mcp-aoai-web-browsingA minimal Model Context Protocol 🖥️ server/client🧑‍💻with Azure OpenAI and 🌐 web browser control via Playwright.

mcp-aoai-web-browsing is a minimal server/client application implementation utilizing the Model Context Protocol (MCP) and Azure OpenAI. The MCP server is built with FastMCP and utilizes Playwright for web browser control. The application enables secure, controlled interactions between AI applications and local or remote resources. The MCP response about tools will be converted to the OpenAI function calling format, and a bridge converts the MCP server response to the OpenAI function calling format. This customization is based on the MCP-LLM Bridge implementation. The client_bridge supports both in-process and external (stdio) MCP server connections, allowing it to be reused by different clients, such as Claude Code, VS Code, and custom scripts. The application also supports using external clients, such as Claude Desktop, through the mcp.json configuration. Additionally, the client_bridge can connect to external MCP servers via stdio from Python. This application utilizes Azure OpenAI for deployment, and configuration requires filling in values in the .env file for Azure OpenAI endpoints, API keys, and other necessary information. The application supports various configuration options, including using different libraries for dependency management and renaming configuration files. Overall, mcp-aoai-web-browsing is a tool for utilizing the Model Context Protocol and Azure OpenAI for web browsing and other purposes.