MCP Server Azure Ai Search

🚀 Azure AI Agent Service + Azure AI Search MCP Server

An MCP (Model Context Protocol) server that enables Claude Desktop to use Azure AI services to search your content. You can choose to use the Azure AI Agent Service (with document search and web search capabilities) or directly integrate Azure AI Search.

🚀 Quick Start

This project aims to connect Claude Desktop with Azure's AI search capabilities through an MCP server. Users can choose between two implementation methods based on their needs:

1. Use Azure AI Agent Service: Offers a powerful agent service that supports complex query processing and multi - data source search.
2. Directly integrate Azure AI Search: Quickly access Azure search capabilities, suitable for simple scenarios.

✨ Features

Goals

• Achieve seamless connection between Claude Desktop and Azure AI services.
• Provide flexible configuration options to meet the needs of different users.

Functionality

• Support search across multiple data sources (including documents and web content).
• Efficient query processing capabilities.
• Easy to configure and expand.

📦 Installation

Install Dependencies

1. Use Azure AI Agent Service:

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

2. Directly integrate Azure AI Search:

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

💻 Usage Examples

Configuration Examples

Configuration for Using Azure AI Agent Service

{
  "mcpServers": {
    "azure-agent-service": {
      "command": "python",
      "args": ["path/to/agent_service.py"],
      "env": {
        "AZURE_AGENT_ENDPOINT": "https://your-endpoint.azure.com",
        "API_KEY": "your-api-key"
      }
    }
  }
}

Configuration for Directly Integrating Azure AI Search

{
  "mcpServers": {
    "azure-search": {
      "command": "python",
      "args": ["path/to/search_server.py"],
      "env": {
        "AZURE_SEARCH_ENDPOINT": "https://your-search-service.search.windows.net",
        "INDEX_NAME": "your-index-name",
        "API_KEY": "your-api-key"
      }
    }
  }
}

Start the Server

1. Use Azure AI Agent Service:

   python agent_service.py
   

2. Directly integrate Azure AI Search:

   python search_server.py
   

Verify Functionality

• Enter the following sample queries in Claude Desktop:
  • "Search for information about AI in the Azure search index."
  • "Find the latest developments in large language models (LLMs)."
  • "Use hybrid search to find information about neural networks."

📚 Documentation

Troubleshooting

Server Not Displayed

• Check the Claude Desktop logs (located at %APPDATA%\Claude\logs\mcp*.log).
• Ensure that the file paths and environment variables are configured correctly.

Issues with Azure AI Agent Service

• Verify that the Azure AI project is configured correctly.
• Ensure that the connection is correctly created and enabled.

Customize the Server

1. Modify tool instructions: Adjust the instructions of the agent service to change how it processes queries.
2. Add new tools: Integrate additional functionality using the @mcp.tool() decorator.
3. Customize the response format: Edit the response format returned by Claude Desktop.

📄 License

This project is open - source under the MIT License.