Agentic Prompt Engineer for Google ADK Hackathon

This project is a command-line tool that automates prompt generation, evaluation, and iterative refinement for businesses implementing chatbots and AI tools. It leverages Google Gemini LLMs for prompt evaluation and refinement.

Features

• Generate prompts for a given use case or task
• Evaluate prompt quality automatically using Gemini LLM
• Iteratively refine prompts based on evaluation feedback
• Command-line interface (CLI)

Setup

1. Clone this repository.
2. Install dependencies:

   pip install -r requirements.txt

3. Set your Gemini API key as an environment variable:

   $env:GEMINI_API_KEY = "your_api_key_here"

Usage

Run the CLI tool:

python main.py

You will be prompted to enter a task/use case. The agent will generate, evaluate, and refine prompts automatically.

File Structure

• main.py - Entry point and CLI
• prompt_generator.py - Prompt generation logic
• prompt_evaluator.py - Prompt evaluation logic
• prompt_refiner.py - Iterative refinement logic
• gemini_api.py - Handles Gemini LLM API communication

License

MIT License