FastMCP Multi-Tenancy
A serverless, multi-tenant implementation of MCP servers that runs on Vercel with fluid compute mode, allowing multiple users to connect to the same endpoint while maintaining session state through Redis.
What is FastMCP Multi-Tenancy?
FastMCP Multi-Tenancy is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to serverless, multi-tenant implementation of mcp servers that runs on vercel with fluid compute mode, allowing multiple users to connect to the same endpoint while maintaining session state through redi...
A serverless, multi-tenant implementation of MCP servers that runs on Vercel with fluid compute mode, allowing multiple users to connect to the same endpoint while maintaining session state through Redis.
This server falls under the Cloud Services category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- A serverless, multi-tenant implementation of MCP servers tha
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx fastmcp-multi-tenancyConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use FastMCP Multi-Tenancy
FastMCP Multi-Tenancy (MCPToolKit) is a Python framework for building serverless, multi-tenant MCP servers that run on Vercel with fluid compute mode, using Redis for distributed session state so multiple users can connect to the same endpoint simultaneously without cross-contaminating session data. It provides built-in OAuth integration for providers like Notion and Slack, human approval workflows for high-risk operations, and a decorator-based API so developers can mark individual tools as requiring authentication without managing token storage manually.
Prerequisites
- Python 3.9 or newer
- Redis instance (local or cloud, e.g. Upstash, Redis Cloud) with a REDIS_URL connection string
- Vercel account for serverless deployment (optional for local use)
- OAuth provider credentials (Notion, Slack) if using authenticated tools
Install the MCPToolKit package
Install the package and its dependencies, including the MCP Python SDK.
pip install mcp-python-sdk mcptoolkitSet up a Redis instance
Provision a Redis instance. For local development you can use Docker; for production use Upstash or Redis Cloud. Note the connection URL.
# Local Redis via Docker
docker run -d -p 6379:6379 redis:alpine
export REDIS_URL=redis://localhost:6379/0Define your MCP server with MCPToolKit
Create a server.py file. Public tools are decorated with @server.tool() while authenticated tools also use @requires_auth with the provider name.
from mcptoolkit import MCPToolKit, requires_auth
server = MCPToolKit(
name="My MCP Server",
redis_url="redis://localhost:6379/0"
)
@server.tool()
def public_tool() -> str:
return "No auth required"
@server.tool()
@requires_auth(provider="notion")
def notion_tool() -> str:
return "Requires Notion OAuth"Deploy to Vercel
Deploy your server.py to Vercel with fluid compute mode enabled. Set REDIS_URL as an environment variable in your Vercel project settings.
vercel deploy --prodConfigure your MCP client
Point your MCP client to the deployed Vercel endpoint URL for HTTP/SSE transport, or run locally for development.
{
"mcpServers": {
"fastmcp-multi-tenancy": {
"url": "https://your-project.vercel.app/mcp",
"transport": "sse"
}
}
}FastMCP Multi-Tenancy Examples
Client configuration
MCP client config for connecting to a deployed MCPToolKit server via SSE transport on Vercel.
{
"mcpServers": {
"fastmcp-multi-tenancy": {
"url": "https://your-project.vercel.app/mcp",
"transport": "sse",
"env": {
"REDIS_URL": "redis://your-redis-host:6379/0"
}
}
}
}Prompts to try
Example prompts for interacting with a multi-tenant MCPToolKit server.
- "Connect my Notion account so you can read my workspace pages"
- "List all pages in my Notion database using the authenticated tool"
- "Send a message to the #general channel in my Slack workspace"
- "Run the public status check tool to see if the server is healthy"
- "Approve the pending high-risk operation to delete old records"Troubleshooting FastMCP Multi-Tenancy
Session state is not persisting between requests
Verify that REDIS_URL is correctly set and that the Redis instance is reachable from your deployment. On Vercel, set REDIS_URL as an environment variable in the project settings — local .env files are not read in production. Test Redis connectivity with 'redis-cli -u $REDIS_URL ping'.
OAuth authentication flow fails or tokens are not refreshed
Ensure your OAuth provider credentials (client ID and secret for Notion or Slack) are configured in your server and set as environment variables. MCPToolKit handles token refresh automatically, but the initial OAuth redirect URL must be registered in your provider's allowed redirect URIs.
Vercel deployment times out or cold starts are slow
Enable Vercel Fluid Compute mode in your project settings, which is required for long-lived SSE connections. Without fluid compute, connections may be terminated by Vercel's default function timeout.
Frequently Asked Questions about FastMCP Multi-Tenancy
What is FastMCP Multi-Tenancy?
FastMCP Multi-Tenancy is a Model Context Protocol (MCP) server that serverless, multi-tenant implementation of mcp servers that runs on vercel with fluid compute mode, allowing multiple users to connect to the same endpoint while maintaining session state through redis. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install FastMCP Multi-Tenancy?
Follow the installation instructions on the FastMCP Multi-Tenancy GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with FastMCP Multi-Tenancy?
FastMCP Multi-Tenancy works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is FastMCP Multi-Tenancy free to use?
Yes, FastMCP Multi-Tenancy is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
FastMCP Multi-Tenancy Alternatives — Similar Cloud Services Servers
Looking for alternatives to FastMCP Multi-Tenancy? Here are other popular cloud services servers you can use with Claude, Cursor, and VS Code.
Open WebUI
★ 138.2kUser-friendly AI Interface (Supports Ollama, OpenAI API, ...)
Anything LLM
★ 60.4kThe all-in-one AI productivity accelerator. On device and privacy first with no annoying setup or configuration.
LocalAI
★ 46.4kLocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required.
Nacos
★ 33.0kan easy-to-use dynamic service discovery, configuration and service management platform for building AI cloud native applications.
Xiaozhi ESP32
★ 26.7k本项目为xiaozhi-esp32提供后端服务，帮助您快速搭建ESP32设备控制服务器。Backend service for xiaozhi-esp32, helps you quickly build an ESP32 device control server.
Gateway
★ 11.8kA blazing fast AI Gateway with integrated guardrails. Route to 1,600+ LLMs, 50+ AI Guardrails with 1 fast & friendly API.
Browse More Cloud Services MCP Servers
Explore all cloud services servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up FastMCP Multi-Tenancy in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use FastMCP Multi-Tenancy?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.