SymptomAI-MCP

chihebguesmi11/SymptomAI-MCP

3.3

If you are the rightful owner of SymptomAI-MCP and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

The Model Context Protocol (MCP) server is a modular and extensible framework designed to facilitate the orchestration of AI tools and workflows, particularly in the context of agentic AI systems.

Tools
3
Resources
0
Prompts
0

SymptomAI-MCP: Symptom Checker with Model Context Protocol

Python FastAPI LLM AI Agent Gemini HTML CSS JSON-RPC

Overview

This project demonstrates an end-to-end architecture using the Model Context Protocol (MCP) integrated with FastAPI to prototype a Symptom Checker AI Assistant. The system connects a browser-based frontend with an MCP-powered backend, illustrating how agentic AI workflows can be modularized and extended with tools.

The project is designed as both a learning resource for MCP and a practical health AI prototype.

Architecture

Architecture GIF

Flow:
Browser ā†’ FastAPI ā†’ MCP Client ā‡† MCP Server (Tools) ā†’ Results

  • Browser (Frontend): Collects user input.
  • FastAPI (Backend): Handles HTTP routes, acts as entry point.
  • MCP Client: Async bridge between FastAPI and MCP.
  • MCP Server: Hosts modular tools (Extractor, Matcher, Advisor).
  • Results Layer: Outputs structured health insights (Symptoms, Diseases, Advice).

The repository includes an animated architecture GIF visualizing the system flow.

Protocols and Standards Used

  • HTTP/REST ā†’ Communication between Browser and FastAPI.
  • Python async ā†’ For MCP client bridging.
  • JSON-RPC over stdio ā†’ Protocol for MCP client-server communication.
  • MCP (Model Context Protocol) ā†’ Standard for tool orchestration.

Technologies and Frameworks

  • FastAPI (backend web framework, async-first).
  • MCP (Model Context Protocol for AI tool orchestration).
  • Python 3.10+ (core runtime).
  • Matplotlib + NumPy (visualization and architecture animation).
  • ImageIO (GIF generation).
  • HTML5 video and images (frontend UI banner integration).

Features

  • End-to-end MCP architecture demo with FastAPI.

  • Browser-based input and results rendering.

  • Modular MCP server tools (Extractor, Matcher, Advisor).

  • Animated architecture diagram for clear visualization.

  • Applied to a Symptom Checker prototype:

    • Symptom extraction
    • Disease matching
    • Advice generation
    • Web search with TAVILY

Repository Structure

Project Structure

This project is organized into the following key directories and files:

.
ā”œā”€ā”€ app.py                   # The main FastAPI application entry point.
ā”œā”€ā”€ mcp_client/              # Contains the implementation for the MCP client.
ā”œā”€ā”€ mcp_server/              # Houses the MCP server, including various tools and the LLM (Large Language Model) integration.
ā”œā”€ā”€ static/                  # Stores static frontend assets, such as images and video.
ā”œā”€ā”€ templates/               # Contains the HTML templates for the frontend.
ā”œā”€ā”€ mcp_architecture_gif.gif   # An animated visualization of the MCP architecture.
ā””ā”€ā”€ README.md                # The main README file for this repository.