What this list covers
Automation MCP servers are the action layer of an agentic system. Playwright MCP and Puppeteer MCP drive a real browser, click elements, fill forms, and capture screenshots. Chrome DevTools MCP exposes the full Chrome DevTools Protocol. Together, they let an agent end-to-end test a web app, automate a SaaS workflow, or verify that a deploy worked.
- Playwright MCP for cross-browser automation
- Puppeteer MCP for Chrome-only automation
- Chrome DevTools MCP for console, network, and profiling
- Fetch MCP for one-off HTTP requests
The full list
3 Model Context Protocol servers, ordered by adoption.
Playwright MCP
Reliable browser automation for AI agents
A browser automation MCP server built on Playwright. Agents can navigate pages, click elements, fill forms, capture screenshots, and inspect the DOM with deterministic waits and accessibility-aware locators.
Puppeteer MCP
Headless Chrome automation via MCP
A Puppeteer-based MCP server for scraping, screenshotting, and interacting with web pages. A practical alternative to Playwright when teams already standardize on the Puppeteer ecosystem.
Chrome DevTools MCP
Inspect and script the live browser
Expose Chrome DevTools Protocol commands through MCP. Agents can capture console output, profile performance, manipulate cookies, and execute JavaScript against a live Chrome instance.
Frequently asked questions
Playwright MCP vs Puppeteer MCP — which is better?
Playwright MCP is the more popular choice because it supports multiple browsers (Chromium, Firefox, WebKit) and has better auto-waiting. Puppeteer MCP is a good fit if you already standardize on the Puppeteer ecosystem.
Can an agent click a button with MCP?
Yes. Playwright MCP and Puppeteer MCP expose a click tool that locates an element (by selector, role, or text) and clicks it. You can also hover, fill inputs, take screenshots, and execute JavaScript.