MCP.so
登录

Playwright Mcp

@microsoft

代码仓库主页

关于 Playwright Mcp

Playwright MCP server

#playwright#browser-automation#webscraping

基本信息

分类

浏览器自动化

许可证

Apache-2.0

运行时

node

传输方式

stdio

发布者

microsoft

提交者

vivek

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "playwright": {
      "command": "npx",
      "args": [
        "@playwright/mcp@latest"
      ]
    }
  }
}

工具

15

Navigate to a URL

Go back to the previous page

Go forward to the next page

Choose one or multiple files to upload

Capture accessibility snapshot of the current page, this is better than screenshot

Perform click on a web page

Hover over element on page

Type text into editable element

Select an option in a dropdown

Take a screenshot of the current page. You can't perform actions based on the screenshot, use browser_snapshot for actions.

Press a key on the keyboard

Wait for a specified time in seconds

Save page as PDF

Close the page

Install the browser specified in the config. Call this if you get an error about the browser not being installed.

概览

What is Playwright MCP?

A Model Context Protocol (MCP) server that provides browser automation capabilities using Playwright. It enables LLMs to interact with web pages through structured accessibility tree snapshots, bypassing the need for screenshots or visually-tuned models.

How to use Playwright MCP?

Install the server with your MCP‑compatible client (e.g., VS Code, Cursor, Claude Desktop). Use the standard JSON configuration: "command": "npx" and "args": ["@playwright/mcp@latest"]. Detailed setup guides for each client are in the README.

Key features of Playwright MCP

  • Fast and lightweight – uses accessibility tree, not pixel input.
  • LLM‑friendly – no vision models needed.
  • Deterministic tool application – avoids ambiguity of screenshot‑based approaches.
  • Runs on Node.js 18 or newer.
  • Supports a wide range of MCP clients.

Use cases of Playwright MCP

  • Exploratory automation with persistent browser state.
  • Self‑healing tests that iterate over page structure.
  • Long‑running autonomous workflows requiring continuous context.
  • General browser automation driven by language models.

FAQ from Playwright MCP

How does Playwright MCP differ from the Playwright CLI?

The MCP variant is suited for agentic loops that benefit from persistent state, rich introspection, and iterative reasoning over page structure. The CLI + SKILLs approach is more token‑efficient for high‑throughput coding agents.

What are the system requirements?

Node.js 18 or newer is required. No additional runtime dependencies are mentioned.

Which clients can I use with Playwright MCP?

It works with VS Code, Cursor, Windsurf, Claude Desktop, Goose, Junie, Copilot, Gemini CLI, Warp, and many other MCP‑compatible tools. Detailed per‑client installation instructions are provided.

How do I install Playwright MCP?

Run the standard MCP configuration: npx @playwright/mcp@latest. Specific commands for each client (e.g., claude mcp add, code --add-mcp) are documented in the README.

Are there any security or host filtering options?

Yes, you can pass --allowed-hosts <hosts...> to limit which hosts the server can serve from. The environment variable PLAYWRIGHT_MCP_ALLOWED_HOSTS can also be used.

评论

浏览器自动化 分类下的更多 MCP 服务器

Fetcher MCP

jae-jae

MCP server for fetch web page content using Playwright headless browser.

1.1K

Fetch MCP Server

zcaceres

A flexible HTTP fetching Model Context Protocol server.

797

X Twitter Scraper

Xquik-dev

Twitter scraper API skill for tweet search, advanced Twitter search, profile tweets, follower export, media download, monitors, webhooks, MCP, and posting automation: send tweets and replies.

135

Webpage Screenshot MCP Server

ananddtyagi

An MCP that gives your agent the ability to snap a screenshot of webpages. Useful when you want your agent to check its progress during development.

58

Fetch MCP

jae-jae

MCP server for fetch web page content using Playwright headless browser.

1.1K