AI Agents Capstone Project - Health Resource Assistant

A health resource assistant AI agent built using Google's Agent Development Kit (ADK) and Gemini API. The agent searches through a FAQ knowledge base to answer health-related questions about topics like free flu shots, Medicaid providers, mental health resources, and emergency services.

Features

• FAQ knowledge base with categorized health information (general health, mental health, emergency services)
• MCP (Model Context Protocol) toolset integration for file search capabilities
• Interactive testing interface with conversation history
• Automated evaluation framework with scoring metrics
• Performance visualization and analysis

Requirements

• Google Gemini API key
• Python packages: google-adk, google-genai, mcp

Setup

1. Get your Gemini API key

   • Create an API key at Google AI Studio

2. Install dependencies

   pip install -q google-adk google-genai mcp

3. Configure authentication

   • Set your API key as an environment variable GOOGLE_API_KEY

Usage

The notebook contains a complete implementation including:

• FAQ knowledge base creation with health resource information
• AI agent configuration using Google's Agent Development Kit
• Interactive testing interface for asking health-related questions
• Automated evaluation system with scoring metrics

Project Structure

• Creates /tmp/faq-knowledge/faqs directory for FAQ storage
• Implements categorized health FAQs (general health, mental health, emergency services)
• Uses MCP toolset for file search capabilities
• Includes performance evaluation and visualization

Author

Charles Rice

License

Licensed under the Apache License, Version 2.0