MCP Gradio Client Proof of Concept

This repository showcases a Model Context Protocol (MCP) client implemented using Gradio, demonstrating interaction with MCP servers via STDIO and SSE. MCP standardizes communication between language models and tools, enabling AI assistants with tool integration.

Features include a Gradio interface, STDIO/SSE support, dynamic tool loading, and debugging.

Installation requires Python 3.12+, Node.js, uvicorn, NPX, and an OpenAI API key. Steps involve cloning the repository, creating a virtual environment, installing dependencies using pip install -r requirements.txt, setting up environment variables in a .env file, and running the app with python gradio_ui.py.

Configuration uses a config.json file to define MCP servers, specifying type (stdio/sse), command/URL, arguments, and environment variables. STDIO servers are instantiated by Gradio, while SSE servers must be manually started.

Usage involves starting SSE servers, running the Gradio application, and interacting through the web interface. The project is licensed under the MIT License.