Spark
Sign In
MCP Connector

Orchestrate MCP Servers for LLM Autonomy

Meta-MCP server letting LLMs autonomously discover, install, and orchestrate other MCP servers to expand their own capabilities.

Connect
Works with github
Phillip Sitbon
Own this? Claim it

Maintainer of this project? Claim this page to edit the listing.

72
Spark score
out of 100
Updated 5 months ago
Version 1.0.0
Models
universal
Add to Favorites

Why it matters

Empower AI assistants to autonomously discover, install, and manage multiple MCP servers, expanding their capabilities on demand. This provides a universal hub for LLM-driven operational expansion.

Outcomes

What it gets done

01

Discover and install various MCP servers.

02

Orchestrate the deployment and management of MCP servers.

03

Enable LLMs to self-expand their functional scope.

04

Browse and script MCP server interactions via CLI.

Install

Add it to your toolbox

Run in your project directory:

curl -fsSL https://spark.entire.vc/get/vb-magg-the-mcp-aggregator | bash

Capabilities

Tools your agent gets

discover_mcp_servers

Discover available MCP servers that can be installed and used.

install_mcp_server

Install a specified MCP server to expand LLM capabilities.

orchestrate_mcp_servers

Orchestrate and manage multiple MCP servers working together.

browse_mcp_servers

Browse available MCP servers using the mbro CLI tool.

Overview

Magg: The MCP Aggregator MCP Server

Magg is a meta-MCP server that lets an LLM autonomously discover, install, and orchestrate other MCP servers, with an mbro CLI for browsing available servers. Use it when an assistant should dynamically expand its own toolset as tasks demand; teams needing a fixed, pre-audited toolset should vet autonomous installs carefully.

What it does

Magg is a meta-MCP server that acts as a universal hub for MCP servers themselves, enabling an LLM to autonomously discover, install, and orchestrate other MCP servers on demand - effectively letting an AI assistant expand its own toolset as new tasks require new capabilities. It includes mbro, an MCP server browser CLI with scripting capabilities for exploring and managing the available MCP server ecosystem from the command line.

When to use - and when NOT to

Use this connector when you want an AI assistant to dynamically pull in whichever specialized MCP servers a task needs, rather than pre-configuring every possible tool up front.

Do not use it if you need a fixed, fully audited set of tools known in advance - Magg's value is in-session capability expansion, which trades predictability for flexibility; teams with strict tool-approval requirements should vet what Magg can install before relying on autonomous discovery.

Inputs and outputs

Inputs are task descriptions or capability needs that drive discovery of relevant MCP servers. Outputs are newly installed and orchestrated MCP servers made available to the LLM for subsequent tool calls.

Capabilities

  • Autonomous discovery of relevant MCP servers for a given task
  • Installation and orchestration of discovered MCP servers on demand
  • mbro: a scriptable CLI for browsing and managing MCP servers

How to install

Refer to the project repository at github.com/sitbon/ma for setup instructions specific to your MCP client.

Who it's for

AI agent builders who want their assistant to self-expand its available tools by discovering and installing other MCP servers as tasks demand.

Source README

A meta-MCP server that operates as a universal hub, enabling LLMs to autonomously discover, install, and orchestrate multiple MCP servers-essentially giving AI assistants the ability to expand their own capabilities on demand. Includes mbro, a powerful MCP server browser CLI with scripting capabilities

FAQ

Common questions

Discussion

Questions & comments · 0

Sign In Sign in to leave a comment.