{"type":"mcp_server","name":"mcp-server-azure-ai-agents","description":"This repository offers two Model Context Protocol (MCP) server implementations, integrating Claude Desktop with Azure AI Search for document and web content search, enhanced by AI and direct connections.","category":"Search & Knowledge","language":"Python","stars":54,"forks":2,"owner":"farzad528","github_url":"https://github.com/farzad528/mcp-server-azure-ai-agents","homepage":null,"setup":"## Setup\n\n**Azure AI Agent Service Implementation (Recommended)**\n\n1.  **Project Directory:**\n\n   ```bash\n   mkdir mcp-server-azure-ai-search\n   cd mcp-server-azure-ai-search\n   ```\n\n2.  **Create a `.env` File:**\n\n   ```bash\n   echo \"PROJECT_CONNECTION_STRING=your-project-connection-string\" > .env\n   echo \"MODEL_DEPLOYMENT_NAME=your-model-deployment-name\" >> .env\n   echo \"AI_SEARCH_CONNECTION_NAME=your-search-connection-name\" >> .env\n   echo \"BING_CONNECTION_NAME=your-bing-connection-name\" >> .env\n   echo \"AI_SEARCH_INDEX_NAME=your-index-name\" >> .env\n   ```\n\n3.  **Set Up Virtual Environment:**\n\n   ```bash\n   uv venv\n   .venv\\Scripts\\activate\n   uv pip install \"mcp[cli]\" azure-identity python-dotenv azure-ai-projects\n   ```\n\n4.  **Use the `azure_ai_agent_service_server.py` script** for integration with Azure AI Agent Service.\n\n### Azure AI Agent Service Setup\n\nBefore using the implementation, you need to:\n\n1.  **Create an Azure AI Project:**\n    - Go to the Azure Portal and create a new Azure AI Project\n    - Note the project connection string and model deployment name\n\n2.  **Create an Azure AI Search Connection:**\n    - In your Azure AI Project, add a connection to your Azure AI Search service\n    - Note the connection name and index name\n\n3.  **Create a Bing Web Search Connection:**\n    - In your Azure AI Project, add a connection to Bing Search service\n    - Note the connection name\n\n4.  **Authenticate with Azure:**\n   ```bash\n   az login\n   ```\n\n### Configuring Claude Desktop\n\n```json\n{\n  \"mcpServers\": {\n    \"azure-ai-agent\": {\n      \"command\": \"C:\\\\path\\\\to\\\\.venv\\\\Scripts\\\\python.exe\",\n      \"args\": [\"C:\\\\path\\\\to\\\\azure_ai_agent_service_server.py\"],\n      \"env\": {\n        \"PROJECT_CONNECTION_STRING\": \"your-project-connection-string\",\n        \"MODEL_DEPLOYMENT_NAME\": \"your-model-deployment-name\",\n        \"AI_SEARCH_CONNECTION_NAME\": \"your-search-connection-name\",\n        \"BING_CONNECTION_NAME\": \"your-bing-connection-name\",\n        \"AI_SEARCH_INDEX_NAME\": \"your-index-name\"\n      }\n    }\n  }\n}\n```\n\n**Direct Azure AI Search Implementation**\n\n1.  **Create a different `.env` File:**\n\n   ```bash\n   echo \"AZURE_SEARCH_SERVICE_ENDPOINT=https://your-service-name.search.windows.net\" > .env\n   echo \"AZURE_SEARCH_INDEX_NAME=your-index-name\" >> .env\n   echo \"AZURE_SEARCH_API_KEY=your-api-key\" >> .env\n   ```\n\n2.  **Install Dependencies:**\n\n   ```bash\n   uv pip install \"mcp[cli]\" azure-search-documents==11.5.2 azure-identity python-dotenv\n   ```\n\n3.  **Use the `azure_search_server.py` script** for direct integration with Azure AI Search.\n\n4.  **Configure Claude Desktop:**\n\n   ```json\n   {\n     \"mcpServers\": {\n       \"azure-search\": {\n         \"command\": \"C:\\\\path\\\\to\\\\.venv\\\\Scripts\\\\python.exe\",\n         \"args\": [\"C:\\\\path\\\\to\\\\azure_search_server.py\"],\n         \"env\": {\n           \"AZURE_SEARCH_SERVICE_ENDPOINT\": \"https://your-service-name.search.windows.net\",\n           \"AZURE_SEARCH_INDEX_NAME\": \"your-index-name\",\n           \"AZURE_SEARCH_API_KEY\": \"your-api-key\"\n         }\n       }\n     }\n   }\n   ```","tools":"## Available Tools\n\n1.  AI-Enhanced Search (Azure AI Agent Service optimizes search results with intelligent processing)\n2.  Multiple Data Sources (Search both your private documents and the public web)\n3.  Source Citations (Web search results include citations to original sources)\n4.  Flexible Implementation (Choose between Azure AI Agent Service or direct Azure AI Search integration)\n5.  Seamless Claude Integration (All search capabilities accessible through Claude Desktop's interface)\n6.  Customizable (Easy to extend or modify search behavior)\n7.  Azure AI Search Tool (Search your indexed documents with AI-enhanced results, available with Azure AI Agent Service)\n8.  Bing Web Grounding Tool (Search the web with source citations, available with Azure AI Agent Service)\n9.  Keyword Search (Exact lexical matches, available with direct Azure AI Search integration)\n10. Vector Search (Semantic similarity using embeddings, available with direct Azure AI Search integration)\n11. Hybrid Search (Combination of keyword and vector searches, available with direct Azure AI Search integration)","faq":null,"created_at":"2025-03-14T12:32:16+00:00","updated_at":"2025-03-28T07:26:28+00:00","source_url":"https://model-context-protocol.com/servers/mcp-azure-ai-search-agent-servers","related_articles":[]}