kartikeykamboj/AI-Agent-Coordination-Decision-Engine ? reverse-engineered prompt

Build me a Python project for an AI agent coordination and decision engine that can take a user request, route it through a workflow orchestrator, and get a helpful final answer back from a set of specialized agents. I want a coordinator that decides whether the request should go to a planner, a researcher, or a decision agent, plus prompt templates, short term memory, and Google Gemini as the model behind the responses.

Please keep it organized like a real team project, with clear modules for agents, memory, prompts, workflow, and the main app entry point. Add a few basic tests so I can verify the pieces work, and set it up so it is easy to run locally with a simple install and a single command. If you need to check current LangChain or Gemini docs online, go ahead.

Also include readable documentation and make sure the whole thing feels like an enterprise ready foundation for adding more agents and tools later.