Click 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., "@MCP_Python how to deploy Azure Container Apps?"
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.
MCP (Model Context Protocol) server deployed on Azure Functions that exposes a LearnAgent tool: any MCP-compatible client (Claude Code, VS Code, custom agents) can call it to search and retrieve official Microsoft Learn documentation in real time.
Built with Azure AI Foundry as the LLM backend. Two components:
agent_client.py — local agent that queries Microsoft Learn docs via MCP
server/ — MCP server deployed on Azure Functions, exposes a LearnAgent tool
Endpoint:
https://azure-function-jnb-agents.azurewebsites.net/runtime/webhooks/mcpTransport: Streamable HTTP (MCP protocol 2025-03-26)
Auth: Azure Functions system key in header:
x-functions-key: <mcp_extension_key>Get the key:
az functionapp keys list \
--name azure-function-jnb-agents \
--resource-group NovilloBenitoJaime \
--query systemKeys.mcp_extension -o tsvTool | Input | Description |
|
| Searches and retrieves official Microsoft Learn documentation |
from mcp import ClientSession
from mcp.client.streamable_http import streamablehttp_client
import asyncio
KEY = "<mcp_extension_key>"
URL = "https://azure-function-jnb-agents.azurewebsites.net/runtime/webhooks/mcp"
async def main():
async with streamablehttp_client(URL, headers={"x-functions-key": KEY}) as (r, w, _):
async with ClientSession(r, w) as session:
await session.initialize()
result = await session.call_tool("LearnAgent", {"query": "How to deploy Azure Container Apps?"})
print(result.content[0].text)
asyncio.run(main())Install: pip install mcp
mcp.json){
"mcpServers": {
"LearnAgent": {
"url": "https://azure-function-jnb-agents.azurewebsites.net/runtime/webhooks/mcp",
"headers": {
"x-functions-key": "<mcp_extension_key>"
}
}
}
}KEY="<mcp_extension_key>"
# Initialize
curl -X POST https://azure-function-jnb-agents.azurewebsites.net/runtime/webhooks/mcp \
-H "Content-Type: application/json" \
-H "x-functions-key: $KEY" \
-d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2025-03-26","capabilities":{},"clientInfo":{"name":"test","version":"1.0"}}}'
# Call tool
curl -X POST https://azure-function-jnb-agents.azurewebsites.net/runtime/webhooks/mcp \
-H "Content-Type: application/json" \
-H "x-functions-key: $KEY" \
-d '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"LearnAgent","arguments":{"query":"Azure Blob Storage Python quickstart"}}}'Related MCP server: Azure Functions MCP Server
agent_client.py)Connects to https://learn.microsoft.com/api/mcp directly and uses Azure AI Foundry as LLM.
Requirements: Python 3.9+, az login
python -m venv venv
venv\Scripts\activate
pip install -r requirements.txtCreate .env:
FOUNDRY_PROJECT_ENDPOINT=https://<resource>.services.ai.azure.com/api/projects/<project>
FOUNDRY_MODEL=gpt-4o-minipython agent_client.pycd server
# Edit requirements.txt if needed
Compress-Archive -Path function_app.py,host.json,requirements.txt -DestinationPath deploy.zip -Force
az functionapp deployment source config-zip \
--name azure-function-jnb-agents \
--resource-group NovilloBenitoJaime \
--src deploy.zip --build-remote trueRequired app settings on the Function App:
Setting | Value |
| Azure AI Foundry project endpoint |
| Model deployment name |
The Function App needs System Assigned Managed Identity with Azure AI Developer role on the Foundry resource.
Component | Package |
MCP transport |
|
Agent orchestration |
|
Azure Functions hosting |
|
Azure AI Foundry LLM |
|
Azure auth |
|
This server cannot be deployed
Official Microsoft Learn MCP Server – real-time, trusted docs & code samples for AI and LLMs.
MCP server for agentverse documentation, generated by doc2mcp.
The MCP server for Azure DevOps, bringing the power of Azure DevOps directly to your agents.
MCP server for AI agents to plan, verify, and deploy Cloudflare-native apps.
