Playwright MCP

What is Playwright MCP?

Playwright MCP is a Model Context Protocol server published by Microsoft as the @playwright/mcp npm package under the Apache-2.0 license. It lets an MCP client such as VS Code, Cursor, Windsurf, Claude Desktop, Claude Code, Codex, Cline, Goose or Gemini CLI browse and operate web pages through Playwright. Instead of screenshots, it hands the model the page's accessibility tree, so no vision model is needed and tool calls act on specific elements deterministically. That makes it suited to exploratory automation, self-healing tests and long-running autonomous workflows where persistent browser state matters. For coding agents that care about context size, the README points to the Playwright CLI with skills as a more token-efficient alternative. Setup is a short JSON config that runs npx @playwright/mcp@latest, and it requires Node.js 18 or newer. It is free to install, and no paid plans are listed.

Use cases

• Let an LLM browse and click through websites
• Automate web tasks from an AI agent
• Explore a site to draft tests
• Build self-healing test workflows
• Scrape structured page content with agents
• Run long autonomous browser sessions

Who it's for

• AI agent developers
• QA engineers
• Developers using coding agents
• Automation engineers

Why use Playwright MCP

• Page structure is passed as text, so models can act on pages without needing image understanding.
• Element references make actions less ambiguous than pixel-based clicking.
• One short config works across many MCP clients, so you can switch assistants without redoing setup.
• Persistent browser state suits multi-step exploratory or long-running agent loops.
• It is released by Microsoft under Apache-2.0 and installs from npm without a subscription.

Playwright MCP features

• Browser automation exposed as MCP tools
• Structured accessibility snapshots instead of screenshots
• No vision model required
• Deterministic element-based actions
• Works with any MCP client
• Configured with a standard npx command
• Allowed and blocked origin options
• Option to block service workers
• Runs on top of Playwright's browser engines

Playwright MCP integrations

• VS Code
• Cursor
• Windsurf
• Claude Desktop
• Claude Code
• Codex
• Cline
• Goose
• Gemini CLI
• GitHub Copilot
• Amp
• Antigravity

How to get started with Playwright MCP

1. Make sure Node.js 18 or newer is installed.
2. Add the standard MCP config that runs npx @playwright/mcp@latest, or use your client's one-line add command such as claude mcp add playwright.
3. Start your MCP client, which launches the server and exposes the browser tools.
4. Ask your assistant to open a page, click, type or read content; it receives accessibility snapshots with element references.
5. Adjust options such as allowed origins or browser choice in the server arguments when you need tighter control.

Playwright MCP limitations

• Requires Node.js 18 or newer
• Accessibility trees and tool schemas add to the model's context, so the README suggests the Playwright CLI for token-sensitive coding agents
• Version 0.0.83 is still a 0.0.x release
• Community forks and the Cloudflare package are separate projects with their own features

Playwright MCP pricing

Playwright MCP is open source, so you can use and self-host it for free.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

Playwright MCP alternatives

The main Playwright MCP alternatives listed on Debuly are Browserbase, Stagehand, Browser Use, Firecrawl, AutoGPT and Agent Zero. The table compares pricing and platforms; order is A–Z.

1. 

   Agent Zero

   Open-source agentic framework that runs an autonomous AI assistant in Docker with its own tools, memory, plugins and skills.

   Automation and agentsDeveloper platformsCoding
   Open sourceVisit Agent Zero (official site)
2. 

   AutoGPT

   Agent platform with a plain-English AutoPilot, a visual workflow builder and self-hosting, plus a $50/month Pro plan with a 7-day trial.

   Automation and agentsDeveloper platformsProductivity
   FreemiumVisit AutoGPT (official site)
3. 

   Browser Use

   Cloud browsers and hosted web agents for AI, plus an MIT-licensed open-source browser agent, billed pay-as-you-go from $0.02 per hour.

   Developer platformsAutomation and agentsCoding
   PaidVisit Browser Use (official site)
4. 

   Browserbase

   Cloud platform of headless browsers, search and fetch APIs and agent runtime for building AI agents that browse and act on the web.

   Developer platformsAutomation and agentsCoding
   FreemiumVisit Browserbase (official site)
5. 

   Firecrawl

   Web data API for AI agents: search, scrape, crawl and interact with websites and get clean content back, with a free tier.

   Developer platformsResearch and searchAutomation and agents
   FreemiumVisit Firecrawl (official site)
6. 

   Stagehand

   Open-source SDK from Browserbase for building browser agents with act, extract and observe calls in TypeScript, Python or Go.

   Developer platformsAutomation and agentsCoding
   Open sourceVisit Stagehand (official site)

Playwright MCP FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: none collected yet, so a monogram is shown.
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.