1
Microsoft's MCP server for browser automation with Playwright
★ 38k · +196 this week · Apache-2.0 · updated 2026-09-28
9.1/10
What it does Playwright MCP is Microsoft's MCP server for browser automation, built on the Playwright testing library. Instead of screenshots, it gives the agent structured accessibility snapshots of the page, so it can find and use elements by role and name without a vision model. For vibe coders, it is a common way to let an agent open the… Read more →
Pros
- Reliable, fast, no vision model needed
- Official Microsoft project
Cons
- Adds many tools to context
Pricing: Free · browser testing
2
Official Chrome DevTools MCP server: debug, inspect and profile pages from your agent
★ 53k · +281 this week · Apache-2.0 · updated 2026-10-02
8.7/10
What it does Chrome DevTools MCP is the Chrome DevTools team's official server for connecting coding agents to a live Chrome browser. It uses Puppeteer to drive the browser and automatically waits for each action to finish. Beyond browsing, it opens up the DevTools features you would normally use by hand: console messages with source-mapped stack traces, network requests, performance… Read more →
Pros
- Official Chrome team project
- Performance traces and network analysis
Pricing: Free · browser debugging
3
Cloud browsers for agents, with Stagehand natural-language actions
7.8/10
What it does Browserbase MCP gives an agent a browser that runs in Browserbase's cloud, so nothing needs to be installed or opened on your machine. The browser is driven by Stagehand, Browserbase's open-source automation framework. The agent gives plain-language instructions like "click the login button", and Stagehand's model turns them into browser actions. This suits scraping, filling in forms,… Read more →
Pros
- No local browser needed
- Natural-language actions
Cons
- Paid beyond the free tier
Pricing: Free plan, paid from $20/mo · browser cloud