Azure AI Agent Service
Enables connections to Azure AI Agents within any MCP client, allowing users to leverage Azure AI Foundry's models and knowledge tools like Azure AI Search and Bing Web Grounding through a conversational interface.
What is Azure AI Agent Service?
Azure AI Agent Service is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to enables connections to azure ai agents within any mcp client, allowing users to leverage azure ai foundry's models and knowledge tools like azure ai search and bing web grounding through a conversatio...
Enables connections to Azure AI Agents within any MCP client, allowing users to leverage Azure AI Foundry's models and knowledge tools like Azure AI Search and Bing Web Grounding through a conversational interface.
This server falls under the Cloud Services category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Enables connections to Azure AI Agents within any MCP client
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx azure-ai-agent-service-mcp-serverConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Azure AI Agent Service
The Azure AI Foundry MCP Server bridges any MCP-compatible AI client to Azure AI Foundry's full suite of enterprise capabilities, including model catalog exploration, Azure AI Search index management, document operations, fine-tuning job tracking, and evaluation pipelines. It connects to Azure AI Search for knowledge retrieval, supports Bing Web Grounding for live web search, and exposes tools for querying model quotas, deployment guidance, and performance metrics — all through natural language in your MCP client. Enterprise developers and ML engineers use it to manage Azure AI resources, run evaluations, and build knowledge-augmented agents without leaving their AI assistant interface.
Prerequisites
- Python 3 with uv installed (see https://docs.astral.sh/uv/getting-started/installation/)
- An Azure account with Azure AI Foundry access and a deployed resource
- AZURE_AI_SEARCH_ENDPOINT and authentication credentials (API key or service principal) for knowledge tools
- GITHUB_TOKEN for free model testing with rate limits via GitHub Models
- An MCP client such as Claude Desktop, VS Code with GitHub Copilot, or Claude Code
Install uv package manager
The Azure AI Foundry MCP Server is run via uvx, which requires uv. Install it using the official installation script.
# macOS/Linux
curl -LsSf https://astral.sh/uv/install.sh | sh
# Windows (PowerShell)
powershell -ExecutionPolicy Bypass -c "irm https://astral.sh/uv/install.ps1 | iex"
# Verify
uv --versionSet required environment variables
Configure your Azure credentials and search endpoint. For quick testing with GitHub Models, only GITHUB_TOKEN is needed. For knowledge tools, set the Azure AI Search variables.
# For GitHub Models (free tier testing)
export GITHUB_TOKEN=ghp_your_token_here
# For Azure AI Search knowledge tools
export AZURE_AI_SEARCH_ENDPOINT=https://your-search-service.search.windows.net
export SEARCH_AUTHENTICATION_METHOD=api-search-key
export AZURE_AI_SEARCH_API_KEY=your-search-api-key
# For evaluation tools
export EVAL_DATA_DIR=/path/to/eval/data
export AZURE_OPENAI_ENDPOINT=https://your-openai.openai.azure.com
export AZURE_OPENAI_API_KEY=your-key
export AZURE_OPENAI_DEPLOYMENT=gpt-4oRun the MCP server with uvx
Start the Azure AI Foundry MCP server directly from the GitHub repository using uvx. No local clone or install step is needed.
uvx --prerelease=allow \
--from git+https://github.com/azure-ai-foundry/mcp-foundry.git \
run-azure-ai-foundry-mcpConfigure Claude Desktop to use the server
Add the server to your Claude Desktop configuration so it starts automatically when Claude Desktop launches.
{
"mcpServers": {
"azure-ai-foundry": {
"command": "uvx",
"args": [
"--prerelease=allow",
"--from", "git+https://github.com/azure-ai-foundry/mcp-foundry.git",
"run-azure-ai-foundry-mcp"
],
"env": {
"GITHUB_TOKEN": "ghp_your_token",
"AZURE_AI_SEARCH_ENDPOINT": "https://your-search.search.windows.net",
"AZURE_AI_SEARCH_API_KEY": "your-search-key"
}
}
}
}Verify available tools
Ask your MCP client to list available tools to confirm the server is connected and all capability groups are loaded.
# In your MCP client prompt:
# "List all available Azure AI Foundry tools"Azure AI Agent Service Examples
Client configuration
Complete Claude Desktop JSON configuration for the Azure AI Foundry MCP Server with GitHub token and search credentials
{
"mcpServers": {
"azure-ai-agent-service-mcp-server": {
"command": "uvx",
"args": [
"--prerelease=allow",
"--from",
"git+https://github.com/azure-ai-foundry/mcp-foundry.git",
"run-azure-ai-foundry-mcp"
],
"env": {
"GITHUB_TOKEN": "ghp_YOUR_TOKEN_HERE",
"AZURE_AI_SEARCH_ENDPOINT": "https://your-search.search.windows.net",
"SEARCH_AUTHENTICATION_METHOD": "api-search-key",
"AZURE_AI_SEARCH_API_KEY": "YOUR_SEARCH_KEY"
}
}
}
}Prompts to try
Example prompts for exploring Azure AI Foundry capabilities through the MCP server
- "What OpenAI models are available in Azure AI Foundry and what are their capabilities?"
- "Search my Azure AI Search index for documents about quarterly earnings"
- "Show me the status of my fine-tuning jobs and their performance metrics"
- "Compare Phi-4 and GPT-4o models and recommend one for summarization tasks"
- "Run an evaluation on my dataset at /data/eval.jsonl using coherence and groundedness metrics"Troubleshooting Azure AI Agent Service
uvx fails with 'package not found' or prerelease error
Always include the --prerelease=allow flag as shown in the install command. If uv is outdated, update it with uv self update or reinstall from https://astral.sh/uv.
Azure AI Search tools return authentication errors
Verify AZURE_AI_SEARCH_ENDPOINT is the full URL including https:// and .search.windows.net. Confirm SEARCH_AUTHENTICATION_METHOD matches your setup ('api-search-key' or 'service-principal'). For service principal auth, also set AZURE_CLIENT_ID, AZURE_CLIENT_SECRET, and AZURE_TENANT_ID.
Server starts but no tools appear in the MCP client
Check the server logs for startup errors. Ensure the git repository URL is accessible and that your network allows outbound HTTPS to github.com. Try running the uvx command directly in a terminal to see error output before adding it to the client config.
Frequently Asked Questions about Azure AI Agent Service
What is Azure AI Agent Service?
Azure AI Agent Service is a Model Context Protocol (MCP) server that enables connections to azure ai agents within any mcp client, allowing users to leverage azure ai foundry's models and knowledge tools like azure ai search and bing web grounding through a conversational interface. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Azure AI Agent Service?
Follow the installation instructions on the Azure AI Agent Service GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Azure AI Agent Service?
Azure AI Agent Service works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Azure AI Agent Service free to use?
Yes, Azure AI Agent Service is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.
Azure AI Agent Service Alternatives — Similar Cloud Services Servers
Looking for alternatives to Azure AI Agent Service? Here are other popular cloud services servers you can use with Claude, Cursor, and VS Code.
Open WebUI
★ 138.2kUser-friendly AI Interface (Supports Ollama, OpenAI API, ...)
Anything LLM
★ 60.4kThe all-in-one AI productivity accelerator. On device and privacy first with no annoying setup or configuration.
LocalAI
★ 46.4kLocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required.
Nacos
★ 33.0kan easy-to-use dynamic service discovery, configuration and service management platform for building AI cloud native applications.
Xiaozhi ESP32
★ 26.7k本项目为xiaozhi-esp32提供后端服务，帮助您快速搭建ESP32设备控制服务器。Backend service for xiaozhi-esp32, helps you quickly build an ESP32 device control server.
Gateway
★ 11.8kA blazing fast AI Gateway with integrated guardrails. Route to 1,600+ LLMs, 50+ AI Guardrails with 1 fast & friendly API.
Browse More Cloud Services MCP Servers
Explore all cloud services servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Azure AI Agent Service in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Azure AI Agent Service?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.