🏗️ Solution Architecture

Architecture Flow

Client/User ↓ Azure AI Foundry Agent ↓ Planner ↓ Azure OpenAI GPT-5 ↓ Knowledge Base (Azure AI Search) ↓ Enterprise MCP Server ↓ Company Policies / Holiday Calendar / Enterprise Search

🚀 Enterprise Knowledge AI Agent

An enterprise-grade AI assistant built using Microsoft Azure AI Foundry, Azure OpenAI, Azure AI Search, and FastMCP.

This project demonstrates how to build a grounded AI agent capable of answering enterprise questions using Azure AI Search and invoking custom tools through the Model Context Protocol (MCP).

---

✨ Features

• Azure AI Foundry Agent
• Azure OpenAI GPT-5
• Azure AI Search Knowledge Base
• FastMCP Server
• Enterprise Search Tool
• Company Policy Tool
• Holiday Calendar Tool
• Retrieval-Augmented Generation (RAG)
• Grounded Responses
• Tool Calling

---

🏗️ Architecture

Architecture diagram will be added.

---

🛠️ Tech Stack

Technology	Purpose
Python	Backend
Azure AI Foundry	AI Agent
Azure OpenAI	LLM
Azure AI Search	Knowledge Retrieval
FastMCP	MCP Server
JSON	Enterprise Data
GitHub	Version Control

---

📁 Project Structure

Enterprise-Knowledge-AI-Agent/
│
├── mcp_server/
│   ├── config.py
│   ├── search.py
│   ├── server.py
│   ├── tools.py
│   └── data/
│       ├── company_policy.json
│       └── holiday_calendar.json
│
├── requirements.txt
├── README.md
├── test_search.py
└── test_tools.py

---

⚙️ Setup

git clone https://github.com/tarunyadav24/Enterprise-Knowledge-AI-Agent.git

cd Enterprise-Knowledge-AI-Agent

pip install -r requirements.txt

python mcp_server/server.py

---

📌 Future Enhancements

• Planner
• Memory
• Observability
• Evaluations
• Tracing
• Multi-Agent Support
• Microsoft Teams Integration
• SharePoint Integration

---

👨‍💻 Author

Tarun Yadav

• Azure AI Foundry
• Azure OpenAI
• Azure AI Search
• FastMCP
• Agentic AI