azure-sfi-agent 1.0.31

Azure MCP Agent for secure, compliant resource deployment

Azure SFI Agent - MCP Server

An intelligent Model Context Protocol (MCP) server for deploying Azure resources with automatic SFI compliance orchestration.

Features

• 🚀 Interactive Deployment: Agent prompts for missing parameters
• 🔒 Automatic NSP Attachment: Network Security Perimeter for storage, key vault, cosmos-db, sql-db
• 📊 Automatic Log Analytics: Diagnostic settings for monitoring-enabled resources
• ✅ SFI Compliance: Enforced security baselines and governance
• 🎯 Zero Bypass: All deployments go through compliance orchestration

Installation

Via uvx (Recommended for GitHub Copilot)

uvx install azure-sfi-agent

Via pip

pip install azure-sfi-agent

Quick Start

1. Configure in GitHub Copilot (VS Code)

Add to your VS Code settings.json:

{
  "github.copilot.mcpServers": {
    "azure-sfi-agent": {
      "command": "uvx",
      "args": ["azure-sfi-agent"]
    }
  }
}

Or if installed via pip:

{
  "github.copilot.mcpServers": {
    "azure-sfi-agent": {
      "command": "python",
      "args": ["-m", "azure_sfi_agent.server"]
    }
  }
}

2. Login to Azure

az login

3. Use in Copilot Chat

User: "Create a storage account for ADLS"

Agent: 📋 Creating storage-account - Please provide:
       ✓ resource_group: (Azure resource group name)
       ✓ storageAccountName: (required)
       ✓ location: (required)
       ✓ accessTier: (required)

User: "RG: my-rg, name: datalake001, location: eastus, tier: Hot"

Agent: ✅ Deployment succeeded
       ✅ NSP attached: my-rg-nsp
       
       Endpoints:
       - DFS: https://datalake001.dfs.core.windows.net/

Supported Resources

Resource Type	NSP	Log Analytics
storage-account (ADLS)	✅	❌
key-vault	✅	✅
cosmos-db	✅	❌
sql-db	✅	❌
openai	❌	❌
ai-search	❌	✅
ai-foundry	❌	✅
log-analytics	❌	❌

Available Tools

• create_azure_resource() - Interactive resource creation with compliance
• list_permissions() - View active role assignments
• list_resources() - View accessible Azure resources
• create_resource_group() - Create resource group with tagging
• get_bicep_requirements() - Check required parameters for a resource type

Requirements

• Python 3.10+
• Azure CLI installed and authenticated
• PowerShell Core (pwsh) for script execution
• Appropriate Azure RBAC permissions (Contributor role)

License

MIT

Support

For issues and questions, please visit the GitHub repository.