Kernel
A Model Context Protocol server that connects AI assistants with the Kernel platform, enabling them to deploy applications, automate web browsers, and manage cloud resources.
What is Kernel?
Kernel is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to model context protocol server that connects ai assistants with the kernel platform, enabling them to deploy applications, automate web browsers, and manage cloud resources.
A Model Context Protocol server that connects AI assistants with the Kernel platform, enabling them to deploy applications, automate web browsers, and manage cloud resources.
This server falls under the Cloud Services and Browser Automation categories on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- A Model Context Protocol server that connects AI assistants
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx kernel-mcp-serverConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Kernel
Kernel MCP Server connects AI assistants to the Kernel cloud platform, enabling them to control cloud-hosted Chromium browsers, execute Playwright automation scripts, and manage browser profiles, proxies, and deployed apps — all without local infrastructure. It exposes 15 tools spanning browser management, Playwright code execution, shell commands inside browser VMs, and documentation search, making it ideal for developers who want to automate the web or deploy browser-based agents at scale.
Prerequisites
- Node.js 18+ installed for running npx commands
- A Kernel account at https://onkernel.com with API access
- An MCP-compatible client such as Claude Desktop, Cursor, or VS Code with MCP support
- Optional: KERNEL_MCP_DISABLED_TOOLSETS env var if you want to hide sensitive tool families like api_keys
Install via Kernel CLI
Use the Kernel CLI's built-in MCP installer to register the server with your chosen client. Supported targets include cursor, claude, claude-code, vscode, windsurf, zed, and goose.
kernel mcp install --target claudeOr connect via npx mcp-remote
If you prefer a direct connection without the Kernel CLI, use npx mcp-remote pointing at the hosted Kernel MCP endpoint.
npx -y mcp-remote https://mcp.onkernel.com/mcpAdd the server to your MCP client config
In Claude Desktop's config file, add the Kernel MCP server entry. The server runs as a remote endpoint, so you use mcp-remote as the transport.
{
"mcpServers": {
"kernel": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://mcp.onkernel.com/mcp"]
}
}
}Authenticate with Kernel
The Kernel platform requires authentication. Log in via the Kernel CLI or web dashboard to ensure your session token is available for the MCP server to use.
kernel loginOptionally restrict sensitive tool families
If you want to hide tool families like api_keys from the AI assistant, set the KERNEL_MCP_DISABLED_TOOLSETS environment variable before starting the server.
export KERNEL_MCP_DISABLED_TOOLSETS=api_keysVerify the connection
Open your MCP client and ask it to list available tools or use the built-in kernel-concepts prompt to confirm the 15 Kernel tools are accessible, including manage_browsers, execute_playwright_code, and computer_action.
Kernel Examples
Client configuration
Claude Desktop configuration using npx mcp-remote to connect to the Kernel hosted MCP endpoint.
{
"mcpServers": {
"kernel": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://mcp.onkernel.com/mcp"]
}
}
}Prompts to try
Example prompts to explore Kernel's browser automation and cloud management capabilities.
- "Launch a cloud browser and take a screenshot of https://example.com"
- "Execute Playwright code to scrape the top headlines from a news site"
- "Create a new browser profile with saved credentials for my staging environment"
- "List all my running browser sessions and their current status"
- "Deploy a web scraping app to Kernel and return the results"Troubleshooting Kernel
npx mcp-remote fails with a connection error
Ensure you are logged into Kernel (run 'kernel login') and that your network allows outbound HTTPS to mcp.onkernel.com. Check that mcp-remote is the latest version by running 'npx -y mcp-remote --version'.
Tools like manage_api_keys are missing from the tool list
Check if KERNEL_MCP_DISABLED_TOOLSETS is set and includes the tool family you expect. Remove the variable or adjust its value to re-enable the hidden tool families.
execute_playwright_code returns an error about the browser session
Ensure you have an active browser session running on Kernel. Use manage_browsers to create a session first, then run your Playwright code against that session's ID.
Frequently Asked Questions about Kernel
What is Kernel?
Kernel is a Model Context Protocol (MCP) server that model context protocol server that connects ai assistants with the kernel platform, enabling them to deploy applications, automate web browsers, and manage cloud resources. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Kernel?
Follow the installation instructions on the Kernel GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Kernel?
Kernel works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Kernel free to use?
Yes, Kernel is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.
Kernel Alternatives — Similar Cloud Services Servers
Looking for alternatives to Kernel? Here are other popular cloud services servers you can use with Claude, Cursor, and VS Code.
Open WebUI
★ 138.2kUser-friendly AI Interface (Supports Ollama, OpenAI API, ...)
Anything LLM
★ 60.4kThe all-in-one AI productivity accelerator. On device and privacy first with no annoying setup or configuration.
LocalAI
★ 46.4kLocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required.
Nacos
★ 33.0kan easy-to-use dynamic service discovery, configuration and service management platform for building AI cloud native applications.
Xiaozhi ESP32
★ 26.7k本项目为xiaozhi-esp32提供后端服务，帮助您快速搭建ESP32设备控制服务器。Backend service for xiaozhi-esp32, helps you quickly build an ESP32 device control server.
Gateway
★ 11.8kA blazing fast AI Gateway with integrated guardrails. Route to 1,600+ LLMs, 50+ AI Guardrails with 1 fast & friendly API.
Browse More Cloud Services MCP Servers
Explore all cloud services servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Kernel in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Kernel?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.