Playwright

Provides browser automation and data extraction via Playwright for AI-driven workflows.

python

1

GitHub Stars

python

Language

6 months ago

First Indexed

2 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": {
    "nolecram-build_mcp_server": {
      "command": "playwright-mcp-server",
      "args": []
    }
  }
}

Playwright MCP Server gives you a powerful way to control a browser through the Model Context Protocol. It enables AI assistants to automate web interactions, perform tests, scrape data, and manage browser workflows with Playwright.

How to use

You run the Playwright MCP Server as a local process and connect your MCP clients to it. Once started, you can direct the server to navigate pages, click elements, type into forms, take screenshots, wait for page states, execute custom JavaScript, and extract content. Use your MCP client to issue high‑level browser actions and orchestrate multi-step workflows across pages and tabs.

How to install

Prerequisites include Python 3.8 or higher and Node.js for Playwright browser installations.

pip install playwright-mcp-server
playwright install chromium

Configuration and usage tips

Configure your MCP client to connect to the Playwright MCP Server by pointing to the stdio server command. You can also configure the VSCode MCP extension to launch the server directly from your editor.

{
  "mcpServers": {
    "playwright": {
      "command": "playwright-mcp-server",
      "args": []
    }
  }
}

Security and runtime notes

The server runs in headless mode by default and uses isolated browser contexts for each session. Resources are cleaned up automatically, and you can configure timeouts to suit your workload. Avoid logging sensitive data and ensure your client connections are secured.

Examples and practical patterns

Typical usage patterns include web scraping, form automation, and end‑to‑end testing workflows. You can navigate to pages, interact with elements, fill forms, wait for selectors or network events, take screenshots, and extract data.

Troubleshooting

If the server fails to start, verify Python and Playwright are installed, ensure chromium is installed via the Playwright installer, and confirm that your client is configured to connect to the server command. Check for common errors like missing dependencies or permission issues and re-run the install steps.

Available tools

browser_navigate

Navigate to a URL using the browser context.

browser_click

Click on an element specified by a selector, with optional timeout.

browser_type

Type text into an element specified by a selector, with optional timeout.

browser_screenshot

Capture a screenshot of the page or its full page.

browser_get_text

Extract visible text from an element selected by a selector.

browser_wait_for_selector

Wait for a selector to appear, with optional timeout and state.

browser_evaluate

Execute JavaScript in the browser context.

browser_new_tab

Open a new browser tab with an optional URL.

browser_close_tab

Close the current browser tab.

browser_get_title

Get the current page title.

browser_get_url

Get the current URL of the active page.