What does this MCP server do exactly?
It acts as a protected MCP server that authenticates clients via Entra ID, dynamically registers client applications, and uses the on‑behalf‑of flow to securely call Microsoft Graph.
@localden
Basic information
Remote MCP server with auth gated by Azure API Management
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"remote-auth-mcp-apim-py": {
"command": "npx",
"args": [
"@modelcontextprotocol/inspector@0.9.0"
]
}
}
}Tool list not available yet
This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.
Fetch the live tool list by running this server in a temporary sandbox using the button above.
It is an experimental sample that shows how to deploy an Entra ID‑protected MCP server on Azure. The server uses an OAuth 2.0 authorization pattern where the client acquires a token for the server, then uses the on‑behalf‑of flow to exchange it for a Microsoft Graph access token. This is done in a completely secretless manner.
Install the Azure Developer CLI, clone the repository, register the Microsoft.App resource provider, log in with azd auth login, and deploy with azd up. After deployment, copy the endpoint and use the Model Context Protocol Inspector (or Visual Studio Code) with SSE transport and Bearer token authentication to connect. Test the get_graph_user_details tool to retrieve the authenticated user’s Microsoft Graph data.
It acts as a protected MCP server that authenticates clients via Entra ID, dynamically registers client applications, and uses the on‑behalf‑of flow to securely call Microsoft Graph.
No. The README explicitly marks it as an experimental implementation and warns it should not be used in production scenarios.
The server runs on Azure Functions and Azure API Management, deployed with Bicep and Python. Testing requires Node.js (for the MCP Inspector) and the Azure Developer CLI.
User data is retrieved live from Microsoft Graph via the on‑behalf‑of flow. Tokens are cached in API Management with encrypted session keys; no persistent user data is stored locally.
Transport uses Server‑Sent Events (SSE) with a Bearer token. Authentication follows OAuth 2.0 with PKCE, dynamic client registration, and multiple secure cookies for state correlation and CSRF protection.

Let your cloud agents talk to you. Grok Bot, Meta Muse, ChatGPT, Claude, Devin and others report progress to Heard, which reads it out loud on your Mac, and they pick up the replies you send back.

Launch your product from your AI agent. Connect Claude Code, Codex, Cursor or ChatGPT to Launchdog over MCP and your agent drafts the listing, picks a launch week and books it. 50 launches a week, ranked only by human vo
Model Context Protocol (MCP) server for TranscriptFetch: fetch YouTube transcripts, search, channels, and playlists from any MCP client.

Catch intent conflicts before code conflicts. The open-source coordination protocol for coding agents, built above Git.

Verified distributed compute for humans, applications, and agents. Submit Blender, Octane, and video workloads, route them across qualified GPUs, track asynchronous execution, download finished artifacts, and verify FARP
Comments