Prompt Paws: A Cozy 60 FPS AI Agent Bakery Game Teaching RAG, Guardrails & Context Caching

Prompt Paws is an open-source, browser-based simulation game designed to teach developers about complex AI system architecture through a cozy, pastel-themed bakery setting. Created for the Hacktoberfest Open-Source AI Challenge, the game uses a 60 FPS HTML5 canvas engine to demonstrate production-grade AI concepts. Players manage various stations that represent real-world AI engineering layers, such as vector search for RAG, input guardrails for prompt sanitization, and context caching for token efficiency. By mapping technical challenges to bakery tasks, the platform provides an interactive sandbox for learning how to build reliable, cost-effective agentic workflows. The project includes a built-in architecture linter, 'Chef Whiskers,' which provides real-time feedback on potential system vulnerabilities. The source code is fully available on GitHub, encouraging developers and educators to fork the project to create their own custom AI engineering curriculum.