supermcp
🚀 SuperMCP - Create multiple isolated MCP servers using a single connector. Build powerful Model Context Protocol integrations for databases (PostgreSQL, MSSQL) with FastAPI backend, React dashboard, and token-based auth. Perfect for multi-tenant apps and AI assistants.
claude mcp add --transport stdio dhanababum-supermcp python -m supermcp \ --env DATABASE_URL="postgresql://user:password@host:port/db" \ --env TOKEN_SECRET="your-token-secret" \ --env OAUTH_CLIENT_ID="your-client-id" \ --env OAUTH_CLIENT_SECRET="your-client-secret"
How to use
SuperMCP is a unified platform that lets you run multiple isolated MCP servers from a single connector backend. Each server instance can have its own credentials, connection settings, and security tokens while sharing resources like connection pools and the underlying connector infrastructure. This design simplifies multi-tenant deployments and reduces infrastructure overhead by centralizing management through one unified dashboard and API surface. You can deploy a single connector (e.g., for PostgreSQL or MSSQL) and then configure as many isolated MCP servers as needed, each with its own tools and access controls.
With SuperMCP, you’ll interact with MCP-compatible connectors that expose Tools (such as list_tables, execute_query, and test_connection) and Templates for common workflows. The system uses a FastAPI-based backend and a React-based frontend to manage connectors, servers, tokens, and configurations. Connectors register themselves via the MCP protocol, and each server instance operates independently in terms of credentials and access, while sharing pooling and orchestration features for efficiency.
How to install
Prerequisites:
- Python 3.12+ installed on your system
- Git installed
- Optional: Docker and Docker Compose for containerized deployment
Step 1: Clone the repository
git clone https://github.com/dhanababum-supermcp.git
cd dhanababum-supermcp
Step 2: Set up a Python virtual environment (recommended)
python -m venv venv
source venv/bin/activate # macOS/Linux
venv\Scripts\activate # Windows
Step 3: Install dependencies
pip install -r requirements.txt
Step 4: Configure environment variables
- Create a .env file or export variables in your shell
export DATABASE_URL=postgresql://user:password@host:port/db
export TOKEN_SECRET=your-token-secret
export OAUTH_CLIENT_ID=your-client-id
export OAUTH_CLIENT_SECRET=your-client-secret
Step 5: Run the MCP server
uvicorn app.main:app --reload --port 9000
Optional Step 6: Run with Docker Compose (if provided in repo)
docker-compose up -d
Step 7: Verify startup by visiting
- Frontend: http://localhost:3000
- Backend API: http://localhost:9000
Step 8: Initialize and create MCP servers using the dashboard or API to start registering connectors and configuring isolated servers.
Additional notes
Notes and tips:
- The architecture supports multiple isolated MCP servers from a single connector; ensure each server has unique tokens and credentials for secure isolation.
- If you encounter authentication issues, verify TOKEN_SECRET and OAuth credentials are correctly set.
- For large-scale deployments, consider enabling Docker deployment with Docker Compose for consistent environments and easier scaling.
- Regularly inspect connection pools and per-server limits to optimize performance and avoid resource exhaustion.
- When upgrading, review changes in MCP protocol support and any breaking changes in connectors or the FastAPI backend.
Related MCP Servers
gcp-storage
A Model Context Protocol (MCP) server that provides seamless integration with Google Cloud Storage, enabling AI assistants to perform file operations, manage buckets, and interact with GCS resources directly.
simple
A python implementation of the Model Context Protocol (MCP) server with fastmcp, fastapi and streamablehttp.
smolagents -demo
A comprehensive demonstration of Model Context Protocol (MCP) servers and clients, showcasing AI-powered tools and multi-agent systems for code analysis, security scanning, and intelligent automation.
mcp-proxy
A modular Python client for connecting to MCP servers, integrating with Anthropic Claude, and exposing a FastAPI-based API for handling queries and tool calls.
lightspeed_mcp
Lightspeed MCP: FastAPI + FastMCP integration with automatic discovery of api routes, tools/resources/prompts.
math -learning
Educational MCP server with math operations, matrix algebra, data visualization, and persistent workspace using FastMCP 3.0