Best MCP Servers for Browser Automation
MCP servers that let AI agents control browsers, automate web tasks, and extract page data.
Browser automation MCP servers let your AI agent open pages, click elements, fill forms, and extract data without writing custom scraping code. They are ideal for research, QA, and workflow automation.
Top picks
n8n Workflow Security Scanner
Statically analyzes exported n8n automation workflows for security issues and returns structured findings.
seo-tools-mcp
Five general-purpose stdio MCP servers for SEO: access to SERP, Wordstat, Google Search Console, Yandex.Webmaster and Yandex.Metrica straight from Claude Code (or any MCP client). All tools are read-only, output is strict JSON. Not tied to a specific site: defaults (GSC property, Webmaster host, Metrica counter) are configured on the fly.
@grec0/mcp-secure-env-elicit
A secure wrapper that launches other MCP servers, replacing environment secrets with placeholders and eliciting them via an encrypted browser form, never exposing secrets to the model or client.
Agentic Browser
Enables agents to control a real Chromium browser with semantic tools, providing compact observations and outcome-verified actions for web interaction tasks.
opencli-mcp
MCP server wrapping OpenCLI to let agents control a real Windows Chrome browser, reusing existing login sessions and extensions, with DOM/AX snapshot, action, and network capabilities.
cdt
Enables driving an isolated, already-logged-in Microsoft Edge browser from the shell by reusing your daily Edge's login cookies through a browser-extension bridge.
How to choose
- Prioritize transport: stdio servers run locally; HTTP/remote servers need no install but require network trust.
- Check stars & maintenance: Higher GitHub stars and recent commits usually signal a more reliable project.
- Review security scope: File-system and browser servers can access sensitive data; run them in isolated environments.
- Match your client: Not all MCP clients support every transport; verify compatibility with Claude Desktop, Cursor, or your own agent.
Editorial picks are based on public metadata and keyword matching. They do not constitute endorsements; always review source and license before use.