SSOJet MCP Server
OfficialClick on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@SSOJet MCP Serveradd 10 and 20"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Model Context Protocol (MCP) Server (SSOJet)
This is a Model Context Protocol (MCP) server powered by SSOJet for authentication. Users must first sign in with SSOJet. Once authenticated, they can access and use secure tools such as the add tool exposed by this server.
Configuration
SSOJet Setup
Go to your SSOJet dashboard.
Create a new Single page web application.
Set the callback URL for local development:
http://localhost:8788/callbackNote the following details from your app:
Client ID
Client Secret
Issuer URL (e.g.
https://<your-tenant>.auth.ssojet.com/v1/)
These will be used to configure your server.
Set up a KV Namespace
This project uses a Cloudflare KV namespace to store token metadata:
wrangler kv:namespace create "OAUTH_KV"Then, add the KV binding to your wrangler.jsonc.
Related MCP server: MCP Authentication Demo
Environment Variables
The following environment variables must be configured to run the server:
Variable | Description |
| The Client ID from your SSOJet application |
| The Client Secret from your SSOJet application |
| The issuer URL (e.g. |
| Scopes to request (e.g. |
| Use |
| Not required in this case (unless your tool makes API calls externally) |
Development
Create a .dev.vars file in the root of your project:
SSOJET_CLIENT_ID=<your_ssojet_client_id>
SSOJET_CLIENT_SECRET=<your_ssojet_client_secret>
SSOJET_ISSUER=https://<your-tenant>.auth.ssojet.com/v1/
SSOJET_SCOPE="openid profile email"
NODE_ENV=developmentThen run the MCP server locally:
npm run devTool Available
The server currently provides a single tool:
Tool | Description |
| Adds two numbers together. Useful for simple math operations. |
Once the user signs in via SSOJet, this tool becomes accessible through compatible MCP clients such as the Workers AI LLM Playground.
Testing with MCP Inspector
You can test your server locally with MCP Inspector:
Set the Transport to
sseSet the URL to:
http://localhost:8788/sseA popup will appear for SSOJet authentication
Once logged in, you’ll see the available tools
Deploying to Cloudflare
Before deploying, set the necessary secrets in your Cloudflare environment:
wrangler secret put SSOJET_CLIENT_ID
wrangler secret put SSOJET_CLIENT_SECRET
wrangler secret put SSOJET_ISSUER
wrangler secret put SSOJET_SCOPEDeploy with:
npm run deployThen, in the SSOJet dashboard, add your deployed callback URL:
https://mcp-ssojet-oidc.<your-subdomain>.workers.dev/callbackTo use the deployed server with MCP Inspector or the LLM Playground, use this endpoint:
https://mcp-ssojet-oidc.<your-subdomain>.workers.dev/sseTroubleshooting
Cloudflare Worker Logs
You can inspect logs and errors using Cloudflare’s observability dashboard:
SSOJet Logs
Visit your SSOJet dashboard and check the Logs section to diagnose authentication issues.
Common Issues
❌ Invalid credentials: Double-check that secrets match your SSOJet application.
❌ Missing callback URL: Ensure all callback URLs are added in your SSOJet dashboard.
❌ Tool not showing: Make sure you're authenticated and using the correct endpoint.
❌ Local connection failed: Ensure the MCP server is running on
http://localhost:8788.
This server cannot be deployed
Maintenance
Related MCP Connectors
- StytchOAuthdev.stytch.mcp
The Stytch MCP server is a reference implementation that demonstrates remote MCP server authentication and authorization using Stytch Connected Apps. It provides OAuth 2.1-compliant authorization (including PKCE), Dynamic Client Registration, and validates Stytch-issued access tokens to enable AI agents to securely interact with external services through permissioned access, supporting scopes like openid, email, profile, and manage:project_data.
The Remote MCP server acts as a standardized bridge between LLM applications (like Claude, ChatGPT, and Cursor) and external services, enabling AI agents to access external tools and resources. Its primary capability is providing a centralized search tool to discover other MCP servers and their respective tools. Unlike local implementations, it runs remotely with OAuth authentication and permission controls for security.
MCP server connecting AI agents to 100+ apps (Gmail, Slack, Notion, GitHub) via one-click OAuth.
MCP server for secureFlows: token-free URL builders and integration-linting tools for AI agents.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceMCP server with example tools for arithmetic and greeting, designed for deployment on Amazon Bedrock AgentCore Runtime with OAuth authentication.4-
- FlicenseNot gradedqualityDmaintenanceA production-ready MCP server with OAuth 2.1 authentication, enabling authenticated MCP tool calls via Google OAuth.-
- FlicenseNot gradedqualityCmaintenanceRemote MCP server with built-in OAuth authentication via Cloudflare Access, enabling secure tool usage like math and image generation with user-based access control.-
- FlicenseNot gradedqualityDmaintenanceRemote MCP server with built-in OAuth authentication via Cloudflare Access, enabling secure tool invocation after user sign-in.-