mcp-aoai-web-browsing

MCP Server & Client implementation for using Azure OpenAI

This project provides a minimal server/client application utilizing the Model Context Protocol (MCP) and Azure OpenAI. It leverages FastMCP for the server and Playwright for web browser control, enabling AI applications to interact with web resources in a controlled manner.

Key features:

• Model Context Protocol (MCP): Enables secure, controlled interactions between AI applications and local or remote resources.
• Azure OpenAI Integration: Utilizes Azure OpenAI for AI processing.
• FastMCP Server: Built with FastMCP, providing a fast and Pythonic way to build MCP servers.
• Playwright Web Control: Uses Playwright for end-to-end testing and web browser automation.
• OpenAI Function Calling Format: Converts MCP responses to the OpenAI function calling format for seamless integration.
• Customizable MCP-LLM Bridge: Customizes the MCP-LLM Bridge implementation for converting MCP server responses.

Use cases:

• AI-powered web browsing and data extraction.
• Automated testing of web applications.
• Secure interaction between AI models and web resources.
• Building custom AI tools with web integration.