Enables automated web searching and data extraction from DuckDuckGo search results.
Enables automated web searching and data extraction from Google search results.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Fourth Playwright MCP Server Log in to Fourth and extract the weekly sales report table."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
General-purpose web automation MCP server built on FastMCP v3 and @playwright/mcp.
Gives Claude browser automation capabilities over HTTP — navigate websites, fill forms, extract data, and interact with any web application.
Claude.ai / Cowork Plugin
│ Streamable HTTP
▼
FastMCP v3 (Python)
│
├── 70+ playwright_* tools (proxied via subprocess)
├── 19 web_* orchestration tools
├── 1 browser_wait_for wrapper
└── Skills (MCP resources)Related MCP server: Cloudflare Playwright MCP
Tool | Purpose |
| Navigate to login page, discover form fields (auto_fill option) |
| Check if user is authenticated |
| Navigate + SPA-aware content waiting |
| Wait for page content after any action |
| Identify menus, tabs, pagination |
| Extract table data (rows/csv/markdown, DOM mode) |
| Extract targeted page content |
| Extract links via DOM (structured data) |
| Identify form fields, types, refs |
| Batch-fill form fields |
| Search Google/DuckDuckGo with filters |
| Search + multi-page content extraction |
| Clean article text via DOM parsing |
| Page metadata (OG, JSON-LD, meta tags) |
| Save page as PDF |
| Save browser session state |
| Restore saved session |
| Run JavaScript with error handling |
| CSS selector-based data extraction |
All 70+ @playwright/mcp tools are available with a playwright_ prefix:
playwright_browser_navigate, playwright_browser_snapshot, playwright_browser_click, playwright_browser_type, playwright_browser_take_screenshot, and more.
Tool | Purpose |
| Type-safe wrapper for |
The server deploys to Railway via Dockerfile. Set these environment variables:
Variable | Required | Default | Description |
| No | Set by Railway | HTTP port |
| No |
| Bind address |
| No | — | OAuth client ID (omit to disable auth) |
| No | — | OAuth client secret |
| No | Google OIDC | Authorization endpoint |
| No | Google OIDC | Token endpoint |
| No |
| Server public URL (for OAuth redirects) |
# Install dependencies
uv sync
# Start server
uv run python src/server.py
# → http://localhost:8000/healthdocker build -t playwright-web-mcp .
docker run -p 8000:8000 playwright-web-mcpAdd as a Custom Connector in Claude.ai Project settings:
URL: https://<your-app>.railway.app/mcp
Auth: OAuth 2.1 (if configured)
Reference the MCP endpoint in your plugin's .mcp.json:
{
"mcpServers": {
"web": {
"url": "https://<your-app>.railway.app/mcp"
}
}
}# Run tests
uv run pytest
# Lint
uv run ruff check src/
# Type check
uv run ruff format --check src/Internal use — Fourth Enterprises.
This server cannot be installed
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Hosted MCP server connecting claude.ai, ChatGPT and other AI apps to your own computer
Hosted Amazon Seller and Vendor MCP server for Claude, ChatGPT, Cursor, Codex, Gemini, Copilot.
Hosted Amazon Seller Central and Amazon Ads MCP server for Claude, ChatGPT, Cursor, and agents.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/Rev4nchist/fourth-playwright-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server