Provides ability to discover and integrate MCP servers hosted on GitHub repositories, fetching setup instructions and configuration details directly from source.
Enables installation and management of Python-based MCP servers published on PyPI, with version tracking and dependency management.
Supports Python-based MCP servers with automatic configuration and tool proxying capabilities.
🧲 MAGG - The MCP (Model Context Protocol) Aggregator
An MCP server that manages and aggregates other MCP servers, enabling LLMs to dynamically extend their own capabilities.
What is MAGG?
MAGG (MCP Aggregator) is a meta-MCP server that acts as a central hub for managing multiple MCP servers. It provides tools that allow LLMs to:
- Search for new MCP servers and discover setup instructions
- Add and configure MCP servers dynamically
- Enable/disable servers on demand
- Aggregate tools from multiple servers under unified prefixes
- Persist configurations across sessions
Think of MAGG as a "package manager for LLM tools" - it lets AI assistants install and manage their own capabilities at runtime.
Key Features
- Self-Service Tool Management: LLMs can search for and add new MCP servers without human intervention
- Automatic Tool Proxying: Tools from added servers are automatically exposed with configurable prefixes
- Smart Configuration: Uses MCP sampling to intelligently configure servers from just a URL
- Persistent Configuration: Maintains server configurations in
.magg/config.json
- Multiple Transport Support: Works with stdio, HTTP, and other MCP transports
Installation
Prerequisites
- Python 3.13 or higher
uv
,poetry
, orpip
Run Directly from GitHub
The easiest way to run MAGG is directly from GitHub using uvx
:
Local Development
For development, clone the repository and install in editable mode:
Usage
Running MAGG
MAGG can run in two modes:
- Stdio Mode (default) - For integration with Claude Desktop, Cline, Cursor, etc.:
- HTTP Mode - For system-wide access or web integrations:
Available Tools
Once MAGG is running, it exposes the following tools to LLMs:
magg_list_servers
- List all configured MCP serversmagg_add_server
- Add a new MCP servermagg_remove_server
- Remove a servermagg_enable_server
/magg_disable_server
- Toggle server availabilitymagg_search_servers
- Search for MCP servers onlinemagg_list_tools
- List all available tools from all serversmagg_smart_configure
- Intelligently configure a server from a URLmagg_analyze_servers
- Analyze configured servers and suggest improvements
Configuration
MAGG stores its configuration in .magg/config.json
in your current working directory. This allows for project-specific tool configurations.
Example configuration:
Adding Servers
Servers can be added in several ways:
- Using the LLM (recommended):
- Manual configuration via
magg_add_server
: - Direct config editing: Edit
.magg/config.json
directly
Documentation
For more documentation, see docs/.
This server cannot be installed
hybrid server
The server is able to function both locally and remotely, depending on the configuration or use case.
A meta-MCP server that manages and aggregates other MCP servers, enabling LLMs to dynamically extend their own capabilities by searching for, adding, and configuring tool servers.
Related MCP Servers
- -securityAlicense-qualityThe Metaplex MCP Server facilitates access to Metaplex documentation and repository details, enabling users to search and interact with various Metaplex resources through the MCP protocol.Last updated -31TypeScriptThe Unlicense
- AsecurityAlicenseAqualityA proxy server that unifies multiple MCP servers, enabling seamless tool, prompt, and resource management via the MetaMCP App.Last updated -1,489119TypeScriptApache 2.0
- AsecurityAlicenseAqualityAn MCP server that provides LLMs access to other LLMsLast updated -41412JavaScriptMIT License
- -securityFlicense-qualityAn MCP server that fetches real-time documentation for popular libraries like Langchain, Llama-Index, MCP, and OpenAI, allowing LLMs to access updated library information beyond their knowledge cut-off dates.Last updated -Python