Uploaded July 2026 | Updated September 2026, 1 week ago

---

Learn how to build an AI agent that queries your Azure Cosmos DB data using natural language — fully secured with Entra ID. In this demo, we deploy the open-source Azure Cosmos DB MCP Toolkit, connect it to a retail database with products, orders, and inventory data, and wire it up to an agent in Microsoft Foundry. No custom code, no API wrappers — just deploy and go.

Links:
🔗 Azure Cosmos DB MCP Toolkit (GitHub): github.com/AzureCosmosDB/MCPToolKit
🔗 Quick Start Guide: github.com/AzureCosmosDB/MCPToolKit/blob/main/docs/QUICK-START.md
🔗 Microsoft Foundry: ai.azure.com
🔗 Azure Cosmos DB: learn.microsoft.com/azure/cosmos-db
🔗 Model Context Protocol (MCP): modelcontextprotocol.io

What's covered:
00:00 — Intro & the problem
00:30 — Retail data in Cosmos DB
01:00 — What is the MCP Toolkit
01:30 — One-click deployment
02:30 — Connecting to AI Foundry
03:30 — Querying data with natural language
04:30 — Vector search & cross-collection analytics
05:00 — Wrap-up & call to action

The Azure Cosmos DB MCP Toolkit is open source, MIT licensed, and ready to use today. Clone the repo, deploy in minutes, and give your AI agents secure access to your Azure Cosmos DB data.
#azurecosmosdb #MCP #AIAgents #ModelContextProtocol #MicrosoftFoundry #Azure