🦁 MCPZoo

A single docker container to run, manage, and monitor MCP servers in isolation.

🚀 Quick Start

docker run -p 7998-8199:7998-8199 -e APP_USERNAME=admin -e APP_PASSWORD=changeme123 veloper/mcpzoo:latest

1. Open http://localhost:7999 and login with $APP_USERNAME / $APP_PASSWORD
2. Navigate to "Servers" → Click "Add Server"
3. Click "Sync Servers"
4. Connect...

Each server is proxied on a unique port over the http-streamable transport. For example, if you added two servers Context7 and Fetch, they would be available at:

{
    "mcpServers": {
        "context7" : {
            "type": "http",
            "url": "http://localhost:8010/mcp",
        },
        "fetch" : {
            "type": "http",
            "url": "http://localhost:8011/mcp",
        }
    }
}

📖 Overview

MCPZoo simplifies the management of multiple Model Context Protocol (MCP) servers by providing a web-based interface to configure, monitor, and control MCP servers and their dependencies.

✨ Features

• 🖥️ Server Management: Quick web UI to add and config STDIO/HTTP/SSE MCP servers.
• 🔐 Secure Login: JWT auth for protected web access.
• 🔗 Proxy Access: HTTP-streamable ports (8010-8199) for AI client connections.
• ⚙️ Dependencies: Isolated setups via MISE-EN-PLACE for clean environments.
• 📊 System Monitoring: Live process trees, CPU/memory charts, and history.
• 🐳 Docker Ready: One-container deploy with Supervisor for easy control.

🏗️ Architecture

graph TD
    subgraph Host Machine
        HOST_BROWSER[Web Browser]
        HOST_AI_CLIENT["AI Client<br/>(e.g. Claude Code)"]
        subgraph Docker Container
            subgraph PORTS
                PORT_BF[7998 & 7999]
                PORT_S[8010-8199]
            end
            subgraph Supervisord
                subgraph Overmind
                    MCPZOO_F[Frontend UI]
                    MCPZOO_B[Backend API]
                    MCPZOO_A[Agent]
                end
                subgraph MISE1["MISE"]
                    subgraph FMP1[FastMCP Proxy]
                        MCP_SERVER_CONTEXT7["Context7 (HTTP)"]
                    end
                end
                subgraph MISE2["MISE"]
                    subgraph FMP2[FastMCP Proxy]
                        MCP_SERVER_FETCH["Fetch (STDIO)"]
                    end
                end 
            end
        end
    end
    subgraph Internet
        REMOTE_CONTEXT7["https://mcp.context7.com/mcp"]
    end

    HOST_BROWSER <-->|HTTP Requests| PORT_BF

    PORT_BF <--> MCPZOO_F
    PORT_BF <--> MCPZOO_B

    HOST_AI_CLIENT <-->|"HTTP-Steamable"| PORT_S
    PORT_S <--> MCP_SERVER_CONTEXT7
    PORT_S <--> MCP_SERVER_FETCH
    MCP_SERVER_CONTEXT7 <-->|"HTTP-Streamable"| REMOTE_CONTEXT7

🔧 Configuration

Environment Variables

# Application
APP_USERNAME=admin
APP_PASSWORD=changeme123

# Database
SQLITE_PATH=./data/mcpzoo.db

# JWT
JWT_SECRET=your-secret-key-here-change-in-production
JWT_ALGORITHM=HS256
JWT_EXPIRATION_DAYS=7
JWT_TOKEN_REFRESH_DAYS=7

# Server Ports (internal)
BACKEND_WEB_PORT=7998
FRONTEND_WEB_PORT=7999

# Logging
LOG_LEVEL=INFO

# Process Snapshot Retention
PROCESS_SNAPSHOT_RETENTION_MINUTES=15

🤝 Contributing

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Commit changes (git commit -m 'Add amazing feature')
4. Push to branch (git push origin feature/amazing-feature)
5. Open a Pull Request

📄 License

This project is licensed under the BSD-3-Clause License. See the LICENSE file for details.

🌟 Acknowledgments

Built on the shoulders of giants: Docker, Supervisor, FastAPI, React JS, FastAPI, FastMCP, Shedcn UI, Overmind, MISE-EN-PLACE, Python, Linux, and the open source community.

MCPZoo - Taming MCP servers since 2026 🦁