AI Projects

A collection of AI agent projects and integrations focusing on Azure infrastructure automation, AI-powered tooling, and agentic workflows.

📚 Projects

🔧 Azure Infrastructure Deployment Agents

azure_infra_deploy_agent

Original LangGraph-based Azure deployment agent with notebook interface.

• Location: langgraph/azure_infra_deploy_agent/
• Type: Jupyter notebook-based workflow
• Features: Natural language to Azure Bicep with human-in-the-loop approval
• Supported Services: Storage Account, Key Vault, App Service Plan, Application Insights, Function App (hardcoded)
• 📖 Detailed README

azure_deployment_agent_with_azure_mcp

Enhanced deployment agent using Azure MCP for universal Azure service support.

• Location: langgraph/azure_deployment_agent_with_azure_mcp/
• Type: Modular Python package with CLI and Web UI
• Key Innovation: MCP-powered dynamic schema discovery — supports ANY Azure service
• Interfaces: CLI (main.py) and Gradio Web UI (gradio_app.py)
• 📖 Detailed README

---

🏗️ Foundry Hosted Agents

Azure AI Foundry hosted agent integration with LangChain and MCP.

• Location: foundry-hosted-agents/
• Type: Python agent service
• Key Features: Azure AI Project integration, multi-server MCP client support

---

🚧 AI SRE Agent

(In development)

• Location: ai_sre_agent/

---

🛠️ Common Prerequisites

• Python: 3.10+ recommended
• Azure CLI: For Azure deployment projects (az login required)
• Node.js: For MCP server integration (some projects)
• Virtual Environment: .venv at repository root

📖 Documentation

• CLAUDE.md — Guidance for AI assistants working with this repository
• Individual project READMEs — Detailed setup, architecture, and usage for each project

🚀 Quick Start

Each project has its own setup instructions in its respective README. Generally:

# Create virtual environment (if not exists)
python -m venv .venv

# Activate virtual environment
.venv\Scripts\Activate.ps1  # Windows PowerShell
# or
source .venv/bin/activate    # Linux/Mac

# Navigate to project and install dependencies
cd <project-folder>
pip install -r requirements.txt

# Configure environment variables
cp .env.example .env
# Edit .env with your credentials

📁 Repository Structure

AI_Projects/
├── langgraph/                              # LangGraph-based agents
│   ├── azure_infra_deploy_agent/          # Original notebook-based deployment agent
│   └── azure_deployment_agent_with_azure_mcp/  # MCP-enhanced deployment agent
├── foundry-hosted-agents/                 # Azure AI Foundry agents
├── .venv/                                 # Shared virtual environment
└── README.md                              # This file