RiscadoA/singlestore-summer-hackathon-2023
LLM agent stuck in a barebones Pokemon-like world forced to fulfill useless goals
This project creates an interactive simulation where a single AI agent navigates a simple, Pokemon-like virtual world. Users provide a goal, and the agent uses its understanding of the world, stored in a database, to plan and execute actions like walking or interacting with objects. It's designed for researchers or enthusiasts exploring how AI models make decisions and learn from environmental feedback in a controlled environment.
No commits in the last 6 months.
Use this if you are experimenting with AI agent behavior, prompt engineering, or the integration of large language models with external knowledge bases for decision-making in simulated environments.
Not ideal if you need a robust multi-agent simulation, a complex game engine, or a practical tool for real-world automation, as the agent's capabilities are currently limited and prone to repetition.
Stars
3
Forks
—
Language
Python
License
—
Category
Last pushed
Jul 21, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/RiscadoA/singlestore-summer-hackathon-2023"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.