MAGG
🍎 🪟 🐧 ☁️ 🏠 🐍 - Magg: A meta-MCP server that acts as a universal hub, allowing LLMs to autonomously discover, install, and orchestrate multiple MCP servers - essentially giving AI assistants the power to extend their own capabilities on-demand.
What is MAGG?
MAGG is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to 🍎 🪟 🐧 ☁️ 🏠 🐍 - magg: a meta-mcp server that acts as a universal hub, allowing llms to autonomously discover, install, and orchestrate multiple mcp servers - essentially giving ai assistants the p...
🍎 🪟 🐧 ☁️ 🏠 🐍 - Magg: A meta-MCP server that acts as a universal hub, allowing LLMs to autonomously discover, install, and orchestrate multiple MCP servers - essentially giving AI assistants the power to extend their own capabilities on-demand.
This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- 🍎 🪟 🐧 ☁️ 🏠 🐍 - Magg: A meta-MCP server that acts as a univer
Use Cases
Maintainer
Works with
Installation
NPM
npx -y maggManual Installation
npx -y maggConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use MAGG
Magg (MCP Aggregator) is a meta-MCP server written in Python that acts as a universal hub, letting LLMs autonomously discover, install, configure, and orchestrate multiple MCP servers at runtime without human intervention. It exposes tools like magg_search, magg_add, magg_enable, and magg_status so AI assistants can extend their own capabilities on demand — effectively functioning as a package manager for LLM tools. Configurations persist in .magg/config.json across sessions, and Magg supports stdio, HTTP, and Docker transports, making it suitable for both local development and production deployments.
Prerequisites
- Python 3.12 or higher (3.13+ recommended)
- uv package manager installed from astral.sh/uv
- An MCP client such as Claude Desktop, Cline, or Cursor
- Optional: Docker for containerized HTTP transport deployments
- Optional: RSA key pair for JWT bearer token authentication on HTTP transport
Install Magg via uv
Install Magg as a globally available tool using uv tool install. This is the recommended approach as it isolates the environment and makes the magg CLI available system-wide.
uv tool install maggStart the Magg server in stdio mode
Run magg serve to start the server using stdio transport, which is compatible with Claude Desktop, Cline, Cursor, and most MCP clients. For system-wide HTTP access, use the --http flag instead.
# stdio transport (for MCP clients)
magg serve
# HTTP transport (system-wide access on port 8000)
magg serve --httpAdd the MCP server to your Claude Desktop config
Register Magg in your Claude Desktop (or other client) MCP configuration. Magg runs as a subprocess via stdio, so you point the command at the installed magg binary.
Let the LLM discover and add MCP servers
Once connected, your AI assistant can use Magg's built-in tools to search for new MCP servers, add them by URL or registry name, and enable them. Configurations are persisted in .magg/config.json.
Check server health and manage kits
Use magg_status to see which servers are running and healthy. Bundle related servers into kits with magg_kit_create for easy group loading and unloading.
# Check all server health via the CLI browser
uvx --from magg mbro
# Or run directly from GitHub without installing
uvx --from git+https://github.com/sitbon/magg.git magg serveMAGG Examples
Client configuration
Add Magg to Claude Desktop's MCP configuration. Magg will then expose tools that let the AI assistant discover and add more MCP servers dynamically.
{
"mcpServers": {
"magg": {
"command": "magg",
"args": ["serve"]
}
}
}Prompts to try
Once Magg is running, your AI assistant can manage its own tool ecosystem through these kinds of requests.
- "Search for an MCP server that can browse the web and add it"
- "List all MCP servers currently managed by Magg and their status"
- "Enable the filesystem MCP server so I can read local files"
- "Create a kit called 'research' with the web search and arxiv servers"
- "Check the health of all connected MCP servers"Troubleshooting MAGG
magg command not found after installation
Ensure uv's tool bin directory is in your PATH. Run uv tool dir to find the location, then add it to your shell profile (e.g., export PATH="$HOME/.local/bin:$PATH").
Dynamic tool updates not appearing in Claude Desktop
Claude Desktop does not support MCP notifications for dynamic tool updates. Use the ProxyMCP tool that Magg includes specifically for clients lacking notification support — it proxies the protocol to itself to surface updated tools.
Authentication errors on HTTP transport
When running magg serve --http with authentication enabled, you must provide an RSA private key. Set the MAGG_PRIVATE_KEY environment variable with the key content, or mount ~/.ssh/magg/magg.key when using Docker.
Frequently Asked Questions about MAGG
What is MAGG?
MAGG is a Model Context Protocol (MCP) server that 🍎 🪟 🐧 ☁️ 🏠 🐍 - magg: a meta-mcp server that acts as a universal hub, allowing llms to autonomously discover, install, and orchestrate multiple mcp servers - essentially giving ai assistants the power to extend their own capabilities on-demand. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install MAGG?
Install via npm with the command: npx -y magg. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).
Which AI clients work with MAGG?
MAGG works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is MAGG free to use?
Yes, MAGG is open source and available under the AGPL 3.0 license. You can use it freely in both personal and commercial projects.
MAGG Alternatives — Similar Coding Agents Servers
Looking for alternatives to MAGG? Here are other popular coding agents servers you can use with Claude, Cursor, and VS Code.
Dify
★ 142.2kProduction-ready platform for agentic workflow development.
Ruflo
★ 54.0k🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade architecture, self-learning swarm intelligence, RAG integrat
Goose
★ 45.7kan open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Antigravity Awesome Skills
★ 38.3kInstallable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.
AgentScope
★ 25.5kBuild and run agents you can see, understand and trust.
Serena
★ 24.5kA coding agent toolkit that provides IDE-like semantic code retrieval and editing tools, enabling LLMs to efficiently navigate and modify codebases using symbol-level operations instead of basic file reading and string replacements.
Browse More Coding Agents MCP Servers
Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up MAGG 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 MAGG?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.