Playwright MCP server

Tired of LLMs struggling with web interactions through unreliable screenshots? Give your AI agents precise browser automation with Playwright MCP server—using structured accessibility data instead of visual parsing for faster, accurate web navigation and extraction.

Try Playwright MCP server

POWERING THE WORLD'S TOP DATA-DRIVEN TEAMS

What is Playwright MCP server?

Playwright MCP server connects AI agents with Playwright scraper via the Model Context Protocol. It fetches browser automation capabilities, accessibility snapshots, structured data extraction, web navigation tools, and more.

Accessibility-first browser automation

Leverages Playwright's accessibility tree for faster and lighter web interactions without requiring vision models or screenshots.

Structured data extraction

Enables deterministic web navigation and form filling using structured data instead of pixel-based approaches.

LLM-powered web automation

Provides AI agents with comprehensive browser automation capabilities for testing, data extraction, and intelligent web interactions.

One token to rule them all

Single Apify API key unlocks 6,000+ Actors and MCP servers. One account, one invoice and transparent usage tracking.

Dynamic tool discovery

Is Playwright MCP server not enough? Agents can search, evaluate, and add other Apify tools based on workflow needs.

Smart prompts and resources

Pre-built prompts and resources help agents understand tool capabilities and use cases.

Prerequisites

Before you start, make sure you have the following:

An Apify account

Apify API Token

Get your personal API token from the Integrations section in Apify Console. This token will be used to authorize the MCP server to run Actors on your behalf.

MCP client

An AI agent or client that supports MCP. This could be Anthropic Claude for Desktop, a VS Code extension with MCP support, Apify’s web-based Tester MCP Client, or any custom client implementation. You can also check supported MCP clients in official documentation.

How to set up Playwright MCP server

If you have an MCP client that supports streamable HTTP, connect using OAuth authentication by providing Playwright MCP server URL you can find below. You'll be redirected to your browser to sign in to your Apify account and approve the connection.

1{
2 "mcpServers": {
3   "apify": {
4     "url": "https://mcp.apify.com/?actors=jiri.spilka/playwright-mcp-server"
5   }
6 }
7}

Check documentation

If you don't have an MCP client that supports streamable HTTP, you can still set up Apify's Playwright MCP server. For Claude for Desktop, enable MCP servers via the Developer settings. Then, you need to add an entry for the Apify MCP server. For instance, in Claude’s config file, under mcpServers, add an entry like:

{
    "mcpServers": {
      "actors-mcp-server": {
        "command": "npx",
        "args": [
          "-y", "@apify/actors-mcp-server",
          "--actors", "jiri.spilka/playwright-mcp-server"
        ],
        "env": {
           "APIFY_TOKEN": "YOUR_APIFY_TOKEN"
        }
      }
    }
   }

Check documentation

MCP server configuration

Configure your own Apify MCP server exactly as you need with Actors and tools for easy integration with MCP clients.

Task examples for Playwright MCP server

"Can you take a screenshot of the GitHub homepage?"

"Fill out the login form on this website with my test credentials"

"Click the "Add to Cart" button on this product page"

"Check if the search results loaded properly after I clicked submit"

Sample dataset

Here's a sample of the data you can extract with our Playwright MCP server including structured accessibility data, browser automation controls, web navigation elements, extraction capabilities, and much more without the limitation of visual parsing's official API.

{
  "title": "Web Scraping, Data Extraction and Automation - Apify",
  "#error": false,
  "#debug": {
    "requestId": "fvwscO2UJLdr10B",
    "url": "https://apify.com",
    "loadedUrl": "https://apify.com/",
    "method": "GET",
    "retryCount": 0,
    "errorMessages": null,
    "statusCode": 200
  }
}

Try Playwright MCP Server online

Use the Tester MCP Client, our chat user interface, to interact with Playwright MCP server online. And for free.

Tester MCP Client

jiri.spilka/tester-mcp-client

A model context protocol (MCP) client that connects to any MCP server using Streamable HTTP and displays the conversation in a chat-like UI. It is a standalone Actor server designed for testing MCP servers over Stremable HTTP.

Jiří Spilka

847

5.0

Frequently asked questions

The Playwright MCP server is a tool that enables AI agents and applications to extract text content, form data, interactive elements, and accessibility information from web pages using structured browser automation. It provides programmatic access to comprehensive data beyond the limitations of screenshot-based visual parsing's official API.

To use the Playwright MCP server, create an Apify account and get your API token. Then configure the server in your MCP-compatible client using either OAuth authentication or by adding the configuration with your token.

No, you don't need Playwright API key to use our Playwright MCP server. The server operates through Apify's infrastructure and only requires an Apify account with API token for authentication and access to the service.

You can access structured web data through the Playwright MCP server using accessibility-based extraction rather than visual parsing. Key data types include form elements, interactive components, text content, navigation structures, and page metadata—all retrieved through reliable DOM and accessibility tree analysis instead of screenshot interpretation.

The Playwright MCP server has no specific rate limits imposed by the service itself, but usage is subject to your Apify account's resource limits and pricing plan.