Kernel MCP Server is an MCP server that bridges AI assistants with the Kernel platform, enabling them to deploy and manage Kernel apps in the cloud, launch and control headless Chromium sessions for web automation, monitor deployments and track invocations, search Kernel documentation and inject context, execute arbitrary Playwright code against live browsers, and automatically record video replays of browser automation. The server authenticates users via OAuth 2.0 through Clerk to ensure secure access to Kernel resources. It is open-source and fully managed—Kernel runs the production instance so you don’t need to deploy anything. The MCP server supports multiple transports, including a streamable HTTP endpoint (recommended) and stdio via mcp-remote for clients without MCP support. It integrates with a range of MCP clients (e.g., Claude, Cursor, Goose, VS Code, Smithery) through standard connectors and configurations, making it easy to connect. In short, it provides a scalable, secure bridge between conversational AI and Kernel’s automation, deployment, and documentation capabilities.