Farzad528 MCP Server Azure Ai Agents

🚀 Azure AI Agent Service + Azure AI Search MCP Server

An MCP (Model Context Protocol) server enabling Claude Desktop to utilize Azure AI services for content search. You can choose to use the Azure AI Agent Service (including document search and web search) or directly integrated Azure AI Search.

🚀 Quick Start

This project provides two MCP server implementations to connect Claude Desktop with Azure search capabilities.

✨ Features

• Support multiple search modes: index-based, web, and hybrid.
• Configurable tool instructions to adjust search behavior.
• Flexible response formatting options.
• Highly scalable, allowing easy addition of new features.
• Support for both Chinese and English.

📦 Installation

1. Install Python Environment

Ensure Python 3.8 or a higher version is installed, and install pip.

python --version
pip install --upgrade pip

2. Install Dependencies

For Azure AI Agent Service:

uv pip install "mcp[cli]" azure-core==1.5.0 python-dotenv

For Direct Azure AI Search:

uv pip install "mcp[cli]" azure-search-documents==11.5.2 python-dotenv

3. Download Project Code

Clone the repository from GitHub:

git clone https://github.com/your-organization/azure-ai-mcp.git
cd azure-ai-mcp

📚 Documentation

Overview

This project offers two MCP server implementations to connect Claude Desktop with Azure search features:

1. Azure AI Agent Service Implementation (Recommended) - Leverage the powerful Azure AI Agent Service to provide:

   • Azure AI Search Tool - Search your indexed documents with AI-enhanced results.
   • Bing Web Location Tool - Conduct web searches with source citations.

2. Direct Azure AI Search Implementation - Connect directly to Azure AI Search, including three methods:

   • Keyword Search - Perform exact literal matches.
   • Vector Search - Utilize semantic similarity of embeddings.
   • Hybrid Search - Combine document and web content searches.

Configuration Instructions

Azure AI Agent Service Configuration

Create an env_agent_service file and add the following content, replacing it with your actual information:

AZURE_AI_SUBSCRIPTION_ID=your-subscription-id
AZURE_AI_ENDPOINT=https://your-region.api.cognitive.microsoft.com

Direct Azure AI Search Configuration

Create an env_search_service file and add the following content, replacing it with your actual information:

AZURE_SEARCH_SERVICE_ENDPOINT=https://your-service-name.search.windows.net
AZURE_SEARCH_INDEX_NAME=your-index-name
AZURE_SEARCH_API_KEY=your-api-key

Usage

Start the Server

For Azure AI Agent Service:

uv run python server.py

For Direct Azure AI Search:

uv run python search_server.py

Configuration in Claude Desktop

Add the following content to Claude's configuration file:

Azure AI Agent Service Configuration:

{
  "mcpServers": {
    "azure-agent-service": {
      "command": "C:\\路径\\to\\venv\\Scripts\\python.exe",
      "args": ["C:\\路径\\to\\server.py"],
      "env": {
        "AZURE_AI_SUBSCRIPTION_ID": "your-subscription-id",
        "AZURE_AI_ENDPOINT": "https://your-region.api.cognitive.microsoft.com"
      }
    }
  }
}

Direct Azure AI Search Configuration:

{
  "mcpServers": {
    "azure-search-service": {
      "command": "C:\\路径\\to\\venv\\Scripts\\python.exe",
      "args": ["C:\\路径\\to\\search_server.py"],
      "env": {
        "AZURE_SEARCH_SERVICE_ENDPOINT": "https://your-service-name.search.windows.net",
        "AZURE_SEARCH_INDEX_NAME": "your-index-name",
        "AZURE_SEARCH_API_KEY": "your-api-key"
      }
    }
  }
}

Example Usage

Enter the following queries in Claude Desktop:

• Use the Azure AI Search Tool:

  • "Search for AI-related information in my Azure index."

• Use the Bing Web Location Tool:

  • "Find the latest developments of large language models on the web."

• Use the Hybrid Search Tool:

  • "Find research papers on deep learning."

Frequently Asked Questions

Server Not Displayed

1. Check the Claude Desktop logs (located at %APPDATA%\Claude\logs\mcp*.log).
2. Ensure that the file paths and environment variables are configured correctly.
3. Test running the server directly: python server.py.

Azure AI Agent Service Issues

1. Ensure that the Azure AI project is configured correctly.
2. Verify that the connection exists and is configured correctly.
3. Check the Azure authentication status.

Customization Options

• Modify Tool Instructions - Adjust the instructions for each agent to change how it processes queries.
• Add New Features - Integrate more features using the @mcp.tool() decorator.
• Customize Response Format - Edit the way Claude Desktop returns responses.
• Adjust Web Search Parameters - Modify the focus areas of the Bing search tool.

📄 License

This project is licensed under the MIT License.