MCP browser automation
TL;DR
Browser automation MCP servers let an agent drive a real browser — navigating, clicking, and reading pages. This ability carries significant responsibility because a browser can act on arbitrary web content and reach authenticated sessions.
Capabilities
Browser servers typically expose navigation and interaction tools and return page state for the agent to read. That read-back of live pages is where untrusted content can introduce prompt-injection risk.
Security
A browser connected to a logged-in session can act as the user. Run browser automation in a constrained session and treat everything it reads as potentially hostile content.
In this collection
- Best MCP servers →
- How to choose an MCP server →
- Verified MCP servers →
- MCP server comparison methodology →
- MCP registries, directories, and marketplaces →
- MCP servers for databases →
- Read-only database access with MCP →
- MCP servers for developers →
- MCP servers for search and research →
- MCP servers for productivity →
- MCP servers for DevOps and cloud →
- Enterprise MCP →
- Postgres MCP server guide →
- Playwright MCP →
- MCP server not showing in Claude Desktop →
- Cursor MCP tools not showing →
- MCP connection refused and timeout errors →
- Claude Desktop MCP configuration →
- VS Code MCP configuration →
- Remote MCP servers with OAuth →
- MCP server for GitHub →