/mcp endpoint.
Quickstart
Start the gateway, register your first MCP server, and connect Claude in under 5 minutes.
Installation
Install the MCPJungle binary via Homebrew or Docker on any platform.
Deploy with Docker
Run the MCPJungle server locally or in production using Docker Compose.
CLI Reference
Full reference for every MCPJungle CLI command and flag.
Why MCPJungle?
Managing multiple MCP servers across different AI clients — Claude, Cursor, Copilot, custom agents — means duplicating configuration everywhere and losing visibility into what tools are being called. MCPJungle solves this by acting as a central registry and proxy.One endpoint for everything
Register any number of MCP servers. Every AI agent connects to a single
/mcp endpoint and gets access to all tools automatically.Streamable HTTP and STDIO support
Works with remote HTTP-based MCP servers and local STDIO-based servers like
filesystem, github, and time.Tool Groups
Create subsets of tools for specific agents or clients — give Claude only the tools it needs without exposing everything.
Enterprise Access Control
In enterprise mode, issue per-client access tokens and define exactly which MCP servers each client can access.
Get started in 3 steps
Start the server
Pull the Docker Compose file and start MCPJungle with a persistent Postgres database:
Register an MCP server
Use the CLI to add your first MCP server. This example uses context7:
Explore the docs
Managing MCP servers
Register, update, and remove HTTP and STDIO-based MCP servers from the gateway.
Tool Groups
Expose curated subsets of tools to specific MCP clients using named groups.
Enterprise features
Enable access control, user management, and OpenTelemetry observability.
HTTP API reference
Automate MCPJungle management with the REST API.