awesome-llm-apps/google_adk_tutorials/structured_output_agent

📧 Structured Output in Google ADK

A tutorial demonstrating how to implement structured output using Google's ADK (Agent Development Kit) framework. This example uses an email generator agent to show how to create type-safe, structured responses with Pydantic schemas and Gemini 2.5 Flash model.

Tutorial Features

• 📝 Structured Output Implementation:

  • Learn how to use Pydantic schemas for type-safe output
  • Understand how to define structured response formats
  • See how Google ADK handles structured responses

• 🎯 Email Generator Example:

  • Practical example using email generation as the use case
  • Shows how to create professional email content with proper structure
  • Demonstrates real-world application of structured output

• 🔧 Google ADK Best Practices:

  • Simple agent definition with clear instructions
  • Proper use of output schemas for reliable results
  • Minimal codebase demonstrating core concepts

How to Run

1. Setup Environment

   # Clone the repository
   git clone https://github.com/Shubhamsaboo/awesome-llm-apps.git
   cd awesome-llm-apps/google_adk_tutorials/structured_output_agent/email_generator_agent
   
   # Install dependencies
   pip install -r requirements.txt
   

2. Configure API Keys

   • Get Google AI API key from Google AI Studio
   • Set up your API credentials for Gemini access

3. Run the Agent

   # Start the ADK web interface from the root folder
   cd google_adk_tutorials/structured_output_agent
   adk web
   

   Then:

   1. Open the web interface in your browser
   2. Select the "email_generator_agent"
   3. Enter your email request (e.g. "Write a professional email to schedule a meeting with a client")
   4. The response will be a structured JSON with subject and body fields

Tutorial Overview

This tutorial demonstrates structured output implementation in Google ADK:

1. Agent Definition: Learn how to create a LlmAgent with Gemini 2.5 Flash
2. Output Schema: Understand how to use Pydantic models for structured responses
3. Instructions: See how to write clear prompts for structured output
4. Structured Response: Learn how to handle JSON responses with defined schemas

Code Structure

• agent.py: Contains the main agent definition and Pydantic schema
• __init__.py: Module initialization for easy imports

Dependencies

• google-adk: Google's Agent Development Kit
• pydantic: Data validation and settings management

How Structured Output Works

This tutorial shows how Google ADK handles structured output:

1. Input Processing: Takes natural language requests and processes them through the agent
2. Content Generation: Uses Gemini 2.5 Flash to generate content based on instructions
3. Output Structuring: Automatically formats responses according to the Pydantic schema
4. Response Validation: Ensures the output matches the defined structure and types

This approach demonstrates how to create reliable, type-safe responses in Google ADK applications.