browser instance, with headless mode configurable to either run in the background or be visible for debugging.

close_browser

Terminate the current browser session and release resources.

navigate_to

Direct the browser to a specified URL, with optional wait conditions.

click_element

Click a page element identified by a CSS selector to interact with the UI.

type_text

Type text into a focused input field or specified selector.

fill_form

Fill multiple form fields in a single operation using field selectors and values.

evaluate_javascript

Execute custom JavaScript in the context of the current page.

get_console_logs

Retrieve JavaScript console output, including errors and warnings.

get_network_requests

Capture and inspect network API calls and responses.

get_page_metrics

Obtain performance metrics for the current page and loading process.

take_screenshot

Capture an image of the current browser viewport or full page for visual verification.

wait_for_selector

Pause actions until a specified element appears or changes state.

check_element_state

Verify element visibility, enablement, and other state checks.

get_local_storage

Read values stored in the browser's localStorage.

get_cookies

Inspect cookies to verify authentication or session state.

get_page_content

Retrieve the HTML content of the current page.